document_id
int64 185
2.68k
| context
stringlengths 2.88k
70.8k
| question
stringlengths 11
194
| id
int64 225
5.32k
| answer
stringlengths 1
933
| answer_start
int64 157
69k
|
|---|---|---|---|---|---|
1,584 |
Viral and bacterial co-infection in severe pneumonia triggers innate immune responses and specifically enhances IP-10: a translational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5138590/
SHA: ef3d6cabc804e5eb587b34249b539c1b5efa4cc4
Authors: Hoffmann, Jonathan; Machado, Daniela; Terrier, Olivier; Pouzol, Stephane; Messaoudi, Mélina; Basualdo, Wilma; Espínola, Emilio E; Guillen, Rosa M.; Rosa-Calatrava, Manuel; Picot, Valentina; Bénet, Thomas; Endtz, Hubert; Russomando, Graciela; Paranhos-Baccalà, Gláucia
Date: 2016-12-06
DOI: 10.1038/srep38532
License: cc-by
Abstract: Mixed viral and bacterial infections are widely described in community-acquired pneumonia; however, the clinical implications of co-infection on the associated immunopathology remain poorly studied. In this study, microRNA, mRNA and cytokine/chemokine secretion profiling were investigated for human monocyte-derived macrophages infected in-vitro with Influenza virus A/H1N1 and/or Streptococcus pneumoniae. We observed that the in-vitro co-infection synergistically increased interferon-γ-induced protein-10 (CXCL10, IP-10) expression compared to the singly-infected cells conditions. We demonstrated that endogenous miRNA-200a-3p, whose expression was synergistically induced following co-infection, indirectly regulates CXCL10 expression by targeting suppressor of cytokine signaling-6 (SOCS-6), a well-known regulator of the JAK-STAT signaling pathway. Additionally, in a subsequent clinical pilot study, immunomodulators levels were evaluated in samples from 74 children (≤5 years-old) hospitalized with viral and/or bacterial community-acquired pneumonia. Clinically, among the 74 cases of pneumonia, patients with identified mixed-detection had significantly higher (3.6-fold) serum IP-10 levels than those with a single detection (P = 0.03), and were significantly associated with severe pneumonia (P < 0.01). This study demonstrates that viral and bacterial co-infection modulates the JAK-STAT signaling pathway and leads to exacerbated IP-10 expression, which could play a major role in the pathogenesis of pneumonia.
Text: Scientific RepoRts | 6:38532 | DOI: 10 .1038/srep38532 pathogenesis of several diseases and has been suggested as a potential biomarker of viral infection 10, 11 , late-onset bacterial infection in premature infants 12 , and a promising biomarker of sepsis and septic shock 13, 14 . Combined analysis of IP-10 and IFN-γ has also been reported as a useful biomarker for diagnosis and monitoring therapeutic efficacy in patients with active tuberculosis [15] [16] [17] , and both remain detectable in the urine of patients with pulmonary diseases in the absence of renal dysfunction 18 .
With airway epithelial cells 19 , resident alveolar macrophages (AMs) and blood monocytes-derived macrophages (recruited into tissues under inflammatory conditions 20, 21 ) represent a major line of defense against both pneumococcal (through their high phagocytic capacity [22] [23] [24] ) and influenza infection 25, 26 . So far, no studies have yet focused on the intracellular mechanisms that regulate IP-10 in human blood leukocytes during mixed IAV and SP infection. Several studies indicated that host non-coding small RNAs (including microRNAs) may function as immunomodulators by regulating several pivotal intracellular processes, such as the innate immune response 27 and antiviral activity 28, 29 ; both of these processes are closely related to toll-like receptor (TLR) signaling pathways.
In this study, we firstly investigated the in vitro intracellular mechanisms that mediate the innate immune response in IAV and/or SP infected human monocyte-derived macrophages (MDMs). Using this approach, we observed that mixed-infection of MDMs induces a synergistic production of IP-10 which can be related to a miRNA-200a/JAK-STAT/SOCS-6 regulatory pathway. Subsequently, in a retrospective analysis of clinical samples collected from children ≤ 5 years-old hospitalized with pneumonia, we confirmed that serum IP-10 level could be related to both viral and/or bacterial etiologies and disease severity.
Characteristics of MDMs infected by IAV and/or SP. Initially, we investigated in vitro the impact of single and mixed IAV and SP infection on MDMs. Firstly, active replication of IAV was assessed by qRT-PCR and quantification of new infectious viral particles in the cell supernatants ( Fig. 1a,b ). IAV titer increased over time after single infection with IAV and correlated with increased production of negative-strand IAV RNA. Maximum viral replication was observed at 18-24 hours post-infection, after which time both RNA replication and the quantity of infectious particles decreased. In this in vitro model, subsequent challenge of IAV-infected MDMs with SP had no significant impact on the production of new infectious viral particles (Fig. 1b) . Together, these results indicate permissive and productive infection of MDMs by IAV. Secondly, we evaluated whether MDMs are permissive for both IAV and SP infection. The presence of pneumococci within IAV-and SP-infected primary MDMs was confirmed at 8 h post-infection (Fig. 1c) , suggesting that MDMs are permissive for viral and bacterial co-infection in the early steps of infection. Importantly, confocal co-detection of mixed IAV and SP was only effective following 8 h post-infection due to the bactericidal impact of SP internalization within human macrophages (after 24 h, data not shown). Thirdly, we evaluated the impact of single and mixed infection with IAV and SP on MDM viability. Mixed infection significantly decreased cell viability (65.2 ± 4.5% total cell death at 48 hours post-infection; P < 0.0001) compared to single SP and IAV infection (39.6 ± 1.7% and 17.4 ± 1.1% total cell death, respectively; Fig. 1d ). Taken together, these results confirmed human MDMs are permissive to mixed viral and bacterial infection. mRNA, microRNA and protein expression profiling reveal an overall induction of the host innate immune response following IAV and/or SP infection of MDMs. To investigate the innate immune response orchestrated by IAV-and SP-infected human MDMs, we firstly evaluated the expression of 84 genes involved in the innate and adaptive immune responses (Table S1) ; the major differentially-expressed genes are summarized in Fig. 2a . Expression profiling indicated an overall induction of genes related to the JAK-STAT, NF-Κ β and TLR signaling pathways. Indeed, all interferon-stimulated genes (ISGs) screened, including CXCL10 (fold-change [FC] = 240.9), CCL-2 (FC = 34.2) and MX-1 (FC = 151.4) were upregulated following mixed infection compared to uninfected cells, most of which are closely related to STAT-1 (FC = 52.3), IRF-7 (FC = 6.8) and IFNB1 (FC = 5.2) also found upregulated in mixed infected cells. Secondly, we investigated the endogenous microRNA expression profiles of IAV-and SP-infected MDMs. A selection of microRNAs that were found to be differentially-expressed under different infection conditions are shown in Fig. 2b and Table S2 . MiRNA-200a-3p was overexpressed after both single IAV (FC = 6.9), single SP (FC = 3.7) and mixed IAV/SP infection (FC = 7.3), indicating this miRNA may play a role in the innate immune response to viral and bacterial co-infection. Similar miRNA-200a-3p dysregulation profiles were obtained following IAV and/or SP infections of human macrophages-like (THP-1 monocytes-derived macrophages) or primary MDMs (data not shown). Thirdly, the secreted levels of various antiviral, pro-inflammatory and immunomodulatory cytokines/chemokines were assayed in IAV-and SP-infected-THP-1 and primary MDM cell supernatants. We observed a remarkable correlation between the mRNA and protein expression profiles of single or mixed infected MDMs especially regarding CXCL-10 and IP-10 expression. Indeed, the level of IP-10 was synergistically increased in the supernatant of IAV-infected THP-1 MDMs exposed to SP (mean: 30,589 ± 16,484 pg ml −1 ) compared to single IAV infection (1,439 ± 566.5 pg ml −1 ) and single SP infection (4,472 ± 2,001 pg ml −1 ; P≤ 0.05; Fig. 2c ) at 24 hours after infection. In those cells, IP-10 expression reduced over time (48 to 72 hours), coinciding with a significant higher proportion of necrotic and apoptotic cells (Fig. 1d) . The synergistic expression of IP-10 was similarly observed at 24 hours post-infection using primary MDMs (Fig. 2d) . Significantly increased secretion of the other tested cytokines and chemokines was not observed post-infection, even in mixed infected MDMs (Fig. S1 ). Interestingly, a significant production of IP-10 was also observed in supernatants of primary human airway epithelial cells (HAEC) mixed-infected by IAV and SP compared to the single infections (Fig. 2e) . Taken together, the mRNA and protein profiling results suggested that mixed viral and bacterial infection of MDMs induces a synergistic pro-inflammatory response related to the type-1 interferon and JAK-STAT signaling pathways, with IP-10 as signature of IAV/SP co-infection. Among all microRNAs screened, miR-200a-3p was the most Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 overexpressed in IAV/SP co-infection of human MDMs. In the remainder of this study, we decided to investigate the interconnection between miR-200a-3p expression and the innate immune response.
Endogenous miRNA-200a-3p expression correlates with CXCL10 (IP-10) induction following mixed IAV and SP infection of human MDMs. Using a specific Taqman probe assay targeting miR-200a-3p, we confirmed a significant upregulation of miR-200a-3p following mixed IAV and SP infection of human MDMs (Fig. 3a) . In this experiment, a more marked up-regulation of miR-200a-3p was observed following IAV+ SP compared to results obtained previously (Fig. 2b) . This discrepancy has been attributed to the use of two different approaches to quantify miR-200a-3p expression. The use of a target-specific stem-loop reverse transcription primer in Fig. 3a allows a better sensitivity of miR-200a-3p detection compared to the non-specific fluorescent dye used in Fig. 2b . As the general trend was suggestive of a synergistic induction of miR-200a-3p in response to mixed infection (Fig. 3a) , we hypothesized microRNA-200a-3p may play a role in the regulation of CXCL10 (IP-10), which was also synergistically upregulated in mixed-infected MDMs ( Fig. 2c and d) and primary HAEC ( Statistical analyses were performed using two-way ANOVA with Tukey's post-hoc test; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.
Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 CXCL10 (Fig. 3d) . These results suggested miR-200a-3p indirectly regulates CXCL10 and led us to hypothesize that miR-200a-3p controls a potential repressor of the JAK-STAT signaling pathway. . At 18 h after transfection, the MDMs were singly or mixed infected as described previously. At 8 h post-IAV and/or SP infection, total mRNA was extracted and amplified by PCR using specific primers for the indicated genes. Values represent median ± IQR (a, c) or mean ± SEM (d, e) of three biological replicates. Statistical analyses were performed using a Kruskal-Wallis test (non-parametric, one-way ANOVA with Dunn's post-hoc test) for data presented in (a, c). An ordinary two-way ANOVA (with Tukey's post-hoc multiple comparison test) was used for data presented in (d, e). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. MiRNA-200a-3p indirectly regulates IP-10 expression by targeting SOCS6. As shown in Fig. 2a , several JAK-STAT signaling pathway genes were deregulated in mixed IAV-and SP-infected human MDMs; therefore, we hypothesized that miR-200a-3p directly regulates a regulator of the JAK-STAT signaling pathway. Predictive target analysis indicated that the 3' UTR of suppressor of cytokine signaling-6 (SOCS6) may be targeted by miR-200a-3p (Fig. 3b) . SOCS proteins constitute a class of negative regulators of JAK-STAT signaling pathways that are induced by both cytokines and TLR signaling. MiRNA-200a-3p was not predicted to target any of the other six members of the SOCS gene family. Transfection of human MDMs with MIM-200a downregulated SOCS6 (FC = 0.57) while inhibition of miR-200a-3p (INH-200a) upregulated SOCS6 (FC = 1.55), confirming that miR-200a-3p effectively regulates the expression of SOCS6 (Fig. 3e) . Moreover, SOCS6 was synergistically downregulated in IAV-or IAV/SP-infected MDMs overexpressing miRNA-200a (Fig. 3e) , suggesting that both infection and miR-200a-3p negatively regulate the expression of SOCS6. Finally, western blotting confirmed that expression of SOCS-6 sharply reduced following infection, especially after mixed IAV and SP infection (Fig. 3f) .
These results indicate miR-200a-3p is strongly induced in response to mixed viral and bacterial co-infection, which in turn leads to downregulation of the JAK-STAT regulator SOCS-6 at both the mRNA and protein levels and subsequent upregulation of IP-10.
analyses demonstrated mixed IAV and SP infection of human MDMs and HAEC induced significant production of IP-10. As blood leukocytes and respiratory tract epithelial cells actively contribute to inflammation during pneumonia, we hypothesized the level of IP-10 in serum of patient with pneumonia may be both indicative of mixed respiratory infection and disease severity. As part of a prospective, hospital-based, multicenter case-control study on the etiology of pneumonia among children under 5-years-old, a total of 74 patients (44 male, 30 female) were included in this pilot evaluation. According to WHO guidelines, retrospective analysis indicated 44 (59.5%) children had clinical signs of non-severe pneumonia and 30 (40.5%) children had signs of severe pneumonia. The main patient characteristics at inclusion are shown in Table 1 . Patients with severe pneumonia had significant more recorded episodes of dyspnea (P < 0.001), cyanosis (P = 0.03), lower chest indrawing (P < 0.001), dullness to percussion (P < 0.001) and lethargy (P < 0.001) during chest examination than patient with non-severe pneumonia. Moreover, pleural effusions were significantly more observed among critically ill patients and the duration of hospitalization was significantly longer for the children with severe pneumonia than for those with non-severe pneumonia (P = 0.0015). Two deaths occurred within the group of children retrospectively defined with severe pneumonia. Evaluation of the systemic inflammatory response of the 74 cases is shown in Table 2 . Serum level of CRP, IP-10, PCT, G-CSF, IL-6, IL-8 and MIP-1β were significantly more elevated in serum samples from critically ill patients. Patients with severe pneumonia had significantly higher (4.2-fold) serum IP-10 levels than those with a non-severe pneumonia (P < 0.001) suggesting IP-10 as a promising prognostic marker in pneumonia. Diagnostic accuracy measures for predicting pneumonia severity using blood-based biomarkers are summarized in Table S3 . Briefly, in this study, the optimal IP-10 cut-off value for identifying patient with severe pneumonia was 4,240 pg ml −1 , with an area under the receiver operating characteristic curve of 0.69 (95% CI, 0.57 to 0.82, P < 0.001). Defining as positive a serum IP-10 level above this cut-off resulted in a sensitivity of 63.3%, specificity of 63.6% and a positive likelihood ratio of 1.74. Prognostic values of IP-10 were closed to procalcitonin (PCT; AUC = 0.70; 95% IC, 0.58 to 0.82, P < 0.001) and IL-6 (AUC = 0.70; 95% IC, 0.58-0.83, P < 0.001).
Multiplex PCR-based screening of respiratory and blood samples reveal a high variety of pathogen associations (Table 3) . Respiratory viruses were detected in the nasal aspirates (NAs) of 63/74 patients (85.1%). Etiological bacteria of pneumonia (S. pneumoniae, n = 19; S. aureus, n = 1; or H. influenzae type B, n = 7) were identified via real-time PCR in the blood samples of 27/74 (36.5%) of the patients. Multiplex PCR assays allowed the identification of respiratory bacteria in the blood of 19 patients with negative blood culture results. Among the 74 cases PCR-positive for respiratory pathogens, a single virus or bacteria were detected in the NAs of 7 (9.4%) and 3 (4.0%) patients, respectively; these 10/74 (13.5%) cases were defined as the single infection group. The mixed infection group included the 62/74 (83.8%) cases in which (1) multiple viruses and/or bacteria were identified in NAs (38/74; 51.3%) without any bacteria identified in blood samples or (2) one or more viruses and/or bacteria were identified in NAs and associated with a blood bacteremia (24/74; 32.4%). We evaluated whether IP-10 serum level could correlate with the viral and bacterial etiologies of pneumonia. Patients with mixed infection had significant higher (3.6-fold) IP-10 serum level than patient with single detection (P = 0.03; Table 4 ). A stratified analysis reveals that the highest IP-10 serum level was observed among patients with both several respiratory pathogens identified (mixed-detection group) and severe pneumonia (14,427 pg ml −1 , IQR (3,981-82,994). In detail, a remarkable IP-10 serum level (142,531 pg ml −1 ), representing 33-fold higher above cut-off value predicting pneumonia severity was observed in patient with hRV in NA co-detected with S. pneumoniae (serotype 14) in pleural effusion and blood. In concordance with our in-vitro model of co-infection, a significant IP-10 level (90,338 pg ml −1 ) was quantified in blood sample of patient with severe bacteremic pneumococcal (serotype 14) pneumonia with a positive co-detection of Influenza B virus in NA. Taken together, these results suggest that high serum IP-10 levels are significantly associated with mixed viral and bacterial detection and also related to pneumonia pathogenesis.
This study provides additional in vitro and clinical data to improve our understanding of the immunopathology of mixed viral and bacterial pneumonia (Fig. 4) .
The in vitro model of influenza and pneumococcal superinfection of human MDMs demonstrated that mixed infection synergistically induced release of the pro-inflammatory chemokine IP-10, strongly suggesting human Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 blood leukocytes contribute to the immunopathology of pneumonia. Additionally, transcriptomics and omics analyses provided new data on the inflammatory pathways that are activated during mixed infection and related to synergistic induction of the pro-inflammatory chemokine IP-10 in mixed infected cells. Our observations are consistent with a recent study describing IP-10 induction as host-proteome signature of both viral and bacterial infections 30 . Of the differentially-expressed genes observed in mixed infected MDMs, the transcription factors STAT-1 and IRF-7 appear to play crucial roles in the regulation of interferon-stimulated genes including CXCL10 (IP-10). By focusing on the intracellular mechanisms that regulate inflammatory pathways, we demonstrated a novel role for miRNA-200a-3p in the regulation of CXCL10 (IP-10). These observations are consistent with previous reports showing that RNA virus infection upregulates miR-155 in macrophages and dendritic cells and also regulates suppressor of cytokine signaling 1 (SOCS1), suggesting the existence of a miRNA/JAK-STAT/SOCS regulatory pathway during viral infection 29 . Our study suggests co-infection leads to overexpression of miR-200a-3p, which in turn targets and downregulates the JAK-STAT regulator SOCS-6 and consequently increases CXCL10 (IP-10) expression. Interestingly, a complementary in-silico approach reveals that several microRNAs that were found dysregulated in our experiments of IAV and SP co-infection of MDMs or HAEC, might target several genes of SOCS family and play similar role than miR-200a-3p. Indeed, miRNA-142-3p might target SOCS4, 5, 6 mRNA while miRNA-194-5p might target SOCS2, 3, 4, 5 and 7 mRNA. These observations underline that intra-cellular regulation of IP-10 is not limited to the contribution of a sole microRNA. A complex inter-relationship between numerous host microRNAs and inhibitors of the JAK-STAT signaling pathway occur to control host innate inflammatory response against viral and/or bacterial infections. Clinically, the majority of pediatric CAP cases in this study were associated with both positive viral and/or bacterial detection. Respiratory microorganisms were detected in 97% of cases; 51.3% of which were viral-viral, viral-bacterial or bacterial-bacterial co-detected only in nasal aspirates, 32.4% of which co-detected in both nasal aspirates and blood samples. These data are consistent with previous etiological studies of pediatric CAP 3,31-33 . S. pneumoniae was the major bacteria identified in blood (19/74; 25.7%) and mainly co-detected with respiratory viruses in NAs (16/19; 84.2%). We observed a very high diversity of viral and bacterial associations in biological samples from children with pneumonia. In comparison with IAV and SP14 combination evaluated in-vitro, no pneumonia cases were singly influenza and pneumococcus infected, and no similar co-detection with those two pathogens has been clinically observed. Nevertheless, Influenza B (IVB) virus was identified in 5 patients and two of them had a positive SP co-detection in blood (one non-typable strain and one serotype 14 using our molecular typing test). IVB and SP14 combination seems to be the nearest pathogen co-detection to that in-vitro investigated. Clinically, this co-detection was associated with both a very high IP-10 expression and a very severe pneumonia case definition. Interestingly, our translational pilot evaluation reveals IP-10 expression can be induced by several different viral and/or bacterial combinations. As immune response to each pathogen is different, further in-vitro investigations using different pathogens associations are needed to better characterize the mechanisms involved in the immunopathology of pneumonia.
In this cohort, highest serum IP-10 levels were identified among patients with both several pathogen detected and severe pneumonia, suggesting a significant role of IP-10 on pneumonia pathogenesis. Indeed, high plasma levels of IP-10 have previously been reported in patients with sepsis 12 , and were associated with high mortality rate, especially among patients with CAP 34 . Additionally, the IP-10-CXCR3 axis has been related to acute immune lung injury and lymphocyte apoptosis during the development of severe acute respiratory syndrome (SARS) 35, 36 . Moreover, an in vivo study that modeled influenza and pneumococcal superinfection in mice indicated that pro-inflammatory chemokines, including IP-10, play a crucial role in influenza-induced susceptibility to lung neutrophilia, severe immunopathology and mortality 37 . In this study, markedly elevated IP-10 (92,809 pg ml −1 ) combined with the highest PCT level (74.4 pg ml −1 ) were quantified in the serum sample of a child who died, in whom S. pneumoniae (serotype 9 V) was identified in the blood (PCR and blood culture) and co-detected with Haemophilus influenzae type B in nasal aspirate. These observations suggest an interrelationship between co-detection, elevated serum IP-10 and the pathogenesis of pneumonia.
Several limitations of this pilot translational study need to be acknowledged before concluding mixed infection is related to elevated IP-10 and disease severity. Indeed, although viral shedding (e.g., of HRV and HBoV) is common in asymptomatic children, we were unable to evaluate the levels of immunomodulators in the serum samples of a control group. Moreover, although the samples were collected within the first 24 hours after admission, only a single blood sample was processed for each patient. Therefore, a larger, longitudinal study on the etiology and severity of pneumonia will be necessary to confirm these results. In conclusion, the present findings suggest that mixed respiratory infections and IP-10 may play major, interconnected roles in the pathogenesis of pneumonia. Clinically, assessment and monitoring of induced IP-10 serum level may assist clinicians to improve diagnosis and patient management of severe community-acquired pneumonia.
Viral and bacterial strains. The 10 ng ml −1 M-CSF (Miltenyi Biotec). THP− 1 MDMs were obtained by culturing cells with 10 ng ml -1 phorbol myristate acetate (PMA; Invivogen, Toulouse, France) for 72 hours. Human airway epithelial cells (HAEC, bronchial cell type) originated from a 54-years old woman with no pathology reported (batch number MD056501) were provided by Mucilair (Epithelix, Geneva, Switzerland). Sterility, tissue integrity (TEER), mucus production and cilia beating frequency have been certified by the company. Gene expression profiling. Total cellular mRNA was purified using the RNeasy kit (Qiagen, Hilden, Germany). Reverse-transcription of total mRNA was performed using the RT 2 First Strand Kit (SABiosciences, Hilden, Germany). The expression of 84 genes involved in the human innate and adaptive immune responses was evaluated using the RT 2 profiler ™ PCR Array (SABiosciences) according to the manufacturer's recommendations. The Δ Δ Ct method was applied to calculate the fold changes in gene expression for each gene relative to uninfected control cells using the web-based RT 2 profiler PCR Array Data Analysis software (SABiosciences).
MicroRNA profiling array. Total cellular microRNAs were purified using the miRNeasy Mini kit (Qiagen) and reverse-transcribed using the miScript Reverse Transcription kit (Qiagen). The profiling of 84 miRNAs was performed using the Human Immunopathology miScript miRNA PCR Array kit (Qiagen) according to the manufacturer's instructions. Data were analyzed using the miScript miRNA PCR array data analysis web portal.
In silico miRNA target prediction. MiRNA target genes were retrieved and compiled using TargetScan 38 and microRNA.org resource 39 . The interactions between miRNAs and intracellular pathways were predicted using DIANA-miRPath v2.0 40 .
THP-1 MDMs were seeded in 24-well plates (0.5 × 10 6 per well) in triplicate, exposed to Influenza A H1N1 (A/Solomon islands/3/2006) virus (IAV) under serum-free conditions for 1 hour and then cultured for 3 hours in fresh RPMI-1640 containing 2% FBS. Streptococcus pneumoniae (SP) serotype 14 was added at 4 hours after IAV infection. Gentamicin (10 μ g ml −1 ) was added 2 hours after SP infection (i.e. 6 hours post-influenza infection) and maintained in the culture media throughout the experiment to kill extracellular bacteria and limit bacterial growth. Cell viability was determined by flow-cytometry using the FITC/Annexin V apoptosis detection kit (BD Biosciences), according to the manufacturer's instructions. #4427975) . In this assay, fold changes have been defined by the Δ Δ Ct method using control RNU-44 and -48 as reference microRNAs. Total mRNA was purified from transfected and infected MDMs using the RNeasy kit (Qiagen) and specific primers were used to amplify transforming growth factor beta-2 (TGFB2; F: 5′ -CCATCCCGCCCACTTTCTAC-3′ , R: 5′ -AGCTCAATCCGTTGTTCAGGC-3′ ), SOCS6 (F: 5′ -AAGAATTCATCCCTTGGATTAGGTAAC-3′ , R: 5′ -CAGACTGGAGGTCGTGGAA-3′ ) 41 43 , and 3) absence of wheezing at auscultation, and, 4) first symptoms appearing within the last 14 days, and 5) radiological confirmation of pneumonia as per WHO guidelines 44 . Based on these primary criteria defining pneumonia cases, all 74 cases were retrospectively re-evaluated according to the WHO "Pocket book of hospital care for children" 45 criteria to evaluate pneumonia severity. Cases that died during the study, or who had at least one additional clinical signs including central cyanosis, dullness to percussion during chest examination, prostration/lethargy, pleural effusion observed on chest radiography were retrospectively included in the severe pneumonia group. Patients without any of these additional clinical signs were included in the non-severe pneumonia group. Table 4 . a IP-10 values are expressed in pg ml -1 . IP-10 concentration differences between groups were compared using unpaired Mann-Whitney tests; significant changes (P < 0.05) are in bold. Clinical and molecular analysis. Nasopharyngeal aspirates (NAs) and whole blood samples were collected from children within 24 hours of admission. Whole blood samples were used for complete blood counts, blood culture and multiplex real-time PCR to identify Staphylococcus aureus, Streptococcus pneumoniae and Haemophilus influenzae type B 46 . S. pneumoniae serotypes were defined using a 11 multiplex real-time PCR assay targeting the 40 most frequently represented serotypes or serogroups according to protocol developed by Messaoudi et al. 47 . Serum C-reactive protein (CRP; AssayPro, St. Charles, Missouri, United States) and Procalcitonin (PCT; VIDAS B.R.A.H.M.S; bioMérieux) were quantified from whole-blood samples. Multiplex real-time non quantitative PCR (Fast-Track Diagnostic, Sliema, Malta) was used to detect 19 viruses and five bacteria in the respiratory specimens (NAs and pleural effusions). Mixed detection was defined as 1) PCR-positive for multiple viruses in NAs, 2) positive blood culture or PCR-positive for multiple bacteria in blood or 3) PCR-positive for one or multiple viruses in NAs and one or multiple bacteria in blood (identified by PCR and blood culture).
Ethical approval. The study protocol, informed consent statement, clinical research form, any amendments and all other study documents were submitted to and approved by the Ethical Committee of the Instituto de Investigaciones en Ciencias de la Salud, the Universidad Nacional de Asunción (IICS-UNA) and the Hospital Pediátrico Niños de Acosta Ñu. Informed consent was obtained from all subjects involved in this study. The clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki.
Statistical analysis. The Chi-square test and Fisher's exact test were used to compare categorical variables; continuous variables and non-normally distributed data were compared using the Mann-Whitney U-test; normally distributed data were compared using unpaired t-tests. Comparative analyses between experimental conditions (i.e., MOCK, IAV, SP or IAV + SP) were performed using one-way ANOVA with Tukey's post-hoc test or Kruskal-Wallis analysis with Dunn's post-hoc tests. Receiver operating curve (ROC) analysis was used to determine the optimal cut-off value for IP-10 to differentiate between non-severe and severe pneumonia cases. P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism (La Jolla, California, United States).
|
What followup is needed to confirm the results of the current study?
| 5,213 |
a larger, longitudinal study on the etiology and severity of pneumonia
| 23,800 |
1,584 |
Viral and bacterial co-infection in severe pneumonia triggers innate immune responses and specifically enhances IP-10: a translational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5138590/
SHA: ef3d6cabc804e5eb587b34249b539c1b5efa4cc4
Authors: Hoffmann, Jonathan; Machado, Daniela; Terrier, Olivier; Pouzol, Stephane; Messaoudi, Mélina; Basualdo, Wilma; Espínola, Emilio E; Guillen, Rosa M.; Rosa-Calatrava, Manuel; Picot, Valentina; Bénet, Thomas; Endtz, Hubert; Russomando, Graciela; Paranhos-Baccalà, Gláucia
Date: 2016-12-06
DOI: 10.1038/srep38532
License: cc-by
Abstract: Mixed viral and bacterial infections are widely described in community-acquired pneumonia; however, the clinical implications of co-infection on the associated immunopathology remain poorly studied. In this study, microRNA, mRNA and cytokine/chemokine secretion profiling were investigated for human monocyte-derived macrophages infected in-vitro with Influenza virus A/H1N1 and/or Streptococcus pneumoniae. We observed that the in-vitro co-infection synergistically increased interferon-γ-induced protein-10 (CXCL10, IP-10) expression compared to the singly-infected cells conditions. We demonstrated that endogenous miRNA-200a-3p, whose expression was synergistically induced following co-infection, indirectly regulates CXCL10 expression by targeting suppressor of cytokine signaling-6 (SOCS-6), a well-known regulator of the JAK-STAT signaling pathway. Additionally, in a subsequent clinical pilot study, immunomodulators levels were evaluated in samples from 74 children (≤5 years-old) hospitalized with viral and/or bacterial community-acquired pneumonia. Clinically, among the 74 cases of pneumonia, patients with identified mixed-detection had significantly higher (3.6-fold) serum IP-10 levels than those with a single detection (P = 0.03), and were significantly associated with severe pneumonia (P < 0.01). This study demonstrates that viral and bacterial co-infection modulates the JAK-STAT signaling pathway and leads to exacerbated IP-10 expression, which could play a major role in the pathogenesis of pneumonia.
Text: Scientific RepoRts | 6:38532 | DOI: 10 .1038/srep38532 pathogenesis of several diseases and has been suggested as a potential biomarker of viral infection 10, 11 , late-onset bacterial infection in premature infants 12 , and a promising biomarker of sepsis and septic shock 13, 14 . Combined analysis of IP-10 and IFN-γ has also been reported as a useful biomarker for diagnosis and monitoring therapeutic efficacy in patients with active tuberculosis [15] [16] [17] , and both remain detectable in the urine of patients with pulmonary diseases in the absence of renal dysfunction 18 .
With airway epithelial cells 19 , resident alveolar macrophages (AMs) and blood monocytes-derived macrophages (recruited into tissues under inflammatory conditions 20, 21 ) represent a major line of defense against both pneumococcal (through their high phagocytic capacity [22] [23] [24] ) and influenza infection 25, 26 . So far, no studies have yet focused on the intracellular mechanisms that regulate IP-10 in human blood leukocytes during mixed IAV and SP infection. Several studies indicated that host non-coding small RNAs (including microRNAs) may function as immunomodulators by regulating several pivotal intracellular processes, such as the innate immune response 27 and antiviral activity 28, 29 ; both of these processes are closely related to toll-like receptor (TLR) signaling pathways.
In this study, we firstly investigated the in vitro intracellular mechanisms that mediate the innate immune response in IAV and/or SP infected human monocyte-derived macrophages (MDMs). Using this approach, we observed that mixed-infection of MDMs induces a synergistic production of IP-10 which can be related to a miRNA-200a/JAK-STAT/SOCS-6 regulatory pathway. Subsequently, in a retrospective analysis of clinical samples collected from children ≤ 5 years-old hospitalized with pneumonia, we confirmed that serum IP-10 level could be related to both viral and/or bacterial etiologies and disease severity.
Characteristics of MDMs infected by IAV and/or SP. Initially, we investigated in vitro the impact of single and mixed IAV and SP infection on MDMs. Firstly, active replication of IAV was assessed by qRT-PCR and quantification of new infectious viral particles in the cell supernatants ( Fig. 1a,b ). IAV titer increased over time after single infection with IAV and correlated with increased production of negative-strand IAV RNA. Maximum viral replication was observed at 18-24 hours post-infection, after which time both RNA replication and the quantity of infectious particles decreased. In this in vitro model, subsequent challenge of IAV-infected MDMs with SP had no significant impact on the production of new infectious viral particles (Fig. 1b) . Together, these results indicate permissive and productive infection of MDMs by IAV. Secondly, we evaluated whether MDMs are permissive for both IAV and SP infection. The presence of pneumococci within IAV-and SP-infected primary MDMs was confirmed at 8 h post-infection (Fig. 1c) , suggesting that MDMs are permissive for viral and bacterial co-infection in the early steps of infection. Importantly, confocal co-detection of mixed IAV and SP was only effective following 8 h post-infection due to the bactericidal impact of SP internalization within human macrophages (after 24 h, data not shown). Thirdly, we evaluated the impact of single and mixed infection with IAV and SP on MDM viability. Mixed infection significantly decreased cell viability (65.2 ± 4.5% total cell death at 48 hours post-infection; P < 0.0001) compared to single SP and IAV infection (39.6 ± 1.7% and 17.4 ± 1.1% total cell death, respectively; Fig. 1d ). Taken together, these results confirmed human MDMs are permissive to mixed viral and bacterial infection. mRNA, microRNA and protein expression profiling reveal an overall induction of the host innate immune response following IAV and/or SP infection of MDMs. To investigate the innate immune response orchestrated by IAV-and SP-infected human MDMs, we firstly evaluated the expression of 84 genes involved in the innate and adaptive immune responses (Table S1) ; the major differentially-expressed genes are summarized in Fig. 2a . Expression profiling indicated an overall induction of genes related to the JAK-STAT, NF-Κ β and TLR signaling pathways. Indeed, all interferon-stimulated genes (ISGs) screened, including CXCL10 (fold-change [FC] = 240.9), CCL-2 (FC = 34.2) and MX-1 (FC = 151.4) were upregulated following mixed infection compared to uninfected cells, most of which are closely related to STAT-1 (FC = 52.3), IRF-7 (FC = 6.8) and IFNB1 (FC = 5.2) also found upregulated in mixed infected cells. Secondly, we investigated the endogenous microRNA expression profiles of IAV-and SP-infected MDMs. A selection of microRNAs that were found to be differentially-expressed under different infection conditions are shown in Fig. 2b and Table S2 . MiRNA-200a-3p was overexpressed after both single IAV (FC = 6.9), single SP (FC = 3.7) and mixed IAV/SP infection (FC = 7.3), indicating this miRNA may play a role in the innate immune response to viral and bacterial co-infection. Similar miRNA-200a-3p dysregulation profiles were obtained following IAV and/or SP infections of human macrophages-like (THP-1 monocytes-derived macrophages) or primary MDMs (data not shown). Thirdly, the secreted levels of various antiviral, pro-inflammatory and immunomodulatory cytokines/chemokines were assayed in IAV-and SP-infected-THP-1 and primary MDM cell supernatants. We observed a remarkable correlation between the mRNA and protein expression profiles of single or mixed infected MDMs especially regarding CXCL-10 and IP-10 expression. Indeed, the level of IP-10 was synergistically increased in the supernatant of IAV-infected THP-1 MDMs exposed to SP (mean: 30,589 ± 16,484 pg ml −1 ) compared to single IAV infection (1,439 ± 566.5 pg ml −1 ) and single SP infection (4,472 ± 2,001 pg ml −1 ; P≤ 0.05; Fig. 2c ) at 24 hours after infection. In those cells, IP-10 expression reduced over time (48 to 72 hours), coinciding with a significant higher proportion of necrotic and apoptotic cells (Fig. 1d) . The synergistic expression of IP-10 was similarly observed at 24 hours post-infection using primary MDMs (Fig. 2d) . Significantly increased secretion of the other tested cytokines and chemokines was not observed post-infection, even in mixed infected MDMs (Fig. S1 ). Interestingly, a significant production of IP-10 was also observed in supernatants of primary human airway epithelial cells (HAEC) mixed-infected by IAV and SP compared to the single infections (Fig. 2e) . Taken together, the mRNA and protein profiling results suggested that mixed viral and bacterial infection of MDMs induces a synergistic pro-inflammatory response related to the type-1 interferon and JAK-STAT signaling pathways, with IP-10 as signature of IAV/SP co-infection. Among all microRNAs screened, miR-200a-3p was the most Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 overexpressed in IAV/SP co-infection of human MDMs. In the remainder of this study, we decided to investigate the interconnection between miR-200a-3p expression and the innate immune response.
Endogenous miRNA-200a-3p expression correlates with CXCL10 (IP-10) induction following mixed IAV and SP infection of human MDMs. Using a specific Taqman probe assay targeting miR-200a-3p, we confirmed a significant upregulation of miR-200a-3p following mixed IAV and SP infection of human MDMs (Fig. 3a) . In this experiment, a more marked up-regulation of miR-200a-3p was observed following IAV+ SP compared to results obtained previously (Fig. 2b) . This discrepancy has been attributed to the use of two different approaches to quantify miR-200a-3p expression. The use of a target-specific stem-loop reverse transcription primer in Fig. 3a allows a better sensitivity of miR-200a-3p detection compared to the non-specific fluorescent dye used in Fig. 2b . As the general trend was suggestive of a synergistic induction of miR-200a-3p in response to mixed infection (Fig. 3a) , we hypothesized microRNA-200a-3p may play a role in the regulation of CXCL10 (IP-10), which was also synergistically upregulated in mixed-infected MDMs ( Fig. 2c and d) and primary HAEC ( Statistical analyses were performed using two-way ANOVA with Tukey's post-hoc test; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.
Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 CXCL10 (Fig. 3d) . These results suggested miR-200a-3p indirectly regulates CXCL10 and led us to hypothesize that miR-200a-3p controls a potential repressor of the JAK-STAT signaling pathway. . At 18 h after transfection, the MDMs were singly or mixed infected as described previously. At 8 h post-IAV and/or SP infection, total mRNA was extracted and amplified by PCR using specific primers for the indicated genes. Values represent median ± IQR (a, c) or mean ± SEM (d, e) of three biological replicates. Statistical analyses were performed using a Kruskal-Wallis test (non-parametric, one-way ANOVA with Dunn's post-hoc test) for data presented in (a, c). An ordinary two-way ANOVA (with Tukey's post-hoc multiple comparison test) was used for data presented in (d, e). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. MiRNA-200a-3p indirectly regulates IP-10 expression by targeting SOCS6. As shown in Fig. 2a , several JAK-STAT signaling pathway genes were deregulated in mixed IAV-and SP-infected human MDMs; therefore, we hypothesized that miR-200a-3p directly regulates a regulator of the JAK-STAT signaling pathway. Predictive target analysis indicated that the 3' UTR of suppressor of cytokine signaling-6 (SOCS6) may be targeted by miR-200a-3p (Fig. 3b) . SOCS proteins constitute a class of negative regulators of JAK-STAT signaling pathways that are induced by both cytokines and TLR signaling. MiRNA-200a-3p was not predicted to target any of the other six members of the SOCS gene family. Transfection of human MDMs with MIM-200a downregulated SOCS6 (FC = 0.57) while inhibition of miR-200a-3p (INH-200a) upregulated SOCS6 (FC = 1.55), confirming that miR-200a-3p effectively regulates the expression of SOCS6 (Fig. 3e) . Moreover, SOCS6 was synergistically downregulated in IAV-or IAV/SP-infected MDMs overexpressing miRNA-200a (Fig. 3e) , suggesting that both infection and miR-200a-3p negatively regulate the expression of SOCS6. Finally, western blotting confirmed that expression of SOCS-6 sharply reduced following infection, especially after mixed IAV and SP infection (Fig. 3f) .
These results indicate miR-200a-3p is strongly induced in response to mixed viral and bacterial co-infection, which in turn leads to downregulation of the JAK-STAT regulator SOCS-6 at both the mRNA and protein levels and subsequent upregulation of IP-10.
analyses demonstrated mixed IAV and SP infection of human MDMs and HAEC induced significant production of IP-10. As blood leukocytes and respiratory tract epithelial cells actively contribute to inflammation during pneumonia, we hypothesized the level of IP-10 in serum of patient with pneumonia may be both indicative of mixed respiratory infection and disease severity. As part of a prospective, hospital-based, multicenter case-control study on the etiology of pneumonia among children under 5-years-old, a total of 74 patients (44 male, 30 female) were included in this pilot evaluation. According to WHO guidelines, retrospective analysis indicated 44 (59.5%) children had clinical signs of non-severe pneumonia and 30 (40.5%) children had signs of severe pneumonia. The main patient characteristics at inclusion are shown in Table 1 . Patients with severe pneumonia had significant more recorded episodes of dyspnea (P < 0.001), cyanosis (P = 0.03), lower chest indrawing (P < 0.001), dullness to percussion (P < 0.001) and lethargy (P < 0.001) during chest examination than patient with non-severe pneumonia. Moreover, pleural effusions were significantly more observed among critically ill patients and the duration of hospitalization was significantly longer for the children with severe pneumonia than for those with non-severe pneumonia (P = 0.0015). Two deaths occurred within the group of children retrospectively defined with severe pneumonia. Evaluation of the systemic inflammatory response of the 74 cases is shown in Table 2 . Serum level of CRP, IP-10, PCT, G-CSF, IL-6, IL-8 and MIP-1β were significantly more elevated in serum samples from critically ill patients. Patients with severe pneumonia had significantly higher (4.2-fold) serum IP-10 levels than those with a non-severe pneumonia (P < 0.001) suggesting IP-10 as a promising prognostic marker in pneumonia. Diagnostic accuracy measures for predicting pneumonia severity using blood-based biomarkers are summarized in Table S3 . Briefly, in this study, the optimal IP-10 cut-off value for identifying patient with severe pneumonia was 4,240 pg ml −1 , with an area under the receiver operating characteristic curve of 0.69 (95% CI, 0.57 to 0.82, P < 0.001). Defining as positive a serum IP-10 level above this cut-off resulted in a sensitivity of 63.3%, specificity of 63.6% and a positive likelihood ratio of 1.74. Prognostic values of IP-10 were closed to procalcitonin (PCT; AUC = 0.70; 95% IC, 0.58 to 0.82, P < 0.001) and IL-6 (AUC = 0.70; 95% IC, 0.58-0.83, P < 0.001).
Multiplex PCR-based screening of respiratory and blood samples reveal a high variety of pathogen associations (Table 3) . Respiratory viruses were detected in the nasal aspirates (NAs) of 63/74 patients (85.1%). Etiological bacteria of pneumonia (S. pneumoniae, n = 19; S. aureus, n = 1; or H. influenzae type B, n = 7) were identified via real-time PCR in the blood samples of 27/74 (36.5%) of the patients. Multiplex PCR assays allowed the identification of respiratory bacteria in the blood of 19 patients with negative blood culture results. Among the 74 cases PCR-positive for respiratory pathogens, a single virus or bacteria were detected in the NAs of 7 (9.4%) and 3 (4.0%) patients, respectively; these 10/74 (13.5%) cases were defined as the single infection group. The mixed infection group included the 62/74 (83.8%) cases in which (1) multiple viruses and/or bacteria were identified in NAs (38/74; 51.3%) without any bacteria identified in blood samples or (2) one or more viruses and/or bacteria were identified in NAs and associated with a blood bacteremia (24/74; 32.4%). We evaluated whether IP-10 serum level could correlate with the viral and bacterial etiologies of pneumonia. Patients with mixed infection had significant higher (3.6-fold) IP-10 serum level than patient with single detection (P = 0.03; Table 4 ). A stratified analysis reveals that the highest IP-10 serum level was observed among patients with both several respiratory pathogens identified (mixed-detection group) and severe pneumonia (14,427 pg ml −1 , IQR (3,981-82,994). In detail, a remarkable IP-10 serum level (142,531 pg ml −1 ), representing 33-fold higher above cut-off value predicting pneumonia severity was observed in patient with hRV in NA co-detected with S. pneumoniae (serotype 14) in pleural effusion and blood. In concordance with our in-vitro model of co-infection, a significant IP-10 level (90,338 pg ml −1 ) was quantified in blood sample of patient with severe bacteremic pneumococcal (serotype 14) pneumonia with a positive co-detection of Influenza B virus in NA. Taken together, these results suggest that high serum IP-10 levels are significantly associated with mixed viral and bacterial detection and also related to pneumonia pathogenesis.
This study provides additional in vitro and clinical data to improve our understanding of the immunopathology of mixed viral and bacterial pneumonia (Fig. 4) .
The in vitro model of influenza and pneumococcal superinfection of human MDMs demonstrated that mixed infection synergistically induced release of the pro-inflammatory chemokine IP-10, strongly suggesting human Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 blood leukocytes contribute to the immunopathology of pneumonia. Additionally, transcriptomics and omics analyses provided new data on the inflammatory pathways that are activated during mixed infection and related to synergistic induction of the pro-inflammatory chemokine IP-10 in mixed infected cells. Our observations are consistent with a recent study describing IP-10 induction as host-proteome signature of both viral and bacterial infections 30 . Of the differentially-expressed genes observed in mixed infected MDMs, the transcription factors STAT-1 and IRF-7 appear to play crucial roles in the regulation of interferon-stimulated genes including CXCL10 (IP-10). By focusing on the intracellular mechanisms that regulate inflammatory pathways, we demonstrated a novel role for miRNA-200a-3p in the regulation of CXCL10 (IP-10). These observations are consistent with previous reports showing that RNA virus infection upregulates miR-155 in macrophages and dendritic cells and also regulates suppressor of cytokine signaling 1 (SOCS1), suggesting the existence of a miRNA/JAK-STAT/SOCS regulatory pathway during viral infection 29 . Our study suggests co-infection leads to overexpression of miR-200a-3p, which in turn targets and downregulates the JAK-STAT regulator SOCS-6 and consequently increases CXCL10 (IP-10) expression. Interestingly, a complementary in-silico approach reveals that several microRNAs that were found dysregulated in our experiments of IAV and SP co-infection of MDMs or HAEC, might target several genes of SOCS family and play similar role than miR-200a-3p. Indeed, miRNA-142-3p might target SOCS4, 5, 6 mRNA while miRNA-194-5p might target SOCS2, 3, 4, 5 and 7 mRNA. These observations underline that intra-cellular regulation of IP-10 is not limited to the contribution of a sole microRNA. A complex inter-relationship between numerous host microRNAs and inhibitors of the JAK-STAT signaling pathway occur to control host innate inflammatory response against viral and/or bacterial infections. Clinically, the majority of pediatric CAP cases in this study were associated with both positive viral and/or bacterial detection. Respiratory microorganisms were detected in 97% of cases; 51.3% of which were viral-viral, viral-bacterial or bacterial-bacterial co-detected only in nasal aspirates, 32.4% of which co-detected in both nasal aspirates and blood samples. These data are consistent with previous etiological studies of pediatric CAP 3,31-33 . S. pneumoniae was the major bacteria identified in blood (19/74; 25.7%) and mainly co-detected with respiratory viruses in NAs (16/19; 84.2%). We observed a very high diversity of viral and bacterial associations in biological samples from children with pneumonia. In comparison with IAV and SP14 combination evaluated in-vitro, no pneumonia cases were singly influenza and pneumococcus infected, and no similar co-detection with those two pathogens has been clinically observed. Nevertheless, Influenza B (IVB) virus was identified in 5 patients and two of them had a positive SP co-detection in blood (one non-typable strain and one serotype 14 using our molecular typing test). IVB and SP14 combination seems to be the nearest pathogen co-detection to that in-vitro investigated. Clinically, this co-detection was associated with both a very high IP-10 expression and a very severe pneumonia case definition. Interestingly, our translational pilot evaluation reveals IP-10 expression can be induced by several different viral and/or bacterial combinations. As immune response to each pathogen is different, further in-vitro investigations using different pathogens associations are needed to better characterize the mechanisms involved in the immunopathology of pneumonia.
In this cohort, highest serum IP-10 levels were identified among patients with both several pathogen detected and severe pneumonia, suggesting a significant role of IP-10 on pneumonia pathogenesis. Indeed, high plasma levels of IP-10 have previously been reported in patients with sepsis 12 , and were associated with high mortality rate, especially among patients with CAP 34 . Additionally, the IP-10-CXCR3 axis has been related to acute immune lung injury and lymphocyte apoptosis during the development of severe acute respiratory syndrome (SARS) 35, 36 . Moreover, an in vivo study that modeled influenza and pneumococcal superinfection in mice indicated that pro-inflammatory chemokines, including IP-10, play a crucial role in influenza-induced susceptibility to lung neutrophilia, severe immunopathology and mortality 37 . In this study, markedly elevated IP-10 (92,809 pg ml −1 ) combined with the highest PCT level (74.4 pg ml −1 ) were quantified in the serum sample of a child who died, in whom S. pneumoniae (serotype 9 V) was identified in the blood (PCR and blood culture) and co-detected with Haemophilus influenzae type B in nasal aspirate. These observations suggest an interrelationship between co-detection, elevated serum IP-10 and the pathogenesis of pneumonia.
Several limitations of this pilot translational study need to be acknowledged before concluding mixed infection is related to elevated IP-10 and disease severity. Indeed, although viral shedding (e.g., of HRV and HBoV) is common in asymptomatic children, we were unable to evaluate the levels of immunomodulators in the serum samples of a control group. Moreover, although the samples were collected within the first 24 hours after admission, only a single blood sample was processed for each patient. Therefore, a larger, longitudinal study on the etiology and severity of pneumonia will be necessary to confirm these results. In conclusion, the present findings suggest that mixed respiratory infections and IP-10 may play major, interconnected roles in the pathogenesis of pneumonia. Clinically, assessment and monitoring of induced IP-10 serum level may assist clinicians to improve diagnosis and patient management of severe community-acquired pneumonia.
Viral and bacterial strains. The 10 ng ml −1 M-CSF (Miltenyi Biotec). THP− 1 MDMs were obtained by culturing cells with 10 ng ml -1 phorbol myristate acetate (PMA; Invivogen, Toulouse, France) for 72 hours. Human airway epithelial cells (HAEC, bronchial cell type) originated from a 54-years old woman with no pathology reported (batch number MD056501) were provided by Mucilair (Epithelix, Geneva, Switzerland). Sterility, tissue integrity (TEER), mucus production and cilia beating frequency have been certified by the company. Gene expression profiling. Total cellular mRNA was purified using the RNeasy kit (Qiagen, Hilden, Germany). Reverse-transcription of total mRNA was performed using the RT 2 First Strand Kit (SABiosciences, Hilden, Germany). The expression of 84 genes involved in the human innate and adaptive immune responses was evaluated using the RT 2 profiler ™ PCR Array (SABiosciences) according to the manufacturer's recommendations. The Δ Δ Ct method was applied to calculate the fold changes in gene expression for each gene relative to uninfected control cells using the web-based RT 2 profiler PCR Array Data Analysis software (SABiosciences).
MicroRNA profiling array. Total cellular microRNAs were purified using the miRNeasy Mini kit (Qiagen) and reverse-transcribed using the miScript Reverse Transcription kit (Qiagen). The profiling of 84 miRNAs was performed using the Human Immunopathology miScript miRNA PCR Array kit (Qiagen) according to the manufacturer's instructions. Data were analyzed using the miScript miRNA PCR array data analysis web portal.
In silico miRNA target prediction. MiRNA target genes were retrieved and compiled using TargetScan 38 and microRNA.org resource 39 . The interactions between miRNAs and intracellular pathways were predicted using DIANA-miRPath v2.0 40 .
THP-1 MDMs were seeded in 24-well plates (0.5 × 10 6 per well) in triplicate, exposed to Influenza A H1N1 (A/Solomon islands/3/2006) virus (IAV) under serum-free conditions for 1 hour and then cultured for 3 hours in fresh RPMI-1640 containing 2% FBS. Streptococcus pneumoniae (SP) serotype 14 was added at 4 hours after IAV infection. Gentamicin (10 μ g ml −1 ) was added 2 hours after SP infection (i.e. 6 hours post-influenza infection) and maintained in the culture media throughout the experiment to kill extracellular bacteria and limit bacterial growth. Cell viability was determined by flow-cytometry using the FITC/Annexin V apoptosis detection kit (BD Biosciences), according to the manufacturer's instructions. #4427975) . In this assay, fold changes have been defined by the Δ Δ Ct method using control RNU-44 and -48 as reference microRNAs. Total mRNA was purified from transfected and infected MDMs using the RNeasy kit (Qiagen) and specific primers were used to amplify transforming growth factor beta-2 (TGFB2; F: 5′ -CCATCCCGCCCACTTTCTAC-3′ , R: 5′ -AGCTCAATCCGTTGTTCAGGC-3′ ), SOCS6 (F: 5′ -AAGAATTCATCCCTTGGATTAGGTAAC-3′ , R: 5′ -CAGACTGGAGGTCGTGGAA-3′ ) 41 43 , and 3) absence of wheezing at auscultation, and, 4) first symptoms appearing within the last 14 days, and 5) radiological confirmation of pneumonia as per WHO guidelines 44 . Based on these primary criteria defining pneumonia cases, all 74 cases were retrospectively re-evaluated according to the WHO "Pocket book of hospital care for children" 45 criteria to evaluate pneumonia severity. Cases that died during the study, or who had at least one additional clinical signs including central cyanosis, dullness to percussion during chest examination, prostration/lethargy, pleural effusion observed on chest radiography were retrospectively included in the severe pneumonia group. Patients without any of these additional clinical signs were included in the non-severe pneumonia group. Table 4 . a IP-10 values are expressed in pg ml -1 . IP-10 concentration differences between groups were compared using unpaired Mann-Whitney tests; significant changes (P < 0.05) are in bold. Clinical and molecular analysis. Nasopharyngeal aspirates (NAs) and whole blood samples were collected from children within 24 hours of admission. Whole blood samples were used for complete blood counts, blood culture and multiplex real-time PCR to identify Staphylococcus aureus, Streptococcus pneumoniae and Haemophilus influenzae type B 46 . S. pneumoniae serotypes were defined using a 11 multiplex real-time PCR assay targeting the 40 most frequently represented serotypes or serogroups according to protocol developed by Messaoudi et al. 47 . Serum C-reactive protein (CRP; AssayPro, St. Charles, Missouri, United States) and Procalcitonin (PCT; VIDAS B.R.A.H.M.S; bioMérieux) were quantified from whole-blood samples. Multiplex real-time non quantitative PCR (Fast-Track Diagnostic, Sliema, Malta) was used to detect 19 viruses and five bacteria in the respiratory specimens (NAs and pleural effusions). Mixed detection was defined as 1) PCR-positive for multiple viruses in NAs, 2) positive blood culture or PCR-positive for multiple bacteria in blood or 3) PCR-positive for one or multiple viruses in NAs and one or multiple bacteria in blood (identified by PCR and blood culture).
Ethical approval. The study protocol, informed consent statement, clinical research form, any amendments and all other study documents were submitted to and approved by the Ethical Committee of the Instituto de Investigaciones en Ciencias de la Salud, the Universidad Nacional de Asunción (IICS-UNA) and the Hospital Pediátrico Niños de Acosta Ñu. Informed consent was obtained from all subjects involved in this study. The clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki.
Statistical analysis. The Chi-square test and Fisher's exact test were used to compare categorical variables; continuous variables and non-normally distributed data were compared using the Mann-Whitney U-test; normally distributed data were compared using unpaired t-tests. Comparative analyses between experimental conditions (i.e., MOCK, IAV, SP or IAV + SP) were performed using one-way ANOVA with Tukey's post-hoc test or Kruskal-Wallis analysis with Dunn's post-hoc tests. Receiver operating curve (ROC) analysis was used to determine the optimal cut-off value for IP-10 to differentiate between non-severe and severe pneumonia cases. P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism (La Jolla, California, United States).
|
What is the conclusion of this study?
| 5,214 |
mixed respiratory infections and IP-10 may play major, interconnected roles in the pathogenesis of pneumonia
| 23,964 |
1,584 |
Viral and bacterial co-infection in severe pneumonia triggers innate immune responses and specifically enhances IP-10: a translational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5138590/
SHA: ef3d6cabc804e5eb587b34249b539c1b5efa4cc4
Authors: Hoffmann, Jonathan; Machado, Daniela; Terrier, Olivier; Pouzol, Stephane; Messaoudi, Mélina; Basualdo, Wilma; Espínola, Emilio E; Guillen, Rosa M.; Rosa-Calatrava, Manuel; Picot, Valentina; Bénet, Thomas; Endtz, Hubert; Russomando, Graciela; Paranhos-Baccalà, Gláucia
Date: 2016-12-06
DOI: 10.1038/srep38532
License: cc-by
Abstract: Mixed viral and bacterial infections are widely described in community-acquired pneumonia; however, the clinical implications of co-infection on the associated immunopathology remain poorly studied. In this study, microRNA, mRNA and cytokine/chemokine secretion profiling were investigated for human monocyte-derived macrophages infected in-vitro with Influenza virus A/H1N1 and/or Streptococcus pneumoniae. We observed that the in-vitro co-infection synergistically increased interferon-γ-induced protein-10 (CXCL10, IP-10) expression compared to the singly-infected cells conditions. We demonstrated that endogenous miRNA-200a-3p, whose expression was synergistically induced following co-infection, indirectly regulates CXCL10 expression by targeting suppressor of cytokine signaling-6 (SOCS-6), a well-known regulator of the JAK-STAT signaling pathway. Additionally, in a subsequent clinical pilot study, immunomodulators levels were evaluated in samples from 74 children (≤5 years-old) hospitalized with viral and/or bacterial community-acquired pneumonia. Clinically, among the 74 cases of pneumonia, patients with identified mixed-detection had significantly higher (3.6-fold) serum IP-10 levels than those with a single detection (P = 0.03), and were significantly associated with severe pneumonia (P < 0.01). This study demonstrates that viral and bacterial co-infection modulates the JAK-STAT signaling pathway and leads to exacerbated IP-10 expression, which could play a major role in the pathogenesis of pneumonia.
Text: Scientific RepoRts | 6:38532 | DOI: 10 .1038/srep38532 pathogenesis of several diseases and has been suggested as a potential biomarker of viral infection 10, 11 , late-onset bacterial infection in premature infants 12 , and a promising biomarker of sepsis and septic shock 13, 14 . Combined analysis of IP-10 and IFN-γ has also been reported as a useful biomarker for diagnosis and monitoring therapeutic efficacy in patients with active tuberculosis [15] [16] [17] , and both remain detectable in the urine of patients with pulmonary diseases in the absence of renal dysfunction 18 .
With airway epithelial cells 19 , resident alveolar macrophages (AMs) and blood monocytes-derived macrophages (recruited into tissues under inflammatory conditions 20, 21 ) represent a major line of defense against both pneumococcal (through their high phagocytic capacity [22] [23] [24] ) and influenza infection 25, 26 . So far, no studies have yet focused on the intracellular mechanisms that regulate IP-10 in human blood leukocytes during mixed IAV and SP infection. Several studies indicated that host non-coding small RNAs (including microRNAs) may function as immunomodulators by regulating several pivotal intracellular processes, such as the innate immune response 27 and antiviral activity 28, 29 ; both of these processes are closely related to toll-like receptor (TLR) signaling pathways.
In this study, we firstly investigated the in vitro intracellular mechanisms that mediate the innate immune response in IAV and/or SP infected human monocyte-derived macrophages (MDMs). Using this approach, we observed that mixed-infection of MDMs induces a synergistic production of IP-10 which can be related to a miRNA-200a/JAK-STAT/SOCS-6 regulatory pathway. Subsequently, in a retrospective analysis of clinical samples collected from children ≤ 5 years-old hospitalized with pneumonia, we confirmed that serum IP-10 level could be related to both viral and/or bacterial etiologies and disease severity.
Characteristics of MDMs infected by IAV and/or SP. Initially, we investigated in vitro the impact of single and mixed IAV and SP infection on MDMs. Firstly, active replication of IAV was assessed by qRT-PCR and quantification of new infectious viral particles in the cell supernatants ( Fig. 1a,b ). IAV titer increased over time after single infection with IAV and correlated with increased production of negative-strand IAV RNA. Maximum viral replication was observed at 18-24 hours post-infection, after which time both RNA replication and the quantity of infectious particles decreased. In this in vitro model, subsequent challenge of IAV-infected MDMs with SP had no significant impact on the production of new infectious viral particles (Fig. 1b) . Together, these results indicate permissive and productive infection of MDMs by IAV. Secondly, we evaluated whether MDMs are permissive for both IAV and SP infection. The presence of pneumococci within IAV-and SP-infected primary MDMs was confirmed at 8 h post-infection (Fig. 1c) , suggesting that MDMs are permissive for viral and bacterial co-infection in the early steps of infection. Importantly, confocal co-detection of mixed IAV and SP was only effective following 8 h post-infection due to the bactericidal impact of SP internalization within human macrophages (after 24 h, data not shown). Thirdly, we evaluated the impact of single and mixed infection with IAV and SP on MDM viability. Mixed infection significantly decreased cell viability (65.2 ± 4.5% total cell death at 48 hours post-infection; P < 0.0001) compared to single SP and IAV infection (39.6 ± 1.7% and 17.4 ± 1.1% total cell death, respectively; Fig. 1d ). Taken together, these results confirmed human MDMs are permissive to mixed viral and bacterial infection. mRNA, microRNA and protein expression profiling reveal an overall induction of the host innate immune response following IAV and/or SP infection of MDMs. To investigate the innate immune response orchestrated by IAV-and SP-infected human MDMs, we firstly evaluated the expression of 84 genes involved in the innate and adaptive immune responses (Table S1) ; the major differentially-expressed genes are summarized in Fig. 2a . Expression profiling indicated an overall induction of genes related to the JAK-STAT, NF-Κ β and TLR signaling pathways. Indeed, all interferon-stimulated genes (ISGs) screened, including CXCL10 (fold-change [FC] = 240.9), CCL-2 (FC = 34.2) and MX-1 (FC = 151.4) were upregulated following mixed infection compared to uninfected cells, most of which are closely related to STAT-1 (FC = 52.3), IRF-7 (FC = 6.8) and IFNB1 (FC = 5.2) also found upregulated in mixed infected cells. Secondly, we investigated the endogenous microRNA expression profiles of IAV-and SP-infected MDMs. A selection of microRNAs that were found to be differentially-expressed under different infection conditions are shown in Fig. 2b and Table S2 . MiRNA-200a-3p was overexpressed after both single IAV (FC = 6.9), single SP (FC = 3.7) and mixed IAV/SP infection (FC = 7.3), indicating this miRNA may play a role in the innate immune response to viral and bacterial co-infection. Similar miRNA-200a-3p dysregulation profiles were obtained following IAV and/or SP infections of human macrophages-like (THP-1 monocytes-derived macrophages) or primary MDMs (data not shown). Thirdly, the secreted levels of various antiviral, pro-inflammatory and immunomodulatory cytokines/chemokines were assayed in IAV-and SP-infected-THP-1 and primary MDM cell supernatants. We observed a remarkable correlation between the mRNA and protein expression profiles of single or mixed infected MDMs especially regarding CXCL-10 and IP-10 expression. Indeed, the level of IP-10 was synergistically increased in the supernatant of IAV-infected THP-1 MDMs exposed to SP (mean: 30,589 ± 16,484 pg ml −1 ) compared to single IAV infection (1,439 ± 566.5 pg ml −1 ) and single SP infection (4,472 ± 2,001 pg ml −1 ; P≤ 0.05; Fig. 2c ) at 24 hours after infection. In those cells, IP-10 expression reduced over time (48 to 72 hours), coinciding with a significant higher proportion of necrotic and apoptotic cells (Fig. 1d) . The synergistic expression of IP-10 was similarly observed at 24 hours post-infection using primary MDMs (Fig. 2d) . Significantly increased secretion of the other tested cytokines and chemokines was not observed post-infection, even in mixed infected MDMs (Fig. S1 ). Interestingly, a significant production of IP-10 was also observed in supernatants of primary human airway epithelial cells (HAEC) mixed-infected by IAV and SP compared to the single infections (Fig. 2e) . Taken together, the mRNA and protein profiling results suggested that mixed viral and bacterial infection of MDMs induces a synergistic pro-inflammatory response related to the type-1 interferon and JAK-STAT signaling pathways, with IP-10 as signature of IAV/SP co-infection. Among all microRNAs screened, miR-200a-3p was the most Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 overexpressed in IAV/SP co-infection of human MDMs. In the remainder of this study, we decided to investigate the interconnection between miR-200a-3p expression and the innate immune response.
Endogenous miRNA-200a-3p expression correlates with CXCL10 (IP-10) induction following mixed IAV and SP infection of human MDMs. Using a specific Taqman probe assay targeting miR-200a-3p, we confirmed a significant upregulation of miR-200a-3p following mixed IAV and SP infection of human MDMs (Fig. 3a) . In this experiment, a more marked up-regulation of miR-200a-3p was observed following IAV+ SP compared to results obtained previously (Fig. 2b) . This discrepancy has been attributed to the use of two different approaches to quantify miR-200a-3p expression. The use of a target-specific stem-loop reverse transcription primer in Fig. 3a allows a better sensitivity of miR-200a-3p detection compared to the non-specific fluorescent dye used in Fig. 2b . As the general trend was suggestive of a synergistic induction of miR-200a-3p in response to mixed infection (Fig. 3a) , we hypothesized microRNA-200a-3p may play a role in the regulation of CXCL10 (IP-10), which was also synergistically upregulated in mixed-infected MDMs ( Fig. 2c and d) and primary HAEC ( Statistical analyses were performed using two-way ANOVA with Tukey's post-hoc test; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.
Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 CXCL10 (Fig. 3d) . These results suggested miR-200a-3p indirectly regulates CXCL10 and led us to hypothesize that miR-200a-3p controls a potential repressor of the JAK-STAT signaling pathway. . At 18 h after transfection, the MDMs were singly or mixed infected as described previously. At 8 h post-IAV and/or SP infection, total mRNA was extracted and amplified by PCR using specific primers for the indicated genes. Values represent median ± IQR (a, c) or mean ± SEM (d, e) of three biological replicates. Statistical analyses were performed using a Kruskal-Wallis test (non-parametric, one-way ANOVA with Dunn's post-hoc test) for data presented in (a, c). An ordinary two-way ANOVA (with Tukey's post-hoc multiple comparison test) was used for data presented in (d, e). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. MiRNA-200a-3p indirectly regulates IP-10 expression by targeting SOCS6. As shown in Fig. 2a , several JAK-STAT signaling pathway genes were deregulated in mixed IAV-and SP-infected human MDMs; therefore, we hypothesized that miR-200a-3p directly regulates a regulator of the JAK-STAT signaling pathway. Predictive target analysis indicated that the 3' UTR of suppressor of cytokine signaling-6 (SOCS6) may be targeted by miR-200a-3p (Fig. 3b) . SOCS proteins constitute a class of negative regulators of JAK-STAT signaling pathways that are induced by both cytokines and TLR signaling. MiRNA-200a-3p was not predicted to target any of the other six members of the SOCS gene family. Transfection of human MDMs with MIM-200a downregulated SOCS6 (FC = 0.57) while inhibition of miR-200a-3p (INH-200a) upregulated SOCS6 (FC = 1.55), confirming that miR-200a-3p effectively regulates the expression of SOCS6 (Fig. 3e) . Moreover, SOCS6 was synergistically downregulated in IAV-or IAV/SP-infected MDMs overexpressing miRNA-200a (Fig. 3e) , suggesting that both infection and miR-200a-3p negatively regulate the expression of SOCS6. Finally, western blotting confirmed that expression of SOCS-6 sharply reduced following infection, especially after mixed IAV and SP infection (Fig. 3f) .
These results indicate miR-200a-3p is strongly induced in response to mixed viral and bacterial co-infection, which in turn leads to downregulation of the JAK-STAT regulator SOCS-6 at both the mRNA and protein levels and subsequent upregulation of IP-10.
analyses demonstrated mixed IAV and SP infection of human MDMs and HAEC induced significant production of IP-10. As blood leukocytes and respiratory tract epithelial cells actively contribute to inflammation during pneumonia, we hypothesized the level of IP-10 in serum of patient with pneumonia may be both indicative of mixed respiratory infection and disease severity. As part of a prospective, hospital-based, multicenter case-control study on the etiology of pneumonia among children under 5-years-old, a total of 74 patients (44 male, 30 female) were included in this pilot evaluation. According to WHO guidelines, retrospective analysis indicated 44 (59.5%) children had clinical signs of non-severe pneumonia and 30 (40.5%) children had signs of severe pneumonia. The main patient characteristics at inclusion are shown in Table 1 . Patients with severe pneumonia had significant more recorded episodes of dyspnea (P < 0.001), cyanosis (P = 0.03), lower chest indrawing (P < 0.001), dullness to percussion (P < 0.001) and lethargy (P < 0.001) during chest examination than patient with non-severe pneumonia. Moreover, pleural effusions were significantly more observed among critically ill patients and the duration of hospitalization was significantly longer for the children with severe pneumonia than for those with non-severe pneumonia (P = 0.0015). Two deaths occurred within the group of children retrospectively defined with severe pneumonia. Evaluation of the systemic inflammatory response of the 74 cases is shown in Table 2 . Serum level of CRP, IP-10, PCT, G-CSF, IL-6, IL-8 and MIP-1β were significantly more elevated in serum samples from critically ill patients. Patients with severe pneumonia had significantly higher (4.2-fold) serum IP-10 levels than those with a non-severe pneumonia (P < 0.001) suggesting IP-10 as a promising prognostic marker in pneumonia. Diagnostic accuracy measures for predicting pneumonia severity using blood-based biomarkers are summarized in Table S3 . Briefly, in this study, the optimal IP-10 cut-off value for identifying patient with severe pneumonia was 4,240 pg ml −1 , with an area under the receiver operating characteristic curve of 0.69 (95% CI, 0.57 to 0.82, P < 0.001). Defining as positive a serum IP-10 level above this cut-off resulted in a sensitivity of 63.3%, specificity of 63.6% and a positive likelihood ratio of 1.74. Prognostic values of IP-10 were closed to procalcitonin (PCT; AUC = 0.70; 95% IC, 0.58 to 0.82, P < 0.001) and IL-6 (AUC = 0.70; 95% IC, 0.58-0.83, P < 0.001).
Multiplex PCR-based screening of respiratory and blood samples reveal a high variety of pathogen associations (Table 3) . Respiratory viruses were detected in the nasal aspirates (NAs) of 63/74 patients (85.1%). Etiological bacteria of pneumonia (S. pneumoniae, n = 19; S. aureus, n = 1; or H. influenzae type B, n = 7) were identified via real-time PCR in the blood samples of 27/74 (36.5%) of the patients. Multiplex PCR assays allowed the identification of respiratory bacteria in the blood of 19 patients with negative blood culture results. Among the 74 cases PCR-positive for respiratory pathogens, a single virus or bacteria were detected in the NAs of 7 (9.4%) and 3 (4.0%) patients, respectively; these 10/74 (13.5%) cases were defined as the single infection group. The mixed infection group included the 62/74 (83.8%) cases in which (1) multiple viruses and/or bacteria were identified in NAs (38/74; 51.3%) without any bacteria identified in blood samples or (2) one or more viruses and/or bacteria were identified in NAs and associated with a blood bacteremia (24/74; 32.4%). We evaluated whether IP-10 serum level could correlate with the viral and bacterial etiologies of pneumonia. Patients with mixed infection had significant higher (3.6-fold) IP-10 serum level than patient with single detection (P = 0.03; Table 4 ). A stratified analysis reveals that the highest IP-10 serum level was observed among patients with both several respiratory pathogens identified (mixed-detection group) and severe pneumonia (14,427 pg ml −1 , IQR (3,981-82,994). In detail, a remarkable IP-10 serum level (142,531 pg ml −1 ), representing 33-fold higher above cut-off value predicting pneumonia severity was observed in patient with hRV in NA co-detected with S. pneumoniae (serotype 14) in pleural effusion and blood. In concordance with our in-vitro model of co-infection, a significant IP-10 level (90,338 pg ml −1 ) was quantified in blood sample of patient with severe bacteremic pneumococcal (serotype 14) pneumonia with a positive co-detection of Influenza B virus in NA. Taken together, these results suggest that high serum IP-10 levels are significantly associated with mixed viral and bacterial detection and also related to pneumonia pathogenesis.
This study provides additional in vitro and clinical data to improve our understanding of the immunopathology of mixed viral and bacterial pneumonia (Fig. 4) .
The in vitro model of influenza and pneumococcal superinfection of human MDMs demonstrated that mixed infection synergistically induced release of the pro-inflammatory chemokine IP-10, strongly suggesting human Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 blood leukocytes contribute to the immunopathology of pneumonia. Additionally, transcriptomics and omics analyses provided new data on the inflammatory pathways that are activated during mixed infection and related to synergistic induction of the pro-inflammatory chemokine IP-10 in mixed infected cells. Our observations are consistent with a recent study describing IP-10 induction as host-proteome signature of both viral and bacterial infections 30 . Of the differentially-expressed genes observed in mixed infected MDMs, the transcription factors STAT-1 and IRF-7 appear to play crucial roles in the regulation of interferon-stimulated genes including CXCL10 (IP-10). By focusing on the intracellular mechanisms that regulate inflammatory pathways, we demonstrated a novel role for miRNA-200a-3p in the regulation of CXCL10 (IP-10). These observations are consistent with previous reports showing that RNA virus infection upregulates miR-155 in macrophages and dendritic cells and also regulates suppressor of cytokine signaling 1 (SOCS1), suggesting the existence of a miRNA/JAK-STAT/SOCS regulatory pathway during viral infection 29 . Our study suggests co-infection leads to overexpression of miR-200a-3p, which in turn targets and downregulates the JAK-STAT regulator SOCS-6 and consequently increases CXCL10 (IP-10) expression. Interestingly, a complementary in-silico approach reveals that several microRNAs that were found dysregulated in our experiments of IAV and SP co-infection of MDMs or HAEC, might target several genes of SOCS family and play similar role than miR-200a-3p. Indeed, miRNA-142-3p might target SOCS4, 5, 6 mRNA while miRNA-194-5p might target SOCS2, 3, 4, 5 and 7 mRNA. These observations underline that intra-cellular regulation of IP-10 is not limited to the contribution of a sole microRNA. A complex inter-relationship between numerous host microRNAs and inhibitors of the JAK-STAT signaling pathway occur to control host innate inflammatory response against viral and/or bacterial infections. Clinically, the majority of pediatric CAP cases in this study were associated with both positive viral and/or bacterial detection. Respiratory microorganisms were detected in 97% of cases; 51.3% of which were viral-viral, viral-bacterial or bacterial-bacterial co-detected only in nasal aspirates, 32.4% of which co-detected in both nasal aspirates and blood samples. These data are consistent with previous etiological studies of pediatric CAP 3,31-33 . S. pneumoniae was the major bacteria identified in blood (19/74; 25.7%) and mainly co-detected with respiratory viruses in NAs (16/19; 84.2%). We observed a very high diversity of viral and bacterial associations in biological samples from children with pneumonia. In comparison with IAV and SP14 combination evaluated in-vitro, no pneumonia cases were singly influenza and pneumococcus infected, and no similar co-detection with those two pathogens has been clinically observed. Nevertheless, Influenza B (IVB) virus was identified in 5 patients and two of them had a positive SP co-detection in blood (one non-typable strain and one serotype 14 using our molecular typing test). IVB and SP14 combination seems to be the nearest pathogen co-detection to that in-vitro investigated. Clinically, this co-detection was associated with both a very high IP-10 expression and a very severe pneumonia case definition. Interestingly, our translational pilot evaluation reveals IP-10 expression can be induced by several different viral and/or bacterial combinations. As immune response to each pathogen is different, further in-vitro investigations using different pathogens associations are needed to better characterize the mechanisms involved in the immunopathology of pneumonia.
In this cohort, highest serum IP-10 levels were identified among patients with both several pathogen detected and severe pneumonia, suggesting a significant role of IP-10 on pneumonia pathogenesis. Indeed, high plasma levels of IP-10 have previously been reported in patients with sepsis 12 , and were associated with high mortality rate, especially among patients with CAP 34 . Additionally, the IP-10-CXCR3 axis has been related to acute immune lung injury and lymphocyte apoptosis during the development of severe acute respiratory syndrome (SARS) 35, 36 . Moreover, an in vivo study that modeled influenza and pneumococcal superinfection in mice indicated that pro-inflammatory chemokines, including IP-10, play a crucial role in influenza-induced susceptibility to lung neutrophilia, severe immunopathology and mortality 37 . In this study, markedly elevated IP-10 (92,809 pg ml −1 ) combined with the highest PCT level (74.4 pg ml −1 ) were quantified in the serum sample of a child who died, in whom S. pneumoniae (serotype 9 V) was identified in the blood (PCR and blood culture) and co-detected with Haemophilus influenzae type B in nasal aspirate. These observations suggest an interrelationship between co-detection, elevated serum IP-10 and the pathogenesis of pneumonia.
Several limitations of this pilot translational study need to be acknowledged before concluding mixed infection is related to elevated IP-10 and disease severity. Indeed, although viral shedding (e.g., of HRV and HBoV) is common in asymptomatic children, we were unable to evaluate the levels of immunomodulators in the serum samples of a control group. Moreover, although the samples were collected within the first 24 hours after admission, only a single blood sample was processed for each patient. Therefore, a larger, longitudinal study on the etiology and severity of pneumonia will be necessary to confirm these results. In conclusion, the present findings suggest that mixed respiratory infections and IP-10 may play major, interconnected roles in the pathogenesis of pneumonia. Clinically, assessment and monitoring of induced IP-10 serum level may assist clinicians to improve diagnosis and patient management of severe community-acquired pneumonia.
Viral and bacterial strains. The 10 ng ml −1 M-CSF (Miltenyi Biotec). THP− 1 MDMs were obtained by culturing cells with 10 ng ml -1 phorbol myristate acetate (PMA; Invivogen, Toulouse, France) for 72 hours. Human airway epithelial cells (HAEC, bronchial cell type) originated from a 54-years old woman with no pathology reported (batch number MD056501) were provided by Mucilair (Epithelix, Geneva, Switzerland). Sterility, tissue integrity (TEER), mucus production and cilia beating frequency have been certified by the company. Gene expression profiling. Total cellular mRNA was purified using the RNeasy kit (Qiagen, Hilden, Germany). Reverse-transcription of total mRNA was performed using the RT 2 First Strand Kit (SABiosciences, Hilden, Germany). The expression of 84 genes involved in the human innate and adaptive immune responses was evaluated using the RT 2 profiler ™ PCR Array (SABiosciences) according to the manufacturer's recommendations. The Δ Δ Ct method was applied to calculate the fold changes in gene expression for each gene relative to uninfected control cells using the web-based RT 2 profiler PCR Array Data Analysis software (SABiosciences).
MicroRNA profiling array. Total cellular microRNAs were purified using the miRNeasy Mini kit (Qiagen) and reverse-transcribed using the miScript Reverse Transcription kit (Qiagen). The profiling of 84 miRNAs was performed using the Human Immunopathology miScript miRNA PCR Array kit (Qiagen) according to the manufacturer's instructions. Data were analyzed using the miScript miRNA PCR array data analysis web portal.
In silico miRNA target prediction. MiRNA target genes were retrieved and compiled using TargetScan 38 and microRNA.org resource 39 . The interactions between miRNAs and intracellular pathways were predicted using DIANA-miRPath v2.0 40 .
THP-1 MDMs were seeded in 24-well plates (0.5 × 10 6 per well) in triplicate, exposed to Influenza A H1N1 (A/Solomon islands/3/2006) virus (IAV) under serum-free conditions for 1 hour and then cultured for 3 hours in fresh RPMI-1640 containing 2% FBS. Streptococcus pneumoniae (SP) serotype 14 was added at 4 hours after IAV infection. Gentamicin (10 μ g ml −1 ) was added 2 hours after SP infection (i.e. 6 hours post-influenza infection) and maintained in the culture media throughout the experiment to kill extracellular bacteria and limit bacterial growth. Cell viability was determined by flow-cytometry using the FITC/Annexin V apoptosis detection kit (BD Biosciences), according to the manufacturer's instructions. #4427975) . In this assay, fold changes have been defined by the Δ Δ Ct method using control RNU-44 and -48 as reference microRNAs. Total mRNA was purified from transfected and infected MDMs using the RNeasy kit (Qiagen) and specific primers were used to amplify transforming growth factor beta-2 (TGFB2; F: 5′ -CCATCCCGCCCACTTTCTAC-3′ , R: 5′ -AGCTCAATCCGTTGTTCAGGC-3′ ), SOCS6 (F: 5′ -AAGAATTCATCCCTTGGATTAGGTAAC-3′ , R: 5′ -CAGACTGGAGGTCGTGGAA-3′ ) 41 43 , and 3) absence of wheezing at auscultation, and, 4) first symptoms appearing within the last 14 days, and 5) radiological confirmation of pneumonia as per WHO guidelines 44 . Based on these primary criteria defining pneumonia cases, all 74 cases were retrospectively re-evaluated according to the WHO "Pocket book of hospital care for children" 45 criteria to evaluate pneumonia severity. Cases that died during the study, or who had at least one additional clinical signs including central cyanosis, dullness to percussion during chest examination, prostration/lethargy, pleural effusion observed on chest radiography were retrospectively included in the severe pneumonia group. Patients without any of these additional clinical signs were included in the non-severe pneumonia group. Table 4 . a IP-10 values are expressed in pg ml -1 . IP-10 concentration differences between groups were compared using unpaired Mann-Whitney tests; significant changes (P < 0.05) are in bold. Clinical and molecular analysis. Nasopharyngeal aspirates (NAs) and whole blood samples were collected from children within 24 hours of admission. Whole blood samples were used for complete blood counts, blood culture and multiplex real-time PCR to identify Staphylococcus aureus, Streptococcus pneumoniae and Haemophilus influenzae type B 46 . S. pneumoniae serotypes were defined using a 11 multiplex real-time PCR assay targeting the 40 most frequently represented serotypes or serogroups according to protocol developed by Messaoudi et al. 47 . Serum C-reactive protein (CRP; AssayPro, St. Charles, Missouri, United States) and Procalcitonin (PCT; VIDAS B.R.A.H.M.S; bioMérieux) were quantified from whole-blood samples. Multiplex real-time non quantitative PCR (Fast-Track Diagnostic, Sliema, Malta) was used to detect 19 viruses and five bacteria in the respiratory specimens (NAs and pleural effusions). Mixed detection was defined as 1) PCR-positive for multiple viruses in NAs, 2) positive blood culture or PCR-positive for multiple bacteria in blood or 3) PCR-positive for one or multiple viruses in NAs and one or multiple bacteria in blood (identified by PCR and blood culture).
Ethical approval. The study protocol, informed consent statement, clinical research form, any amendments and all other study documents were submitted to and approved by the Ethical Committee of the Instituto de Investigaciones en Ciencias de la Salud, the Universidad Nacional de Asunción (IICS-UNA) and the Hospital Pediátrico Niños de Acosta Ñu. Informed consent was obtained from all subjects involved in this study. The clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki.
Statistical analysis. The Chi-square test and Fisher's exact test were used to compare categorical variables; continuous variables and non-normally distributed data were compared using the Mann-Whitney U-test; normally distributed data were compared using unpaired t-tests. Comparative analyses between experimental conditions (i.e., MOCK, IAV, SP or IAV + SP) were performed using one-way ANOVA with Tukey's post-hoc test or Kruskal-Wallis analysis with Dunn's post-hoc tests. Receiver operating curve (ROC) analysis was used to determine the optimal cut-off value for IP-10 to differentiate between non-severe and severe pneumonia cases. P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism (La Jolla, California, United States).
|
What suggests that IP-10 plays a significant role on the pathogenesis of pneumonia?
| 5,215 |
highest serum IP-10 levels
| 22,018 |
1,584 |
Viral and bacterial co-infection in severe pneumonia triggers innate immune responses and specifically enhances IP-10: a translational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5138590/
SHA: ef3d6cabc804e5eb587b34249b539c1b5efa4cc4
Authors: Hoffmann, Jonathan; Machado, Daniela; Terrier, Olivier; Pouzol, Stephane; Messaoudi, Mélina; Basualdo, Wilma; Espínola, Emilio E; Guillen, Rosa M.; Rosa-Calatrava, Manuel; Picot, Valentina; Bénet, Thomas; Endtz, Hubert; Russomando, Graciela; Paranhos-Baccalà, Gláucia
Date: 2016-12-06
DOI: 10.1038/srep38532
License: cc-by
Abstract: Mixed viral and bacterial infections are widely described in community-acquired pneumonia; however, the clinical implications of co-infection on the associated immunopathology remain poorly studied. In this study, microRNA, mRNA and cytokine/chemokine secretion profiling were investigated for human monocyte-derived macrophages infected in-vitro with Influenza virus A/H1N1 and/or Streptococcus pneumoniae. We observed that the in-vitro co-infection synergistically increased interferon-γ-induced protein-10 (CXCL10, IP-10) expression compared to the singly-infected cells conditions. We demonstrated that endogenous miRNA-200a-3p, whose expression was synergistically induced following co-infection, indirectly regulates CXCL10 expression by targeting suppressor of cytokine signaling-6 (SOCS-6), a well-known regulator of the JAK-STAT signaling pathway. Additionally, in a subsequent clinical pilot study, immunomodulators levels were evaluated in samples from 74 children (≤5 years-old) hospitalized with viral and/or bacterial community-acquired pneumonia. Clinically, among the 74 cases of pneumonia, patients with identified mixed-detection had significantly higher (3.6-fold) serum IP-10 levels than those with a single detection (P = 0.03), and were significantly associated with severe pneumonia (P < 0.01). This study demonstrates that viral and bacterial co-infection modulates the JAK-STAT signaling pathway and leads to exacerbated IP-10 expression, which could play a major role in the pathogenesis of pneumonia.
Text: Scientific RepoRts | 6:38532 | DOI: 10 .1038/srep38532 pathogenesis of several diseases and has been suggested as a potential biomarker of viral infection 10, 11 , late-onset bacterial infection in premature infants 12 , and a promising biomarker of sepsis and septic shock 13, 14 . Combined analysis of IP-10 and IFN-γ has also been reported as a useful biomarker for diagnosis and monitoring therapeutic efficacy in patients with active tuberculosis [15] [16] [17] , and both remain detectable in the urine of patients with pulmonary diseases in the absence of renal dysfunction 18 .
With airway epithelial cells 19 , resident alveolar macrophages (AMs) and blood monocytes-derived macrophages (recruited into tissues under inflammatory conditions 20, 21 ) represent a major line of defense against both pneumococcal (through their high phagocytic capacity [22] [23] [24] ) and influenza infection 25, 26 . So far, no studies have yet focused on the intracellular mechanisms that regulate IP-10 in human blood leukocytes during mixed IAV and SP infection. Several studies indicated that host non-coding small RNAs (including microRNAs) may function as immunomodulators by regulating several pivotal intracellular processes, such as the innate immune response 27 and antiviral activity 28, 29 ; both of these processes are closely related to toll-like receptor (TLR) signaling pathways.
In this study, we firstly investigated the in vitro intracellular mechanisms that mediate the innate immune response in IAV and/or SP infected human monocyte-derived macrophages (MDMs). Using this approach, we observed that mixed-infection of MDMs induces a synergistic production of IP-10 which can be related to a miRNA-200a/JAK-STAT/SOCS-6 regulatory pathway. Subsequently, in a retrospective analysis of clinical samples collected from children ≤ 5 years-old hospitalized with pneumonia, we confirmed that serum IP-10 level could be related to both viral and/or bacterial etiologies and disease severity.
Characteristics of MDMs infected by IAV and/or SP. Initially, we investigated in vitro the impact of single and mixed IAV and SP infection on MDMs. Firstly, active replication of IAV was assessed by qRT-PCR and quantification of new infectious viral particles in the cell supernatants ( Fig. 1a,b ). IAV titer increased over time after single infection with IAV and correlated with increased production of negative-strand IAV RNA. Maximum viral replication was observed at 18-24 hours post-infection, after which time both RNA replication and the quantity of infectious particles decreased. In this in vitro model, subsequent challenge of IAV-infected MDMs with SP had no significant impact on the production of new infectious viral particles (Fig. 1b) . Together, these results indicate permissive and productive infection of MDMs by IAV. Secondly, we evaluated whether MDMs are permissive for both IAV and SP infection. The presence of pneumococci within IAV-and SP-infected primary MDMs was confirmed at 8 h post-infection (Fig. 1c) , suggesting that MDMs are permissive for viral and bacterial co-infection in the early steps of infection. Importantly, confocal co-detection of mixed IAV and SP was only effective following 8 h post-infection due to the bactericidal impact of SP internalization within human macrophages (after 24 h, data not shown). Thirdly, we evaluated the impact of single and mixed infection with IAV and SP on MDM viability. Mixed infection significantly decreased cell viability (65.2 ± 4.5% total cell death at 48 hours post-infection; P < 0.0001) compared to single SP and IAV infection (39.6 ± 1.7% and 17.4 ± 1.1% total cell death, respectively; Fig. 1d ). Taken together, these results confirmed human MDMs are permissive to mixed viral and bacterial infection. mRNA, microRNA and protein expression profiling reveal an overall induction of the host innate immune response following IAV and/or SP infection of MDMs. To investigate the innate immune response orchestrated by IAV-and SP-infected human MDMs, we firstly evaluated the expression of 84 genes involved in the innate and adaptive immune responses (Table S1) ; the major differentially-expressed genes are summarized in Fig. 2a . Expression profiling indicated an overall induction of genes related to the JAK-STAT, NF-Κ β and TLR signaling pathways. Indeed, all interferon-stimulated genes (ISGs) screened, including CXCL10 (fold-change [FC] = 240.9), CCL-2 (FC = 34.2) and MX-1 (FC = 151.4) were upregulated following mixed infection compared to uninfected cells, most of which are closely related to STAT-1 (FC = 52.3), IRF-7 (FC = 6.8) and IFNB1 (FC = 5.2) also found upregulated in mixed infected cells. Secondly, we investigated the endogenous microRNA expression profiles of IAV-and SP-infected MDMs. A selection of microRNAs that were found to be differentially-expressed under different infection conditions are shown in Fig. 2b and Table S2 . MiRNA-200a-3p was overexpressed after both single IAV (FC = 6.9), single SP (FC = 3.7) and mixed IAV/SP infection (FC = 7.3), indicating this miRNA may play a role in the innate immune response to viral and bacterial co-infection. Similar miRNA-200a-3p dysregulation profiles were obtained following IAV and/or SP infections of human macrophages-like (THP-1 monocytes-derived macrophages) or primary MDMs (data not shown). Thirdly, the secreted levels of various antiviral, pro-inflammatory and immunomodulatory cytokines/chemokines were assayed in IAV-and SP-infected-THP-1 and primary MDM cell supernatants. We observed a remarkable correlation between the mRNA and protein expression profiles of single or mixed infected MDMs especially regarding CXCL-10 and IP-10 expression. Indeed, the level of IP-10 was synergistically increased in the supernatant of IAV-infected THP-1 MDMs exposed to SP (mean: 30,589 ± 16,484 pg ml −1 ) compared to single IAV infection (1,439 ± 566.5 pg ml −1 ) and single SP infection (4,472 ± 2,001 pg ml −1 ; P≤ 0.05; Fig. 2c ) at 24 hours after infection. In those cells, IP-10 expression reduced over time (48 to 72 hours), coinciding with a significant higher proportion of necrotic and apoptotic cells (Fig. 1d) . The synergistic expression of IP-10 was similarly observed at 24 hours post-infection using primary MDMs (Fig. 2d) . Significantly increased secretion of the other tested cytokines and chemokines was not observed post-infection, even in mixed infected MDMs (Fig. S1 ). Interestingly, a significant production of IP-10 was also observed in supernatants of primary human airway epithelial cells (HAEC) mixed-infected by IAV and SP compared to the single infections (Fig. 2e) . Taken together, the mRNA and protein profiling results suggested that mixed viral and bacterial infection of MDMs induces a synergistic pro-inflammatory response related to the type-1 interferon and JAK-STAT signaling pathways, with IP-10 as signature of IAV/SP co-infection. Among all microRNAs screened, miR-200a-3p was the most Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 overexpressed in IAV/SP co-infection of human MDMs. In the remainder of this study, we decided to investigate the interconnection between miR-200a-3p expression and the innate immune response.
Endogenous miRNA-200a-3p expression correlates with CXCL10 (IP-10) induction following mixed IAV and SP infection of human MDMs. Using a specific Taqman probe assay targeting miR-200a-3p, we confirmed a significant upregulation of miR-200a-3p following mixed IAV and SP infection of human MDMs (Fig. 3a) . In this experiment, a more marked up-regulation of miR-200a-3p was observed following IAV+ SP compared to results obtained previously (Fig. 2b) . This discrepancy has been attributed to the use of two different approaches to quantify miR-200a-3p expression. The use of a target-specific stem-loop reverse transcription primer in Fig. 3a allows a better sensitivity of miR-200a-3p detection compared to the non-specific fluorescent dye used in Fig. 2b . As the general trend was suggestive of a synergistic induction of miR-200a-3p in response to mixed infection (Fig. 3a) , we hypothesized microRNA-200a-3p may play a role in the regulation of CXCL10 (IP-10), which was also synergistically upregulated in mixed-infected MDMs ( Fig. 2c and d) and primary HAEC ( Statistical analyses were performed using two-way ANOVA with Tukey's post-hoc test; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.
Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 CXCL10 (Fig. 3d) . These results suggested miR-200a-3p indirectly regulates CXCL10 and led us to hypothesize that miR-200a-3p controls a potential repressor of the JAK-STAT signaling pathway. . At 18 h after transfection, the MDMs were singly or mixed infected as described previously. At 8 h post-IAV and/or SP infection, total mRNA was extracted and amplified by PCR using specific primers for the indicated genes. Values represent median ± IQR (a, c) or mean ± SEM (d, e) of three biological replicates. Statistical analyses were performed using a Kruskal-Wallis test (non-parametric, one-way ANOVA with Dunn's post-hoc test) for data presented in (a, c). An ordinary two-way ANOVA (with Tukey's post-hoc multiple comparison test) was used for data presented in (d, e). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. MiRNA-200a-3p indirectly regulates IP-10 expression by targeting SOCS6. As shown in Fig. 2a , several JAK-STAT signaling pathway genes were deregulated in mixed IAV-and SP-infected human MDMs; therefore, we hypothesized that miR-200a-3p directly regulates a regulator of the JAK-STAT signaling pathway. Predictive target analysis indicated that the 3' UTR of suppressor of cytokine signaling-6 (SOCS6) may be targeted by miR-200a-3p (Fig. 3b) . SOCS proteins constitute a class of negative regulators of JAK-STAT signaling pathways that are induced by both cytokines and TLR signaling. MiRNA-200a-3p was not predicted to target any of the other six members of the SOCS gene family. Transfection of human MDMs with MIM-200a downregulated SOCS6 (FC = 0.57) while inhibition of miR-200a-3p (INH-200a) upregulated SOCS6 (FC = 1.55), confirming that miR-200a-3p effectively regulates the expression of SOCS6 (Fig. 3e) . Moreover, SOCS6 was synergistically downregulated in IAV-or IAV/SP-infected MDMs overexpressing miRNA-200a (Fig. 3e) , suggesting that both infection and miR-200a-3p negatively regulate the expression of SOCS6. Finally, western blotting confirmed that expression of SOCS-6 sharply reduced following infection, especially after mixed IAV and SP infection (Fig. 3f) .
These results indicate miR-200a-3p is strongly induced in response to mixed viral and bacterial co-infection, which in turn leads to downregulation of the JAK-STAT regulator SOCS-6 at both the mRNA and protein levels and subsequent upregulation of IP-10.
analyses demonstrated mixed IAV and SP infection of human MDMs and HAEC induced significant production of IP-10. As blood leukocytes and respiratory tract epithelial cells actively contribute to inflammation during pneumonia, we hypothesized the level of IP-10 in serum of patient with pneumonia may be both indicative of mixed respiratory infection and disease severity. As part of a prospective, hospital-based, multicenter case-control study on the etiology of pneumonia among children under 5-years-old, a total of 74 patients (44 male, 30 female) were included in this pilot evaluation. According to WHO guidelines, retrospective analysis indicated 44 (59.5%) children had clinical signs of non-severe pneumonia and 30 (40.5%) children had signs of severe pneumonia. The main patient characteristics at inclusion are shown in Table 1 . Patients with severe pneumonia had significant more recorded episodes of dyspnea (P < 0.001), cyanosis (P = 0.03), lower chest indrawing (P < 0.001), dullness to percussion (P < 0.001) and lethargy (P < 0.001) during chest examination than patient with non-severe pneumonia. Moreover, pleural effusions were significantly more observed among critically ill patients and the duration of hospitalization was significantly longer for the children with severe pneumonia than for those with non-severe pneumonia (P = 0.0015). Two deaths occurred within the group of children retrospectively defined with severe pneumonia. Evaluation of the systemic inflammatory response of the 74 cases is shown in Table 2 . Serum level of CRP, IP-10, PCT, G-CSF, IL-6, IL-8 and MIP-1β were significantly more elevated in serum samples from critically ill patients. Patients with severe pneumonia had significantly higher (4.2-fold) serum IP-10 levels than those with a non-severe pneumonia (P < 0.001) suggesting IP-10 as a promising prognostic marker in pneumonia. Diagnostic accuracy measures for predicting pneumonia severity using blood-based biomarkers are summarized in Table S3 . Briefly, in this study, the optimal IP-10 cut-off value for identifying patient with severe pneumonia was 4,240 pg ml −1 , with an area under the receiver operating characteristic curve of 0.69 (95% CI, 0.57 to 0.82, P < 0.001). Defining as positive a serum IP-10 level above this cut-off resulted in a sensitivity of 63.3%, specificity of 63.6% and a positive likelihood ratio of 1.74. Prognostic values of IP-10 were closed to procalcitonin (PCT; AUC = 0.70; 95% IC, 0.58 to 0.82, P < 0.001) and IL-6 (AUC = 0.70; 95% IC, 0.58-0.83, P < 0.001).
Multiplex PCR-based screening of respiratory and blood samples reveal a high variety of pathogen associations (Table 3) . Respiratory viruses were detected in the nasal aspirates (NAs) of 63/74 patients (85.1%). Etiological bacteria of pneumonia (S. pneumoniae, n = 19; S. aureus, n = 1; or H. influenzae type B, n = 7) were identified via real-time PCR in the blood samples of 27/74 (36.5%) of the patients. Multiplex PCR assays allowed the identification of respiratory bacteria in the blood of 19 patients with negative blood culture results. Among the 74 cases PCR-positive for respiratory pathogens, a single virus or bacteria were detected in the NAs of 7 (9.4%) and 3 (4.0%) patients, respectively; these 10/74 (13.5%) cases were defined as the single infection group. The mixed infection group included the 62/74 (83.8%) cases in which (1) multiple viruses and/or bacteria were identified in NAs (38/74; 51.3%) without any bacteria identified in blood samples or (2) one or more viruses and/or bacteria were identified in NAs and associated with a blood bacteremia (24/74; 32.4%). We evaluated whether IP-10 serum level could correlate with the viral and bacterial etiologies of pneumonia. Patients with mixed infection had significant higher (3.6-fold) IP-10 serum level than patient with single detection (P = 0.03; Table 4 ). A stratified analysis reveals that the highest IP-10 serum level was observed among patients with both several respiratory pathogens identified (mixed-detection group) and severe pneumonia (14,427 pg ml −1 , IQR (3,981-82,994). In detail, a remarkable IP-10 serum level (142,531 pg ml −1 ), representing 33-fold higher above cut-off value predicting pneumonia severity was observed in patient with hRV in NA co-detected with S. pneumoniae (serotype 14) in pleural effusion and blood. In concordance with our in-vitro model of co-infection, a significant IP-10 level (90,338 pg ml −1 ) was quantified in blood sample of patient with severe bacteremic pneumococcal (serotype 14) pneumonia with a positive co-detection of Influenza B virus in NA. Taken together, these results suggest that high serum IP-10 levels are significantly associated with mixed viral and bacterial detection and also related to pneumonia pathogenesis.
This study provides additional in vitro and clinical data to improve our understanding of the immunopathology of mixed viral and bacterial pneumonia (Fig. 4) .
The in vitro model of influenza and pneumococcal superinfection of human MDMs demonstrated that mixed infection synergistically induced release of the pro-inflammatory chemokine IP-10, strongly suggesting human Scientific RepoRts | 6:38532 | DOI: 10.1038/srep38532 blood leukocytes contribute to the immunopathology of pneumonia. Additionally, transcriptomics and omics analyses provided new data on the inflammatory pathways that are activated during mixed infection and related to synergistic induction of the pro-inflammatory chemokine IP-10 in mixed infected cells. Our observations are consistent with a recent study describing IP-10 induction as host-proteome signature of both viral and bacterial infections 30 . Of the differentially-expressed genes observed in mixed infected MDMs, the transcription factors STAT-1 and IRF-7 appear to play crucial roles in the regulation of interferon-stimulated genes including CXCL10 (IP-10). By focusing on the intracellular mechanisms that regulate inflammatory pathways, we demonstrated a novel role for miRNA-200a-3p in the regulation of CXCL10 (IP-10). These observations are consistent with previous reports showing that RNA virus infection upregulates miR-155 in macrophages and dendritic cells and also regulates suppressor of cytokine signaling 1 (SOCS1), suggesting the existence of a miRNA/JAK-STAT/SOCS regulatory pathway during viral infection 29 . Our study suggests co-infection leads to overexpression of miR-200a-3p, which in turn targets and downregulates the JAK-STAT regulator SOCS-6 and consequently increases CXCL10 (IP-10) expression. Interestingly, a complementary in-silico approach reveals that several microRNAs that were found dysregulated in our experiments of IAV and SP co-infection of MDMs or HAEC, might target several genes of SOCS family and play similar role than miR-200a-3p. Indeed, miRNA-142-3p might target SOCS4, 5, 6 mRNA while miRNA-194-5p might target SOCS2, 3, 4, 5 and 7 mRNA. These observations underline that intra-cellular regulation of IP-10 is not limited to the contribution of a sole microRNA. A complex inter-relationship between numerous host microRNAs and inhibitors of the JAK-STAT signaling pathway occur to control host innate inflammatory response against viral and/or bacterial infections. Clinically, the majority of pediatric CAP cases in this study were associated with both positive viral and/or bacterial detection. Respiratory microorganisms were detected in 97% of cases; 51.3% of which were viral-viral, viral-bacterial or bacterial-bacterial co-detected only in nasal aspirates, 32.4% of which co-detected in both nasal aspirates and blood samples. These data are consistent with previous etiological studies of pediatric CAP 3,31-33 . S. pneumoniae was the major bacteria identified in blood (19/74; 25.7%) and mainly co-detected with respiratory viruses in NAs (16/19; 84.2%). We observed a very high diversity of viral and bacterial associations in biological samples from children with pneumonia. In comparison with IAV and SP14 combination evaluated in-vitro, no pneumonia cases were singly influenza and pneumococcus infected, and no similar co-detection with those two pathogens has been clinically observed. Nevertheless, Influenza B (IVB) virus was identified in 5 patients and two of them had a positive SP co-detection in blood (one non-typable strain and one serotype 14 using our molecular typing test). IVB and SP14 combination seems to be the nearest pathogen co-detection to that in-vitro investigated. Clinically, this co-detection was associated with both a very high IP-10 expression and a very severe pneumonia case definition. Interestingly, our translational pilot evaluation reveals IP-10 expression can be induced by several different viral and/or bacterial combinations. As immune response to each pathogen is different, further in-vitro investigations using different pathogens associations are needed to better characterize the mechanisms involved in the immunopathology of pneumonia.
In this cohort, highest serum IP-10 levels were identified among patients with both several pathogen detected and severe pneumonia, suggesting a significant role of IP-10 on pneumonia pathogenesis. Indeed, high plasma levels of IP-10 have previously been reported in patients with sepsis 12 , and were associated with high mortality rate, especially among patients with CAP 34 . Additionally, the IP-10-CXCR3 axis has been related to acute immune lung injury and lymphocyte apoptosis during the development of severe acute respiratory syndrome (SARS) 35, 36 . Moreover, an in vivo study that modeled influenza and pneumococcal superinfection in mice indicated that pro-inflammatory chemokines, including IP-10, play a crucial role in influenza-induced susceptibility to lung neutrophilia, severe immunopathology and mortality 37 . In this study, markedly elevated IP-10 (92,809 pg ml −1 ) combined with the highest PCT level (74.4 pg ml −1 ) were quantified in the serum sample of a child who died, in whom S. pneumoniae (serotype 9 V) was identified in the blood (PCR and blood culture) and co-detected with Haemophilus influenzae type B in nasal aspirate. These observations suggest an interrelationship between co-detection, elevated serum IP-10 and the pathogenesis of pneumonia.
Several limitations of this pilot translational study need to be acknowledged before concluding mixed infection is related to elevated IP-10 and disease severity. Indeed, although viral shedding (e.g., of HRV and HBoV) is common in asymptomatic children, we were unable to evaluate the levels of immunomodulators in the serum samples of a control group. Moreover, although the samples were collected within the first 24 hours after admission, only a single blood sample was processed for each patient. Therefore, a larger, longitudinal study on the etiology and severity of pneumonia will be necessary to confirm these results. In conclusion, the present findings suggest that mixed respiratory infections and IP-10 may play major, interconnected roles in the pathogenesis of pneumonia. Clinically, assessment and monitoring of induced IP-10 serum level may assist clinicians to improve diagnosis and patient management of severe community-acquired pneumonia.
Viral and bacterial strains. The 10 ng ml −1 M-CSF (Miltenyi Biotec). THP− 1 MDMs were obtained by culturing cells with 10 ng ml -1 phorbol myristate acetate (PMA; Invivogen, Toulouse, France) for 72 hours. Human airway epithelial cells (HAEC, bronchial cell type) originated from a 54-years old woman with no pathology reported (batch number MD056501) were provided by Mucilair (Epithelix, Geneva, Switzerland). Sterility, tissue integrity (TEER), mucus production and cilia beating frequency have been certified by the company. Gene expression profiling. Total cellular mRNA was purified using the RNeasy kit (Qiagen, Hilden, Germany). Reverse-transcription of total mRNA was performed using the RT 2 First Strand Kit (SABiosciences, Hilden, Germany). The expression of 84 genes involved in the human innate and adaptive immune responses was evaluated using the RT 2 profiler ™ PCR Array (SABiosciences) according to the manufacturer's recommendations. The Δ Δ Ct method was applied to calculate the fold changes in gene expression for each gene relative to uninfected control cells using the web-based RT 2 profiler PCR Array Data Analysis software (SABiosciences).
MicroRNA profiling array. Total cellular microRNAs were purified using the miRNeasy Mini kit (Qiagen) and reverse-transcribed using the miScript Reverse Transcription kit (Qiagen). The profiling of 84 miRNAs was performed using the Human Immunopathology miScript miRNA PCR Array kit (Qiagen) according to the manufacturer's instructions. Data were analyzed using the miScript miRNA PCR array data analysis web portal.
In silico miRNA target prediction. MiRNA target genes were retrieved and compiled using TargetScan 38 and microRNA.org resource 39 . The interactions between miRNAs and intracellular pathways were predicted using DIANA-miRPath v2.0 40 .
THP-1 MDMs were seeded in 24-well plates (0.5 × 10 6 per well) in triplicate, exposed to Influenza A H1N1 (A/Solomon islands/3/2006) virus (IAV) under serum-free conditions for 1 hour and then cultured for 3 hours in fresh RPMI-1640 containing 2% FBS. Streptococcus pneumoniae (SP) serotype 14 was added at 4 hours after IAV infection. Gentamicin (10 μ g ml −1 ) was added 2 hours after SP infection (i.e. 6 hours post-influenza infection) and maintained in the culture media throughout the experiment to kill extracellular bacteria and limit bacterial growth. Cell viability was determined by flow-cytometry using the FITC/Annexin V apoptosis detection kit (BD Biosciences), according to the manufacturer's instructions. #4427975) . In this assay, fold changes have been defined by the Δ Δ Ct method using control RNU-44 and -48 as reference microRNAs. Total mRNA was purified from transfected and infected MDMs using the RNeasy kit (Qiagen) and specific primers were used to amplify transforming growth factor beta-2 (TGFB2; F: 5′ -CCATCCCGCCCACTTTCTAC-3′ , R: 5′ -AGCTCAATCCGTTGTTCAGGC-3′ ), SOCS6 (F: 5′ -AAGAATTCATCCCTTGGATTAGGTAAC-3′ , R: 5′ -CAGACTGGAGGTCGTGGAA-3′ ) 41 43 , and 3) absence of wheezing at auscultation, and, 4) first symptoms appearing within the last 14 days, and 5) radiological confirmation of pneumonia as per WHO guidelines 44 . Based on these primary criteria defining pneumonia cases, all 74 cases were retrospectively re-evaluated according to the WHO "Pocket book of hospital care for children" 45 criteria to evaluate pneumonia severity. Cases that died during the study, or who had at least one additional clinical signs including central cyanosis, dullness to percussion during chest examination, prostration/lethargy, pleural effusion observed on chest radiography were retrospectively included in the severe pneumonia group. Patients without any of these additional clinical signs were included in the non-severe pneumonia group. Table 4 . a IP-10 values are expressed in pg ml -1 . IP-10 concentration differences between groups were compared using unpaired Mann-Whitney tests; significant changes (P < 0.05) are in bold. Clinical and molecular analysis. Nasopharyngeal aspirates (NAs) and whole blood samples were collected from children within 24 hours of admission. Whole blood samples were used for complete blood counts, blood culture and multiplex real-time PCR to identify Staphylococcus aureus, Streptococcus pneumoniae and Haemophilus influenzae type B 46 . S. pneumoniae serotypes were defined using a 11 multiplex real-time PCR assay targeting the 40 most frequently represented serotypes or serogroups according to protocol developed by Messaoudi et al. 47 . Serum C-reactive protein (CRP; AssayPro, St. Charles, Missouri, United States) and Procalcitonin (PCT; VIDAS B.R.A.H.M.S; bioMérieux) were quantified from whole-blood samples. Multiplex real-time non quantitative PCR (Fast-Track Diagnostic, Sliema, Malta) was used to detect 19 viruses and five bacteria in the respiratory specimens (NAs and pleural effusions). Mixed detection was defined as 1) PCR-positive for multiple viruses in NAs, 2) positive blood culture or PCR-positive for multiple bacteria in blood or 3) PCR-positive for one or multiple viruses in NAs and one or multiple bacteria in blood (identified by PCR and blood culture).
Ethical approval. The study protocol, informed consent statement, clinical research form, any amendments and all other study documents were submitted to and approved by the Ethical Committee of the Instituto de Investigaciones en Ciencias de la Salud, the Universidad Nacional de Asunción (IICS-UNA) and the Hospital Pediátrico Niños de Acosta Ñu. Informed consent was obtained from all subjects involved in this study. The clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki.
Statistical analysis. The Chi-square test and Fisher's exact test were used to compare categorical variables; continuous variables and non-normally distributed data were compared using the Mann-Whitney U-test; normally distributed data were compared using unpaired t-tests. Comparative analyses between experimental conditions (i.e., MOCK, IAV, SP or IAV + SP) were performed using one-way ANOVA with Tukey's post-hoc test or Kruskal-Wallis analysis with Dunn's post-hoc tests. Receiver operating curve (ROC) analysis was used to determine the optimal cut-off value for IP-10 to differentiate between non-severe and severe pneumonia cases. P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism (La Jolla, California, United States).
|
What cell types help prevent pneumococcal and influenza infection in the lungs?
| 5,216 |
airway epithelial cells 19 , resident alveolar macrophages (AMs) and blood monocytes-derived macrophages
| 2,714 |
1,586 |
In Vitro Bactericidal Activity of 4- and 5-Chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides against MRSA
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4321674/
SHA: f0e6cef57dbae030aea2f324e21e00945ac659cf
Authors: Zadrazilova, Iveta; Pospisilova, Sarka; Pauk, Karel; Imramovsky, Ales; Vinsova, Jarmila; Cizek, Alois; Jampilek, Josef
Date: 2015-01-15
DOI: 10.1155/2015/349534
License: cc-by
Abstract: A series of nine substituted 2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides was assessed as prospective bactericidal agents against three clinical isolates of methicillin-resistant Staphylococcus aureus (MRSA) and S. aureus ATCC 29213 as the reference and quality control strain. The minimum bactericidal concentration was determined by subculturing aliquots from MIC determination onto substance-free agar plates. The bactericidal kinetics of compounds 5-chloro-2-hydroxy-N-[(2S)-3-methyl-1-oxo-1-{[4-(trifluoromethyl)phenyl]amino}butan-2-yl]benzamide (1f), N-{(2S)-1-[(4-bromophenyl)amino]-3-methyl-1-oxobutan-2-yl}-4-chloro-2-hydroxybenzamide (1g), and 4-chloro-N-{(2S)-1-[(3,4-dichlorophenyl)amino]-3-methyl-1-oxobutan-2-yl}-2-hydroxybenzamide (1h) was established by time-kill assay with a final concentration of the compound equal to 1x, 2x, and 4x MIC; aliquots were removed at 0, 4, 6, 8, and 24 h time points. The most potent bactericidal agent was compound 1f exhibiting remarkable rapid concentration-dependent bactericidal effect even at 2x MIC at 4, 6, and 8 h (with a reduction in bacterial count ranging from 3.08 to 3.75 log(10) CFU/mL) and at 4x MIC at 4, 6, 8, and 24 h (5.30 log(10) CFU/mL reduction in bacterial count) after incubation against MRSA 63718. Reliable bactericidal effect against other strains was maintained at 4x MIC at 24 h.
Text: The antibiotic resistance of invasive pathogens has become one of the most challenging and persistent health problems [1] . Methicillin-resistant Staphylococcus aureus (MRSA) has become the most common clinically relevant multiresistant pathogen [2] causing both healthcare-associated and community-acquired bloodstream infections with mortality rates up to 40% [3] .
The prevalence of MRSA is increasing worldwide and, according to the latest information of the European Centre for Disease Prevention and Control from 2012 [4] , can be considered alarming in some European countries, especially in Portugal and Romania, where ≥50% of all S. aureus isolates from invasive infections were identified as MRSA in 2012 (although, e.g., in Romania the prevalence of MRSA was 25-50% in 2010), followed by Italy, Greece, and Poland with 25-50% isolates being MRSA in 2012 (for comparison, in Poland MRSA isolates constituted 10-25% from all S. aureus isolates in 2010).
The treatment failure of vancomycin, the therapeutic anti-MRSA agent of choice, due to the strains with elevated vancomycin minimum inhibitory concentration (MIC) values (i.e., the lowest concentration of an antimicrobial that will inhibit the visible growth of a microorganism) within the susceptible range was described previously [5, 6] . Thus, the emergence of MRSA (and vancomycin-resistant S. aureus in the recent years as well [7] ) makes the discovery of new molecular scaffolds a priority, and the current situation even necessitates the reengineering and repositioning of some old drug families to achieve adequate control of these bacteria [8] . However, for the treatment of S. aureus bloodstream infections, bactericidal antimicrobial agents are considered to be superior to bacteriostatic drugs [9] . This fact should be considered during the development of effective and safe treatment options for MRSA infections.
The history of clinical usage of salicylanilides (2-hydroxy-N-phenylbenzamides) dates back to the 1940s in therapy of tinea capitis, followed by the discovery of their anthelmintic properties in the mid 1950s [10] . Nowadays, salicylanilides (SALs) are a class of aromatic compounds possessing a wide range of interesting pharmacological activities, such as anthelmintic [11] , antibacterial [12, 13] , antimycobacterial [13] , antifungal [14] , and antiviral [15, 16] , among others. Despite being studied since the 1960s, the mechanism of action responsible for biological activities of these compounds has not been explained so far. SALs have been found to inhibit the two-component regulatory systems (TCS) of bacteria [17] . The latest studies specified them also as selective inhibitors of interleukin-12p40 production that plays a specific role in initiation, expansion, and control of cellular response to tuberculosis [18] . Furthermore, salicylanilides have been recognised as inhibitors of some bacterial enzymes, such as sortase A from S. aureus [19] , d-alanine-d-alanine ligase [20] , or transglycosylases from S. aureus (but not from M. tuberculosis) [12] . These enzymes participate in secretion of various proteins or in biosynthesis of bacterial cell wall. Recently, salicylanilides-like derivatives were described to inhibit two enzymes essential for mycobacteria: (i) methionine aminopeptidase, catalyzing a key step of the posttranslational modification of nascent proteins, and (ii) isocitrate lyase, which is essential for the metabolism of fatty acids [21] . Thus, SALs seem to be promising candidates for development of new antibacterial agents with a novel mechanism of action. Such new agents could be a solution to the resistance challenges.
This study is a follow-up paper to a recently published article [13] . The synthesis of the series of novel derivatives of salicylamides, 4-and 5-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides, called diamides due to their skeleton (for general structure see Table 1 ), was described previously [13, 22] , and their antimycobacterial and antibacterial activities against various bacterial species were reported [13] . As these compounds expressed very significant antibacterial activity with low MIC values against clinical isolates of MRSA as representatives of multidrugresistant bacteria, we decided to extend the knowledge about the antibacterial properties of these compounds against MRSA.
The aim of the current study was to assess the overall in vitro bactericidal activity of nine newly synthesized diamides in dependence on time and concentration against clinical isolates of MRSA as representatives of multidrug-resistant bacteria. To the best of our knowledge, this is the first study dealing with the evaluation of novel microbiological characteristics of SAL analogues and revealing their bactericidal effect.
The synthetic pathway of the series of novel diamides was described recently [13, 22] , and their structures (see Table 1 ) were confirmed by IR, NMR, and MS spectrometry, and the purity of the compounds was checked by CHN analysis [13, 22] . [27] ; and MRSA SA 3202 [27] (National Institute of Public Health, Prague, Czech Republic) both of human origin. Suspected colonies were confirmed by PCR; a 108 bp fragment specific for S. aureus was detected [28] . All isolates were tested for the presence of the mecA gene encoding methicillin resistance [29] . These three clinical isolates were classified as vancomycin-susceptible (but with higher MIC of vancomycin equal to 2 g/mL (VA2-MRSA) within the susceptible range for MRSA 63718) methicillinresistant S. aureus (VS-MRSA). For the MICs of vancomycin, see Table 1 . Vancomycin-susceptible methicillin-susceptible Staphylococcus aureus (VS-MSSA) ATCC 29213, obtained from the American Type Culture Collection, was used as the reference and quality control strain. The bacteria were stored at −80 ∘ C and were kept on blood agar plates (Columbia agar base with 5% ovine blood) between experiments. (MBCs) . The MBCs (i.e., the lowest concentrations of antibacterial agents required to kill a particular bacterium) were determined by subculturing aliquots (20 L) from wells with no visible bacterial growth and from control wells of MIC determination onto substance-free Mueller-Hinton agar (MHA) plates. The plates were incubated aerobically at 37 ∘ C for 24 h for colony count. The MBC was defined as the lowest concentration of substance, which produced ≥99.9% killing Table 1 : Chemical structures and in vitro MIC and MBC [ g/mL] values of tested 5-and 4-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides (bactericidal effect of individual compounds against particular strains marked in bold). after 24 h of incubation as compared to the colony count of the starting inoculum [30] . To ensure reproducibility, each MBC assay was performed in at least triplicate on separate occasions.
N H O H N O OH 1 2 R 1 R 3 R 2 Comp. R 1 R 2 R 3 MIC [ g/mL] MBC [ g/mL] 1 2 3 4 1 2 3 4 1a 5-Cl 4-CH 3 (S)-CH 3 >256 >256 >256 >256 >256 >256 >256 >256 1b 5-Cl 4-CH 3 (S)-CH(CH 3 ) 2 >256 >256 32 32 >256 >256 128 >256 1c 5-Cl 4-CH 3 (S)-benzyl >256 >256 >256 >256 >256 >256 >256 >256 1d 5-Cl 4-CH 3 (R)-CH 2 -indolyl >256 >256 >256 >256 >256 >256 >256 >256 1e 5-Cl 4-OCH 3 (S)-CH(CH 3 ) 2 >256 >256 >256 >256 >256 >256 >256 >256 1f 5-Cl 4-CF 3 (S)-CH(CH 3 ) 2 4 2 2 2 4 4 8 4 1g 4-Cl 4-Br (S)-CH(CH 3 ) 2 8 4 4 4 1 6 8 8 8 1h 4-Cl 3,4-Cl (S)-CH(CH 3 ) 2 2 1 1 1 4 1 4 2 1i 4-Cl 3,4-Cl (S)-benzyl 1 1 0.5 0.5 8 1 8 1 AMP - - - >16 >16 >16 0.25 >16 >16 >16 0.25 CPX - - - >16 >16 >16 0.5 >16 >16 >16 0.5 VAN - - - 2 1 1 1 2 1 1 1
Time-kill assays were performed by the broth macrodilution method according to previously described methodology [30] with some modifications. Briefly, flasks containing sterile fresh Mueller-Hinton broth (MHB) with the appropriate antimicrobial agent were inoculated with the test organism in logarithmic growth phase to obtain the starting inoculum with the concentration of approximately 7.5 × 10 6 CFU/mL (actual inoculum concentrations ranged from 0.9 × 10 5 to 2.9 × 10 6 CFU/mL) and a final concentration of the antibiotic equal to 1x, 2x, and 4x MIC in 10 mL volume. For the determination of viable counts, aliquots were removed at 0, 4, 6, 8, and 24 h time points after inoculation, serially diluted in sterile phosphate buffered saline, and aliquots (20 L) were plated on MHA plates in duplicate. Colony counts were performed on plates yielding 6 to 60 colonies, and the mean was calculated. Antimicrobial carry-over was controlled by dilution and visual inspection of the distribution of colonies on the plates with observation of possible inhibition of growth at the site of the initial streaks. The plates were incubated at 37 ∘ C for 24 to 48 h, and the number of colonies was determined. To ensure reproducibility, each time-kill experiment was carried out in duplicate on separate occasions with results presented as the mean of all experiments. The growth control without the addition of antimicrobial agents and the control containing DMSO without any antimicrobial agent to exclude antibacterial activity of this solvent were included. Time-kill curves were constructed by plotting the log 10 CFU per millilitre versus time (over 24 h), and the change in bacterial concentration was determined. The results were analysed by evaluating the numbers of strains that yielded Δ(log 10 CFU/mL) values of −1 (corresponding to 90% killing), −2 (99% killing), and −3 (99.9% killing) at 4, 6, 8, and 24 h compared to counts at 0 h. Bactericidal activity was defined as a reduction of at least 99.9% (≥3 log 10 ) of the total count of CFU/mL in the original inoculum.
Diamides seem to be promising candidates for antibacterial agents with very strong anti-MRSA activity, as it was published recently [13] . In the present study the series of nine newly synthesized diamides was evaluated as prospective bactericidal agents against representatives of multidrugresistant bacteria, three clinical isolates of MRSA, and Staphylococcus aureus ATCC 29213 (methicillin-susceptible) as the reference and quality control strain. Since SALs and their analogues are known as compounds with bacteriostatic effect [31] , this is the first study where SAL-like compounds were considered as prospective bactericidal agents and the dependence of bactericidal effect of these compounds on time and concentration was evaluated. Thus, absolutely novel microbiological characteristics of these compounds were revealed in the present study.
Recently MIC values of diamides expressed as molar concentrations in mol/L were published [13] . To allow comparison with MBC values of the present study, MICs in g/mL were calculated and are recorded in Table 1 along with the activity of reference antibacterial drugs, ampicillin, ciprofloxacin, and vancomycin. Potential bactericidal activity of diamides was assessed using MBC assay [26] . MBC values of all tested compounds are recorded in Table 1 as well.
Based on the obtained results, all compounds assessed as active according to MIC values in our previous study (1f-i) showed low or moderate MBC values against all four strains. The MBC values of these compounds did not exceed the highest tested drug concentration and ranged from 1 to 16 g/mL. In all cases, there were comparable MBC values for the clinical isolates of MRSA and the S. aureus reference strain.
Bactericidal activity is defined as a ratio of MBC to MIC of ≤4 [32] . Table 1 bactericidal activity is expressed in bold.
As mentioned above, SALs are known to exhibit a bacteriostatic effect [31] , so it was very interesting to discover that diamides possess bactericidal activity. The amide bond (-CONH-) can cause interactions with a variety of enzymes [33] ; therefore the presence of two amide bonds could be responsible for the bactericidal effect of diamides against MRSA. The activity of SALs and their analogues results from multiple mechanisms, which are still under investigation; for example, it was found that SALs are capable of inhibiting transglycosylases in later stages of S. aureus (including MRSA) cell wall biosynthesis [12] . These enzymes catalyse the step prior to the transpeptidation in the peptidoglycan biosynthesis and are responsible for polymerization of lipid II, which occurs at the outer face of the membrane [12] . Since antibacterial agents targeting cell wall biosynthesis act as bactericidal agents [30, 34] , the failure in the cell wall biosynthesis due to the inhibition of transglycosylases could be responsible for bactericidal activity of diamides against MRSA.
Based on these findings, antibacterial active diamides with bactericidal effect against all four tested strains as prospective bactericidal agents were chosen for subsequent timekill curve studies to determine the real dependence of bactericidal effect on concentration over time.
1-oxobutan-2-yl}-2-hydroxybenzamide (1h) were tested in time-kill studies at 1x, 2x, and 4x MIC against all MRSA isolates and the S. aureus reference strain. The antibacterial effect of DMSO [35] used as the solvent of the tested compounds was excluded in this assay, as time-kill curves of this solvent were identical or very similar to those of the growth control. The extent of bacterial killing was estimated by the number of these strains showing a decrease ranging from 1 to 3 log 10 CFU/mL in viable cell count at different times after incubation. A summary of these data is presented in Table 2 . Based on these data it can be concluded that the bactericidal potency of tested diamides against all four strains decreased as follows: 1f > 1h > 1g. No bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) was observed at 1x MIC for any strain and time after incubation tested. At 4x MIC from the four strains, compounds 1f, 1 g, and 1h killed 2, 1, and 2 strains, respectively, at 8 h after incubation and 4, 2, and 2 strains, respectively, at 24 h after incubation.
The findings of time-kill studies for each of the four staphylococci strains at exposure to compounds 1f, 1g, and 1h are summarized in Table 3 . Bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) is expressed in bold.
For compound 1f rapid concentration-dependent antibacterial effect was recorded against clinical isolate of MRSA 63718. Time was not the predictive factor influencing the antibacterial activity because log 10 differences in CFU/mL from the starting inoculum were the same for 4x MIC (with the highest efficiency with a reduction in bacterial count of 5.30 log 10 CFU/mL) or very similar for 2x MIC (with a moderate regrowth after 24 h causing a loss of bactericidal activity) over 24 h. The bactericidal effect was maintained even at 2x MIC at 4 h after incubation for this strain (reduction of 3.08 log 10 CFU/mL). For the remaining strains, clinical isolates of MRSA SA 630, MRSA SA 3202, and S. aureus ATCC 29213, reliable bactericidal effect was recorded at 4x MIC at 24 h after incubation for all these strains with a reduction in bacterial count of 3.22, 3.30, and 3.65 log 10 CFU/mL, respectively.
For compound 1g bactericidal effect against MRSA 63718 was noticed at 2x MIC at 6 and 8 h after incubation and at 4x MIC at 4, 6, and 8 h after incubation with a reduction in bacterial count ranging from 3.10 to 3.58 log 10 CFU/mL. The most effective killing was achieved at 6 h for both concentrations. As in the case of compound 1f, a regrowth was observed after 24 h after incubation. For the remaining isolates of MRSA, SA 630 and SA 3202, bactericidal effect occurred only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.38 and 4.01 log 10 CFU/mL, respectively. The highest bactericidal effect was recorded for MRSA SA 3202 at 4x MIC at 24 h after incubation. A reduction consistent with bacteriostatic effect (0.03 to 2.37 log 10 CFU/mL) was observed at other concentrations over time for both isolates. No bactericidal effect was observed for the S. aureus reference strain; compound 1g demonstrated a pattern of bacteriostatic activity against this strain with a reduction in bacterial count ranging from 0.07 to 2.33 log 10 CFU/mL at 4x MIC over time. In other cases, a slight increase in bacterial counts (i.e., overgrowth) compared with the starting inoculum was observed with values ranging from 0.10 to 1.57 log 10 CFU/mL for this reference strain.
For compound 1h bactericidal effect against MRSA 63718 was maintained at 4x MIC at 6 and 8 h after incubation with a reduction in bacterial count of 3.54 and 3.31 log 10 CFU/mL, respectively. The same as for 1g, the most potent bactericidal effect was maintained at 6 h after incubation. Regrowth at 24 h after incubation causing a loss of bactericidal activity was recorded similarly as with previous compounds. The reason for regrowth of the test organism at 24 h in the experiment is unknown. Most probably, selection of resistant mutants is responsible for this phenomenon [30] ; degradation of the drug in the growth medium is not assumed, as regrowth was
Number of strains showing the following log 10 CFU/mL decrease a at the designated incubation time not observed for any other tested strain. For MRSA SA 630 concentration-dependent killing was recorded at 4x MIC at 6, 8, and 24 h after incubation with log 10 differences in CFU/mL from the starting inoculum being very similar over time (ranging from 3.18 to 3.39 log 10 CFU/mL). For MRSA SA 3202 reliable bactericidal effect was maintained only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.02 log 10 CFU/mL. As for compound 1g, bacteriostatic activity against S. aureus reference strain was observed with a reduction in bacterial count ranging from 0.34 to 2.62 log 10 CFU/mL at 2x and 4x MIC. Overgrowth (values ranging from 0.04 to 1.43 log 10 CFU/mL) was recorded at 1x MIC for this strain. It is of note that in all staphylococci strains with similar MICs and MBCs for compounds 1g and 1h the responsiveness to antibacterial activity of these compounds varied with clinical strains of MRSA being effectively killed and the reference strain remaining unaffected at 4x MIC.
There is a discrepancy between bactericidal results of MBC assay compared with time-kill kinetics. This difference could be caused by comparing microtiter (MBC assay) to macrobroth (time-kill assay) dilutions [36] . Moreover, although time-kill assays are more labour intensive and time consuming than MBC assays, they are recognised to provide a greater degree of characterisation of the cell eradication potential of antibacterial agents [37] .
Concerning antibacterial effect, it is not generally important if the antibacterial agent is also bactericidal at higher concentrations, because the inhibition of bacterial proliferation usually achieves a therapeutic effect; the patient's immune system is capable of coping with the infection then [34] . However, bactericidal therapy could produce a better treatment result by rapid reduction of the bacterial load [38] . Moreover, in the case of an immune system disorder (e.g., immunosuppressive therapy, AIDS patients, etc.) bactericidal agents are unequivocally indicated. Considering steadily escalating numbers of immunocompromised patients with endocarditis, meningitis, or osteomyelitis in recent years, it is necessary to achieve bacterial killing and broaden the spectrum of antimicrobial agents with bactericidal active compounds [30] .
The clinical outcome of MRSA bacteraemia is significantly influenced by vancomycin MIC. Treatment failure exceeding 60% for S. aureus with vancomycin MIC of 4 g/mL resulted in the change of susceptibility breakpoint from 4 g/mL to 2 g/mL by the Clinical and Laboratory Standards Institute (CLSI) in 2006 [23] as well as by the US Food and Drug Administration (FDA) in 2008 [39] . It has been recommended that for infections caused by MRSA strains with elevated vancomycin MICs (2 g/mL), alternative therapy should be considered [40] . It is of note that based on time-kill assays in the present study, all tested diamides (particularly compound 1f exhibiting rapid bactericidal concentration-dependent effect even at 2x MIC) were most effective against isolate MRSA 63718, which is the strain with elevated vancomycin MIC of 2 g/mL. The activity against the remaining isolates with vancomycin MIC of 1 g/mL was lower.
Considering the emergence of decreasing vancomycin susceptibility of MRSA isolates and thus the therapeutic efficacy of vancomycin therapy, our aim was to determine the potential bactericidal role of novel antibacterial compounds against MRSA in vitro. Based on the obtained results, diamides can be suitable candidates for such novel bactericidal active compounds presenting a promising starting point for further investigations to ascertain real in vivo activity and the exact mechanism of action.
The present study is the first evidence of bactericidal effect of SAL analogues. Against other strains, reliable bactericidal effect was maintained at 4x MIC at 24 h after incubation. Considering the necessity to broaden the spectrum of bactericidal agents, diamides from the current study with a novel mechanism of action could present a very promising and interesting solution to this challenge for the future.
|
What is the most common, clinically-relevant multiresistant pathogen in both healthcare and community acquired infections?
| 5,226 |
Methicillin-resistant Staphylococcus aureus (MRSA)
| 1,919 |
1,586 |
In Vitro Bactericidal Activity of 4- and 5-Chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides against MRSA
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4321674/
SHA: f0e6cef57dbae030aea2f324e21e00945ac659cf
Authors: Zadrazilova, Iveta; Pospisilova, Sarka; Pauk, Karel; Imramovsky, Ales; Vinsova, Jarmila; Cizek, Alois; Jampilek, Josef
Date: 2015-01-15
DOI: 10.1155/2015/349534
License: cc-by
Abstract: A series of nine substituted 2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides was assessed as prospective bactericidal agents against three clinical isolates of methicillin-resistant Staphylococcus aureus (MRSA) and S. aureus ATCC 29213 as the reference and quality control strain. The minimum bactericidal concentration was determined by subculturing aliquots from MIC determination onto substance-free agar plates. The bactericidal kinetics of compounds 5-chloro-2-hydroxy-N-[(2S)-3-methyl-1-oxo-1-{[4-(trifluoromethyl)phenyl]amino}butan-2-yl]benzamide (1f), N-{(2S)-1-[(4-bromophenyl)amino]-3-methyl-1-oxobutan-2-yl}-4-chloro-2-hydroxybenzamide (1g), and 4-chloro-N-{(2S)-1-[(3,4-dichlorophenyl)amino]-3-methyl-1-oxobutan-2-yl}-2-hydroxybenzamide (1h) was established by time-kill assay with a final concentration of the compound equal to 1x, 2x, and 4x MIC; aliquots were removed at 0, 4, 6, 8, and 24 h time points. The most potent bactericidal agent was compound 1f exhibiting remarkable rapid concentration-dependent bactericidal effect even at 2x MIC at 4, 6, and 8 h (with a reduction in bacterial count ranging from 3.08 to 3.75 log(10) CFU/mL) and at 4x MIC at 4, 6, 8, and 24 h (5.30 log(10) CFU/mL reduction in bacterial count) after incubation against MRSA 63718. Reliable bactericidal effect against other strains was maintained at 4x MIC at 24 h.
Text: The antibiotic resistance of invasive pathogens has become one of the most challenging and persistent health problems [1] . Methicillin-resistant Staphylococcus aureus (MRSA) has become the most common clinically relevant multiresistant pathogen [2] causing both healthcare-associated and community-acquired bloodstream infections with mortality rates up to 40% [3] .
The prevalence of MRSA is increasing worldwide and, according to the latest information of the European Centre for Disease Prevention and Control from 2012 [4] , can be considered alarming in some European countries, especially in Portugal and Romania, where ≥50% of all S. aureus isolates from invasive infections were identified as MRSA in 2012 (although, e.g., in Romania the prevalence of MRSA was 25-50% in 2010), followed by Italy, Greece, and Poland with 25-50% isolates being MRSA in 2012 (for comparison, in Poland MRSA isolates constituted 10-25% from all S. aureus isolates in 2010).
The treatment failure of vancomycin, the therapeutic anti-MRSA agent of choice, due to the strains with elevated vancomycin minimum inhibitory concentration (MIC) values (i.e., the lowest concentration of an antimicrobial that will inhibit the visible growth of a microorganism) within the susceptible range was described previously [5, 6] . Thus, the emergence of MRSA (and vancomycin-resistant S. aureus in the recent years as well [7] ) makes the discovery of new molecular scaffolds a priority, and the current situation even necessitates the reengineering and repositioning of some old drug families to achieve adequate control of these bacteria [8] . However, for the treatment of S. aureus bloodstream infections, bactericidal antimicrobial agents are considered to be superior to bacteriostatic drugs [9] . This fact should be considered during the development of effective and safe treatment options for MRSA infections.
The history of clinical usage of salicylanilides (2-hydroxy-N-phenylbenzamides) dates back to the 1940s in therapy of tinea capitis, followed by the discovery of their anthelmintic properties in the mid 1950s [10] . Nowadays, salicylanilides (SALs) are a class of aromatic compounds possessing a wide range of interesting pharmacological activities, such as anthelmintic [11] , antibacterial [12, 13] , antimycobacterial [13] , antifungal [14] , and antiviral [15, 16] , among others. Despite being studied since the 1960s, the mechanism of action responsible for biological activities of these compounds has not been explained so far. SALs have been found to inhibit the two-component regulatory systems (TCS) of bacteria [17] . The latest studies specified them also as selective inhibitors of interleukin-12p40 production that plays a specific role in initiation, expansion, and control of cellular response to tuberculosis [18] . Furthermore, salicylanilides have been recognised as inhibitors of some bacterial enzymes, such as sortase A from S. aureus [19] , d-alanine-d-alanine ligase [20] , or transglycosylases from S. aureus (but not from M. tuberculosis) [12] . These enzymes participate in secretion of various proteins or in biosynthesis of bacterial cell wall. Recently, salicylanilides-like derivatives were described to inhibit two enzymes essential for mycobacteria: (i) methionine aminopeptidase, catalyzing a key step of the posttranslational modification of nascent proteins, and (ii) isocitrate lyase, which is essential for the metabolism of fatty acids [21] . Thus, SALs seem to be promising candidates for development of new antibacterial agents with a novel mechanism of action. Such new agents could be a solution to the resistance challenges.
This study is a follow-up paper to a recently published article [13] . The synthesis of the series of novel derivatives of salicylamides, 4-and 5-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides, called diamides due to their skeleton (for general structure see Table 1 ), was described previously [13, 22] , and their antimycobacterial and antibacterial activities against various bacterial species were reported [13] . As these compounds expressed very significant antibacterial activity with low MIC values against clinical isolates of MRSA as representatives of multidrugresistant bacteria, we decided to extend the knowledge about the antibacterial properties of these compounds against MRSA.
The aim of the current study was to assess the overall in vitro bactericidal activity of nine newly synthesized diamides in dependence on time and concentration against clinical isolates of MRSA as representatives of multidrug-resistant bacteria. To the best of our knowledge, this is the first study dealing with the evaluation of novel microbiological characteristics of SAL analogues and revealing their bactericidal effect.
The synthetic pathway of the series of novel diamides was described recently [13, 22] , and their structures (see Table 1 ) were confirmed by IR, NMR, and MS spectrometry, and the purity of the compounds was checked by CHN analysis [13, 22] . [27] ; and MRSA SA 3202 [27] (National Institute of Public Health, Prague, Czech Republic) both of human origin. Suspected colonies were confirmed by PCR; a 108 bp fragment specific for S. aureus was detected [28] . All isolates were tested for the presence of the mecA gene encoding methicillin resistance [29] . These three clinical isolates were classified as vancomycin-susceptible (but with higher MIC of vancomycin equal to 2 g/mL (VA2-MRSA) within the susceptible range for MRSA 63718) methicillinresistant S. aureus (VS-MRSA). For the MICs of vancomycin, see Table 1 . Vancomycin-susceptible methicillin-susceptible Staphylococcus aureus (VS-MSSA) ATCC 29213, obtained from the American Type Culture Collection, was used as the reference and quality control strain. The bacteria were stored at −80 ∘ C and were kept on blood agar plates (Columbia agar base with 5% ovine blood) between experiments. (MBCs) . The MBCs (i.e., the lowest concentrations of antibacterial agents required to kill a particular bacterium) were determined by subculturing aliquots (20 L) from wells with no visible bacterial growth and from control wells of MIC determination onto substance-free Mueller-Hinton agar (MHA) plates. The plates were incubated aerobically at 37 ∘ C for 24 h for colony count. The MBC was defined as the lowest concentration of substance, which produced ≥99.9% killing Table 1 : Chemical structures and in vitro MIC and MBC [ g/mL] values of tested 5-and 4-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides (bactericidal effect of individual compounds against particular strains marked in bold). after 24 h of incubation as compared to the colony count of the starting inoculum [30] . To ensure reproducibility, each MBC assay was performed in at least triplicate on separate occasions.
N H O H N O OH 1 2 R 1 R 3 R 2 Comp. R 1 R 2 R 3 MIC [ g/mL] MBC [ g/mL] 1 2 3 4 1 2 3 4 1a 5-Cl 4-CH 3 (S)-CH 3 >256 >256 >256 >256 >256 >256 >256 >256 1b 5-Cl 4-CH 3 (S)-CH(CH 3 ) 2 >256 >256 32 32 >256 >256 128 >256 1c 5-Cl 4-CH 3 (S)-benzyl >256 >256 >256 >256 >256 >256 >256 >256 1d 5-Cl 4-CH 3 (R)-CH 2 -indolyl >256 >256 >256 >256 >256 >256 >256 >256 1e 5-Cl 4-OCH 3 (S)-CH(CH 3 ) 2 >256 >256 >256 >256 >256 >256 >256 >256 1f 5-Cl 4-CF 3 (S)-CH(CH 3 ) 2 4 2 2 2 4 4 8 4 1g 4-Cl 4-Br (S)-CH(CH 3 ) 2 8 4 4 4 1 6 8 8 8 1h 4-Cl 3,4-Cl (S)-CH(CH 3 ) 2 2 1 1 1 4 1 4 2 1i 4-Cl 3,4-Cl (S)-benzyl 1 1 0.5 0.5 8 1 8 1 AMP - - - >16 >16 >16 0.25 >16 >16 >16 0.25 CPX - - - >16 >16 >16 0.5 >16 >16 >16 0.5 VAN - - - 2 1 1 1 2 1 1 1
Time-kill assays were performed by the broth macrodilution method according to previously described methodology [30] with some modifications. Briefly, flasks containing sterile fresh Mueller-Hinton broth (MHB) with the appropriate antimicrobial agent were inoculated with the test organism in logarithmic growth phase to obtain the starting inoculum with the concentration of approximately 7.5 × 10 6 CFU/mL (actual inoculum concentrations ranged from 0.9 × 10 5 to 2.9 × 10 6 CFU/mL) and a final concentration of the antibiotic equal to 1x, 2x, and 4x MIC in 10 mL volume. For the determination of viable counts, aliquots were removed at 0, 4, 6, 8, and 24 h time points after inoculation, serially diluted in sterile phosphate buffered saline, and aliquots (20 L) were plated on MHA plates in duplicate. Colony counts were performed on plates yielding 6 to 60 colonies, and the mean was calculated. Antimicrobial carry-over was controlled by dilution and visual inspection of the distribution of colonies on the plates with observation of possible inhibition of growth at the site of the initial streaks. The plates were incubated at 37 ∘ C for 24 to 48 h, and the number of colonies was determined. To ensure reproducibility, each time-kill experiment was carried out in duplicate on separate occasions with results presented as the mean of all experiments. The growth control without the addition of antimicrobial agents and the control containing DMSO without any antimicrobial agent to exclude antibacterial activity of this solvent were included. Time-kill curves were constructed by plotting the log 10 CFU per millilitre versus time (over 24 h), and the change in bacterial concentration was determined. The results were analysed by evaluating the numbers of strains that yielded Δ(log 10 CFU/mL) values of −1 (corresponding to 90% killing), −2 (99% killing), and −3 (99.9% killing) at 4, 6, 8, and 24 h compared to counts at 0 h. Bactericidal activity was defined as a reduction of at least 99.9% (≥3 log 10 ) of the total count of CFU/mL in the original inoculum.
Diamides seem to be promising candidates for antibacterial agents with very strong anti-MRSA activity, as it was published recently [13] . In the present study the series of nine newly synthesized diamides was evaluated as prospective bactericidal agents against representatives of multidrugresistant bacteria, three clinical isolates of MRSA, and Staphylococcus aureus ATCC 29213 (methicillin-susceptible) as the reference and quality control strain. Since SALs and their analogues are known as compounds with bacteriostatic effect [31] , this is the first study where SAL-like compounds were considered as prospective bactericidal agents and the dependence of bactericidal effect of these compounds on time and concentration was evaluated. Thus, absolutely novel microbiological characteristics of these compounds were revealed in the present study.
Recently MIC values of diamides expressed as molar concentrations in mol/L were published [13] . To allow comparison with MBC values of the present study, MICs in g/mL were calculated and are recorded in Table 1 along with the activity of reference antibacterial drugs, ampicillin, ciprofloxacin, and vancomycin. Potential bactericidal activity of diamides was assessed using MBC assay [26] . MBC values of all tested compounds are recorded in Table 1 as well.
Based on the obtained results, all compounds assessed as active according to MIC values in our previous study (1f-i) showed low or moderate MBC values against all four strains. The MBC values of these compounds did not exceed the highest tested drug concentration and ranged from 1 to 16 g/mL. In all cases, there were comparable MBC values for the clinical isolates of MRSA and the S. aureus reference strain.
Bactericidal activity is defined as a ratio of MBC to MIC of ≤4 [32] . Table 1 bactericidal activity is expressed in bold.
As mentioned above, SALs are known to exhibit a bacteriostatic effect [31] , so it was very interesting to discover that diamides possess bactericidal activity. The amide bond (-CONH-) can cause interactions with a variety of enzymes [33] ; therefore the presence of two amide bonds could be responsible for the bactericidal effect of diamides against MRSA. The activity of SALs and their analogues results from multiple mechanisms, which are still under investigation; for example, it was found that SALs are capable of inhibiting transglycosylases in later stages of S. aureus (including MRSA) cell wall biosynthesis [12] . These enzymes catalyse the step prior to the transpeptidation in the peptidoglycan biosynthesis and are responsible for polymerization of lipid II, which occurs at the outer face of the membrane [12] . Since antibacterial agents targeting cell wall biosynthesis act as bactericidal agents [30, 34] , the failure in the cell wall biosynthesis due to the inhibition of transglycosylases could be responsible for bactericidal activity of diamides against MRSA.
Based on these findings, antibacterial active diamides with bactericidal effect against all four tested strains as prospective bactericidal agents were chosen for subsequent timekill curve studies to determine the real dependence of bactericidal effect on concentration over time.
1-oxobutan-2-yl}-2-hydroxybenzamide (1h) were tested in time-kill studies at 1x, 2x, and 4x MIC against all MRSA isolates and the S. aureus reference strain. The antibacterial effect of DMSO [35] used as the solvent of the tested compounds was excluded in this assay, as time-kill curves of this solvent were identical or very similar to those of the growth control. The extent of bacterial killing was estimated by the number of these strains showing a decrease ranging from 1 to 3 log 10 CFU/mL in viable cell count at different times after incubation. A summary of these data is presented in Table 2 . Based on these data it can be concluded that the bactericidal potency of tested diamides against all four strains decreased as follows: 1f > 1h > 1g. No bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) was observed at 1x MIC for any strain and time after incubation tested. At 4x MIC from the four strains, compounds 1f, 1 g, and 1h killed 2, 1, and 2 strains, respectively, at 8 h after incubation and 4, 2, and 2 strains, respectively, at 24 h after incubation.
The findings of time-kill studies for each of the four staphylococci strains at exposure to compounds 1f, 1g, and 1h are summarized in Table 3 . Bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) is expressed in bold.
For compound 1f rapid concentration-dependent antibacterial effect was recorded against clinical isolate of MRSA 63718. Time was not the predictive factor influencing the antibacterial activity because log 10 differences in CFU/mL from the starting inoculum were the same for 4x MIC (with the highest efficiency with a reduction in bacterial count of 5.30 log 10 CFU/mL) or very similar for 2x MIC (with a moderate regrowth after 24 h causing a loss of bactericidal activity) over 24 h. The bactericidal effect was maintained even at 2x MIC at 4 h after incubation for this strain (reduction of 3.08 log 10 CFU/mL). For the remaining strains, clinical isolates of MRSA SA 630, MRSA SA 3202, and S. aureus ATCC 29213, reliable bactericidal effect was recorded at 4x MIC at 24 h after incubation for all these strains with a reduction in bacterial count of 3.22, 3.30, and 3.65 log 10 CFU/mL, respectively.
For compound 1g bactericidal effect against MRSA 63718 was noticed at 2x MIC at 6 and 8 h after incubation and at 4x MIC at 4, 6, and 8 h after incubation with a reduction in bacterial count ranging from 3.10 to 3.58 log 10 CFU/mL. The most effective killing was achieved at 6 h for both concentrations. As in the case of compound 1f, a regrowth was observed after 24 h after incubation. For the remaining isolates of MRSA, SA 630 and SA 3202, bactericidal effect occurred only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.38 and 4.01 log 10 CFU/mL, respectively. The highest bactericidal effect was recorded for MRSA SA 3202 at 4x MIC at 24 h after incubation. A reduction consistent with bacteriostatic effect (0.03 to 2.37 log 10 CFU/mL) was observed at other concentrations over time for both isolates. No bactericidal effect was observed for the S. aureus reference strain; compound 1g demonstrated a pattern of bacteriostatic activity against this strain with a reduction in bacterial count ranging from 0.07 to 2.33 log 10 CFU/mL at 4x MIC over time. In other cases, a slight increase in bacterial counts (i.e., overgrowth) compared with the starting inoculum was observed with values ranging from 0.10 to 1.57 log 10 CFU/mL for this reference strain.
For compound 1h bactericidal effect against MRSA 63718 was maintained at 4x MIC at 6 and 8 h after incubation with a reduction in bacterial count of 3.54 and 3.31 log 10 CFU/mL, respectively. The same as for 1g, the most potent bactericidal effect was maintained at 6 h after incubation. Regrowth at 24 h after incubation causing a loss of bactericidal activity was recorded similarly as with previous compounds. The reason for regrowth of the test organism at 24 h in the experiment is unknown. Most probably, selection of resistant mutants is responsible for this phenomenon [30] ; degradation of the drug in the growth medium is not assumed, as regrowth was
Number of strains showing the following log 10 CFU/mL decrease a at the designated incubation time not observed for any other tested strain. For MRSA SA 630 concentration-dependent killing was recorded at 4x MIC at 6, 8, and 24 h after incubation with log 10 differences in CFU/mL from the starting inoculum being very similar over time (ranging from 3.18 to 3.39 log 10 CFU/mL). For MRSA SA 3202 reliable bactericidal effect was maintained only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.02 log 10 CFU/mL. As for compound 1g, bacteriostatic activity against S. aureus reference strain was observed with a reduction in bacterial count ranging from 0.34 to 2.62 log 10 CFU/mL at 2x and 4x MIC. Overgrowth (values ranging from 0.04 to 1.43 log 10 CFU/mL) was recorded at 1x MIC for this strain. It is of note that in all staphylococci strains with similar MICs and MBCs for compounds 1g and 1h the responsiveness to antibacterial activity of these compounds varied with clinical strains of MRSA being effectively killed and the reference strain remaining unaffected at 4x MIC.
There is a discrepancy between bactericidal results of MBC assay compared with time-kill kinetics. This difference could be caused by comparing microtiter (MBC assay) to macrobroth (time-kill assay) dilutions [36] . Moreover, although time-kill assays are more labour intensive and time consuming than MBC assays, they are recognised to provide a greater degree of characterisation of the cell eradication potential of antibacterial agents [37] .
Concerning antibacterial effect, it is not generally important if the antibacterial agent is also bactericidal at higher concentrations, because the inhibition of bacterial proliferation usually achieves a therapeutic effect; the patient's immune system is capable of coping with the infection then [34] . However, bactericidal therapy could produce a better treatment result by rapid reduction of the bacterial load [38] . Moreover, in the case of an immune system disorder (e.g., immunosuppressive therapy, AIDS patients, etc.) bactericidal agents are unequivocally indicated. Considering steadily escalating numbers of immunocompromised patients with endocarditis, meningitis, or osteomyelitis in recent years, it is necessary to achieve bacterial killing and broaden the spectrum of antimicrobial agents with bactericidal active compounds [30] .
The clinical outcome of MRSA bacteraemia is significantly influenced by vancomycin MIC. Treatment failure exceeding 60% for S. aureus with vancomycin MIC of 4 g/mL resulted in the change of susceptibility breakpoint from 4 g/mL to 2 g/mL by the Clinical and Laboratory Standards Institute (CLSI) in 2006 [23] as well as by the US Food and Drug Administration (FDA) in 2008 [39] . It has been recommended that for infections caused by MRSA strains with elevated vancomycin MICs (2 g/mL), alternative therapy should be considered [40] . It is of note that based on time-kill assays in the present study, all tested diamides (particularly compound 1f exhibiting rapid bactericidal concentration-dependent effect even at 2x MIC) were most effective against isolate MRSA 63718, which is the strain with elevated vancomycin MIC of 2 g/mL. The activity against the remaining isolates with vancomycin MIC of 1 g/mL was lower.
Considering the emergence of decreasing vancomycin susceptibility of MRSA isolates and thus the therapeutic efficacy of vancomycin therapy, our aim was to determine the potential bactericidal role of novel antibacterial compounds against MRSA in vitro. Based on the obtained results, diamides can be suitable candidates for such novel bactericidal active compounds presenting a promising starting point for further investigations to ascertain real in vivo activity and the exact mechanism of action.
The present study is the first evidence of bactericidal effect of SAL analogues. Against other strains, reliable bactericidal effect was maintained at 4x MIC at 24 h after incubation. Considering the necessity to broaden the spectrum of bactericidal agents, diamides from the current study with a novel mechanism of action could present a very promising and interesting solution to this challenge for the future.
|
What is the treatment of choice for MRSA infections?
| 5,227 |
vancomycin
| 2,785 |
1,586 |
In Vitro Bactericidal Activity of 4- and 5-Chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides against MRSA
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4321674/
SHA: f0e6cef57dbae030aea2f324e21e00945ac659cf
Authors: Zadrazilova, Iveta; Pospisilova, Sarka; Pauk, Karel; Imramovsky, Ales; Vinsova, Jarmila; Cizek, Alois; Jampilek, Josef
Date: 2015-01-15
DOI: 10.1155/2015/349534
License: cc-by
Abstract: A series of nine substituted 2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides was assessed as prospective bactericidal agents against three clinical isolates of methicillin-resistant Staphylococcus aureus (MRSA) and S. aureus ATCC 29213 as the reference and quality control strain. The minimum bactericidal concentration was determined by subculturing aliquots from MIC determination onto substance-free agar plates. The bactericidal kinetics of compounds 5-chloro-2-hydroxy-N-[(2S)-3-methyl-1-oxo-1-{[4-(trifluoromethyl)phenyl]amino}butan-2-yl]benzamide (1f), N-{(2S)-1-[(4-bromophenyl)amino]-3-methyl-1-oxobutan-2-yl}-4-chloro-2-hydroxybenzamide (1g), and 4-chloro-N-{(2S)-1-[(3,4-dichlorophenyl)amino]-3-methyl-1-oxobutan-2-yl}-2-hydroxybenzamide (1h) was established by time-kill assay with a final concentration of the compound equal to 1x, 2x, and 4x MIC; aliquots were removed at 0, 4, 6, 8, and 24 h time points. The most potent bactericidal agent was compound 1f exhibiting remarkable rapid concentration-dependent bactericidal effect even at 2x MIC at 4, 6, and 8 h (with a reduction in bacterial count ranging from 3.08 to 3.75 log(10) CFU/mL) and at 4x MIC at 4, 6, 8, and 24 h (5.30 log(10) CFU/mL reduction in bacterial count) after incubation against MRSA 63718. Reliable bactericidal effect against other strains was maintained at 4x MIC at 24 h.
Text: The antibiotic resistance of invasive pathogens has become one of the most challenging and persistent health problems [1] . Methicillin-resistant Staphylococcus aureus (MRSA) has become the most common clinically relevant multiresistant pathogen [2] causing both healthcare-associated and community-acquired bloodstream infections with mortality rates up to 40% [3] .
The prevalence of MRSA is increasing worldwide and, according to the latest information of the European Centre for Disease Prevention and Control from 2012 [4] , can be considered alarming in some European countries, especially in Portugal and Romania, where ≥50% of all S. aureus isolates from invasive infections were identified as MRSA in 2012 (although, e.g., in Romania the prevalence of MRSA was 25-50% in 2010), followed by Italy, Greece, and Poland with 25-50% isolates being MRSA in 2012 (for comparison, in Poland MRSA isolates constituted 10-25% from all S. aureus isolates in 2010).
The treatment failure of vancomycin, the therapeutic anti-MRSA agent of choice, due to the strains with elevated vancomycin minimum inhibitory concentration (MIC) values (i.e., the lowest concentration of an antimicrobial that will inhibit the visible growth of a microorganism) within the susceptible range was described previously [5, 6] . Thus, the emergence of MRSA (and vancomycin-resistant S. aureus in the recent years as well [7] ) makes the discovery of new molecular scaffolds a priority, and the current situation even necessitates the reengineering and repositioning of some old drug families to achieve adequate control of these bacteria [8] . However, for the treatment of S. aureus bloodstream infections, bactericidal antimicrobial agents are considered to be superior to bacteriostatic drugs [9] . This fact should be considered during the development of effective and safe treatment options for MRSA infections.
The history of clinical usage of salicylanilides (2-hydroxy-N-phenylbenzamides) dates back to the 1940s in therapy of tinea capitis, followed by the discovery of their anthelmintic properties in the mid 1950s [10] . Nowadays, salicylanilides (SALs) are a class of aromatic compounds possessing a wide range of interesting pharmacological activities, such as anthelmintic [11] , antibacterial [12, 13] , antimycobacterial [13] , antifungal [14] , and antiviral [15, 16] , among others. Despite being studied since the 1960s, the mechanism of action responsible for biological activities of these compounds has not been explained so far. SALs have been found to inhibit the two-component regulatory systems (TCS) of bacteria [17] . The latest studies specified them also as selective inhibitors of interleukin-12p40 production that plays a specific role in initiation, expansion, and control of cellular response to tuberculosis [18] . Furthermore, salicylanilides have been recognised as inhibitors of some bacterial enzymes, such as sortase A from S. aureus [19] , d-alanine-d-alanine ligase [20] , or transglycosylases from S. aureus (but not from M. tuberculosis) [12] . These enzymes participate in secretion of various proteins or in biosynthesis of bacterial cell wall. Recently, salicylanilides-like derivatives were described to inhibit two enzymes essential for mycobacteria: (i) methionine aminopeptidase, catalyzing a key step of the posttranslational modification of nascent proteins, and (ii) isocitrate lyase, which is essential for the metabolism of fatty acids [21] . Thus, SALs seem to be promising candidates for development of new antibacterial agents with a novel mechanism of action. Such new agents could be a solution to the resistance challenges.
This study is a follow-up paper to a recently published article [13] . The synthesis of the series of novel derivatives of salicylamides, 4-and 5-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides, called diamides due to their skeleton (for general structure see Table 1 ), was described previously [13, 22] , and their antimycobacterial and antibacterial activities against various bacterial species were reported [13] . As these compounds expressed very significant antibacterial activity with low MIC values against clinical isolates of MRSA as representatives of multidrugresistant bacteria, we decided to extend the knowledge about the antibacterial properties of these compounds against MRSA.
The aim of the current study was to assess the overall in vitro bactericidal activity of nine newly synthesized diamides in dependence on time and concentration against clinical isolates of MRSA as representatives of multidrug-resistant bacteria. To the best of our knowledge, this is the first study dealing with the evaluation of novel microbiological characteristics of SAL analogues and revealing their bactericidal effect.
The synthetic pathway of the series of novel diamides was described recently [13, 22] , and their structures (see Table 1 ) were confirmed by IR, NMR, and MS spectrometry, and the purity of the compounds was checked by CHN analysis [13, 22] . [27] ; and MRSA SA 3202 [27] (National Institute of Public Health, Prague, Czech Republic) both of human origin. Suspected colonies were confirmed by PCR; a 108 bp fragment specific for S. aureus was detected [28] . All isolates were tested for the presence of the mecA gene encoding methicillin resistance [29] . These three clinical isolates were classified as vancomycin-susceptible (but with higher MIC of vancomycin equal to 2 g/mL (VA2-MRSA) within the susceptible range for MRSA 63718) methicillinresistant S. aureus (VS-MRSA). For the MICs of vancomycin, see Table 1 . Vancomycin-susceptible methicillin-susceptible Staphylococcus aureus (VS-MSSA) ATCC 29213, obtained from the American Type Culture Collection, was used as the reference and quality control strain. The bacteria were stored at −80 ∘ C and were kept on blood agar plates (Columbia agar base with 5% ovine blood) between experiments. (MBCs) . The MBCs (i.e., the lowest concentrations of antibacterial agents required to kill a particular bacterium) were determined by subculturing aliquots (20 L) from wells with no visible bacterial growth and from control wells of MIC determination onto substance-free Mueller-Hinton agar (MHA) plates. The plates were incubated aerobically at 37 ∘ C for 24 h for colony count. The MBC was defined as the lowest concentration of substance, which produced ≥99.9% killing Table 1 : Chemical structures and in vitro MIC and MBC [ g/mL] values of tested 5-and 4-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides (bactericidal effect of individual compounds against particular strains marked in bold). after 24 h of incubation as compared to the colony count of the starting inoculum [30] . To ensure reproducibility, each MBC assay was performed in at least triplicate on separate occasions.
N H O H N O OH 1 2 R 1 R 3 R 2 Comp. R 1 R 2 R 3 MIC [ g/mL] MBC [ g/mL] 1 2 3 4 1 2 3 4 1a 5-Cl 4-CH 3 (S)-CH 3 >256 >256 >256 >256 >256 >256 >256 >256 1b 5-Cl 4-CH 3 (S)-CH(CH 3 ) 2 >256 >256 32 32 >256 >256 128 >256 1c 5-Cl 4-CH 3 (S)-benzyl >256 >256 >256 >256 >256 >256 >256 >256 1d 5-Cl 4-CH 3 (R)-CH 2 -indolyl >256 >256 >256 >256 >256 >256 >256 >256 1e 5-Cl 4-OCH 3 (S)-CH(CH 3 ) 2 >256 >256 >256 >256 >256 >256 >256 >256 1f 5-Cl 4-CF 3 (S)-CH(CH 3 ) 2 4 2 2 2 4 4 8 4 1g 4-Cl 4-Br (S)-CH(CH 3 ) 2 8 4 4 4 1 6 8 8 8 1h 4-Cl 3,4-Cl (S)-CH(CH 3 ) 2 2 1 1 1 4 1 4 2 1i 4-Cl 3,4-Cl (S)-benzyl 1 1 0.5 0.5 8 1 8 1 AMP - - - >16 >16 >16 0.25 >16 >16 >16 0.25 CPX - - - >16 >16 >16 0.5 >16 >16 >16 0.5 VAN - - - 2 1 1 1 2 1 1 1
Time-kill assays were performed by the broth macrodilution method according to previously described methodology [30] with some modifications. Briefly, flasks containing sterile fresh Mueller-Hinton broth (MHB) with the appropriate antimicrobial agent were inoculated with the test organism in logarithmic growth phase to obtain the starting inoculum with the concentration of approximately 7.5 × 10 6 CFU/mL (actual inoculum concentrations ranged from 0.9 × 10 5 to 2.9 × 10 6 CFU/mL) and a final concentration of the antibiotic equal to 1x, 2x, and 4x MIC in 10 mL volume. For the determination of viable counts, aliquots were removed at 0, 4, 6, 8, and 24 h time points after inoculation, serially diluted in sterile phosphate buffered saline, and aliquots (20 L) were plated on MHA plates in duplicate. Colony counts were performed on plates yielding 6 to 60 colonies, and the mean was calculated. Antimicrobial carry-over was controlled by dilution and visual inspection of the distribution of colonies on the plates with observation of possible inhibition of growth at the site of the initial streaks. The plates were incubated at 37 ∘ C for 24 to 48 h, and the number of colonies was determined. To ensure reproducibility, each time-kill experiment was carried out in duplicate on separate occasions with results presented as the mean of all experiments. The growth control without the addition of antimicrobial agents and the control containing DMSO without any antimicrobial agent to exclude antibacterial activity of this solvent were included. Time-kill curves were constructed by plotting the log 10 CFU per millilitre versus time (over 24 h), and the change in bacterial concentration was determined. The results were analysed by evaluating the numbers of strains that yielded Δ(log 10 CFU/mL) values of −1 (corresponding to 90% killing), −2 (99% killing), and −3 (99.9% killing) at 4, 6, 8, and 24 h compared to counts at 0 h. Bactericidal activity was defined as a reduction of at least 99.9% (≥3 log 10 ) of the total count of CFU/mL in the original inoculum.
Diamides seem to be promising candidates for antibacterial agents with very strong anti-MRSA activity, as it was published recently [13] . In the present study the series of nine newly synthesized diamides was evaluated as prospective bactericidal agents against representatives of multidrugresistant bacteria, three clinical isolates of MRSA, and Staphylococcus aureus ATCC 29213 (methicillin-susceptible) as the reference and quality control strain. Since SALs and their analogues are known as compounds with bacteriostatic effect [31] , this is the first study where SAL-like compounds were considered as prospective bactericidal agents and the dependence of bactericidal effect of these compounds on time and concentration was evaluated. Thus, absolutely novel microbiological characteristics of these compounds were revealed in the present study.
Recently MIC values of diamides expressed as molar concentrations in mol/L were published [13] . To allow comparison with MBC values of the present study, MICs in g/mL were calculated and are recorded in Table 1 along with the activity of reference antibacterial drugs, ampicillin, ciprofloxacin, and vancomycin. Potential bactericidal activity of diamides was assessed using MBC assay [26] . MBC values of all tested compounds are recorded in Table 1 as well.
Based on the obtained results, all compounds assessed as active according to MIC values in our previous study (1f-i) showed low or moderate MBC values against all four strains. The MBC values of these compounds did not exceed the highest tested drug concentration and ranged from 1 to 16 g/mL. In all cases, there were comparable MBC values for the clinical isolates of MRSA and the S. aureus reference strain.
Bactericidal activity is defined as a ratio of MBC to MIC of ≤4 [32] . Table 1 bactericidal activity is expressed in bold.
As mentioned above, SALs are known to exhibit a bacteriostatic effect [31] , so it was very interesting to discover that diamides possess bactericidal activity. The amide bond (-CONH-) can cause interactions with a variety of enzymes [33] ; therefore the presence of two amide bonds could be responsible for the bactericidal effect of diamides against MRSA. The activity of SALs and their analogues results from multiple mechanisms, which are still under investigation; for example, it was found that SALs are capable of inhibiting transglycosylases in later stages of S. aureus (including MRSA) cell wall biosynthesis [12] . These enzymes catalyse the step prior to the transpeptidation in the peptidoglycan biosynthesis and are responsible for polymerization of lipid II, which occurs at the outer face of the membrane [12] . Since antibacterial agents targeting cell wall biosynthesis act as bactericidal agents [30, 34] , the failure in the cell wall biosynthesis due to the inhibition of transglycosylases could be responsible for bactericidal activity of diamides against MRSA.
Based on these findings, antibacterial active diamides with bactericidal effect against all four tested strains as prospective bactericidal agents were chosen for subsequent timekill curve studies to determine the real dependence of bactericidal effect on concentration over time.
1-oxobutan-2-yl}-2-hydroxybenzamide (1h) were tested in time-kill studies at 1x, 2x, and 4x MIC against all MRSA isolates and the S. aureus reference strain. The antibacterial effect of DMSO [35] used as the solvent of the tested compounds was excluded in this assay, as time-kill curves of this solvent were identical or very similar to those of the growth control. The extent of bacterial killing was estimated by the number of these strains showing a decrease ranging from 1 to 3 log 10 CFU/mL in viable cell count at different times after incubation. A summary of these data is presented in Table 2 . Based on these data it can be concluded that the bactericidal potency of tested diamides against all four strains decreased as follows: 1f > 1h > 1g. No bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) was observed at 1x MIC for any strain and time after incubation tested. At 4x MIC from the four strains, compounds 1f, 1 g, and 1h killed 2, 1, and 2 strains, respectively, at 8 h after incubation and 4, 2, and 2 strains, respectively, at 24 h after incubation.
The findings of time-kill studies for each of the four staphylococci strains at exposure to compounds 1f, 1g, and 1h are summarized in Table 3 . Bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) is expressed in bold.
For compound 1f rapid concentration-dependent antibacterial effect was recorded against clinical isolate of MRSA 63718. Time was not the predictive factor influencing the antibacterial activity because log 10 differences in CFU/mL from the starting inoculum were the same for 4x MIC (with the highest efficiency with a reduction in bacterial count of 5.30 log 10 CFU/mL) or very similar for 2x MIC (with a moderate regrowth after 24 h causing a loss of bactericidal activity) over 24 h. The bactericidal effect was maintained even at 2x MIC at 4 h after incubation for this strain (reduction of 3.08 log 10 CFU/mL). For the remaining strains, clinical isolates of MRSA SA 630, MRSA SA 3202, and S. aureus ATCC 29213, reliable bactericidal effect was recorded at 4x MIC at 24 h after incubation for all these strains with a reduction in bacterial count of 3.22, 3.30, and 3.65 log 10 CFU/mL, respectively.
For compound 1g bactericidal effect against MRSA 63718 was noticed at 2x MIC at 6 and 8 h after incubation and at 4x MIC at 4, 6, and 8 h after incubation with a reduction in bacterial count ranging from 3.10 to 3.58 log 10 CFU/mL. The most effective killing was achieved at 6 h for both concentrations. As in the case of compound 1f, a regrowth was observed after 24 h after incubation. For the remaining isolates of MRSA, SA 630 and SA 3202, bactericidal effect occurred only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.38 and 4.01 log 10 CFU/mL, respectively. The highest bactericidal effect was recorded for MRSA SA 3202 at 4x MIC at 24 h after incubation. A reduction consistent with bacteriostatic effect (0.03 to 2.37 log 10 CFU/mL) was observed at other concentrations over time for both isolates. No bactericidal effect was observed for the S. aureus reference strain; compound 1g demonstrated a pattern of bacteriostatic activity against this strain with a reduction in bacterial count ranging from 0.07 to 2.33 log 10 CFU/mL at 4x MIC over time. In other cases, a slight increase in bacterial counts (i.e., overgrowth) compared with the starting inoculum was observed with values ranging from 0.10 to 1.57 log 10 CFU/mL for this reference strain.
For compound 1h bactericidal effect against MRSA 63718 was maintained at 4x MIC at 6 and 8 h after incubation with a reduction in bacterial count of 3.54 and 3.31 log 10 CFU/mL, respectively. The same as for 1g, the most potent bactericidal effect was maintained at 6 h after incubation. Regrowth at 24 h after incubation causing a loss of bactericidal activity was recorded similarly as with previous compounds. The reason for regrowth of the test organism at 24 h in the experiment is unknown. Most probably, selection of resistant mutants is responsible for this phenomenon [30] ; degradation of the drug in the growth medium is not assumed, as regrowth was
Number of strains showing the following log 10 CFU/mL decrease a at the designated incubation time not observed for any other tested strain. For MRSA SA 630 concentration-dependent killing was recorded at 4x MIC at 6, 8, and 24 h after incubation with log 10 differences in CFU/mL from the starting inoculum being very similar over time (ranging from 3.18 to 3.39 log 10 CFU/mL). For MRSA SA 3202 reliable bactericidal effect was maintained only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.02 log 10 CFU/mL. As for compound 1g, bacteriostatic activity against S. aureus reference strain was observed with a reduction in bacterial count ranging from 0.34 to 2.62 log 10 CFU/mL at 2x and 4x MIC. Overgrowth (values ranging from 0.04 to 1.43 log 10 CFU/mL) was recorded at 1x MIC for this strain. It is of note that in all staphylococci strains with similar MICs and MBCs for compounds 1g and 1h the responsiveness to antibacterial activity of these compounds varied with clinical strains of MRSA being effectively killed and the reference strain remaining unaffected at 4x MIC.
There is a discrepancy between bactericidal results of MBC assay compared with time-kill kinetics. This difference could be caused by comparing microtiter (MBC assay) to macrobroth (time-kill assay) dilutions [36] . Moreover, although time-kill assays are more labour intensive and time consuming than MBC assays, they are recognised to provide a greater degree of characterisation of the cell eradication potential of antibacterial agents [37] .
Concerning antibacterial effect, it is not generally important if the antibacterial agent is also bactericidal at higher concentrations, because the inhibition of bacterial proliferation usually achieves a therapeutic effect; the patient's immune system is capable of coping with the infection then [34] . However, bactericidal therapy could produce a better treatment result by rapid reduction of the bacterial load [38] . Moreover, in the case of an immune system disorder (e.g., immunosuppressive therapy, AIDS patients, etc.) bactericidal agents are unequivocally indicated. Considering steadily escalating numbers of immunocompromised patients with endocarditis, meningitis, or osteomyelitis in recent years, it is necessary to achieve bacterial killing and broaden the spectrum of antimicrobial agents with bactericidal active compounds [30] .
The clinical outcome of MRSA bacteraemia is significantly influenced by vancomycin MIC. Treatment failure exceeding 60% for S. aureus with vancomycin MIC of 4 g/mL resulted in the change of susceptibility breakpoint from 4 g/mL to 2 g/mL by the Clinical and Laboratory Standards Institute (CLSI) in 2006 [23] as well as by the US Food and Drug Administration (FDA) in 2008 [39] . It has been recommended that for infections caused by MRSA strains with elevated vancomycin MICs (2 g/mL), alternative therapy should be considered [40] . It is of note that based on time-kill assays in the present study, all tested diamides (particularly compound 1f exhibiting rapid bactericidal concentration-dependent effect even at 2x MIC) were most effective against isolate MRSA 63718, which is the strain with elevated vancomycin MIC of 2 g/mL. The activity against the remaining isolates with vancomycin MIC of 1 g/mL was lower.
Considering the emergence of decreasing vancomycin susceptibility of MRSA isolates and thus the therapeutic efficacy of vancomycin therapy, our aim was to determine the potential bactericidal role of novel antibacterial compounds against MRSA in vitro. Based on the obtained results, diamides can be suitable candidates for such novel bactericidal active compounds presenting a promising starting point for further investigations to ascertain real in vivo activity and the exact mechanism of action.
The present study is the first evidence of bactericidal effect of SAL analogues. Against other strains, reliable bactericidal effect was maintained at 4x MIC at 24 h after incubation. Considering the necessity to broaden the spectrum of bactericidal agents, diamides from the current study with a novel mechanism of action could present a very promising and interesting solution to this challenge for the future.
|
What enzyme is essential for the metabolism of fatty acids?
| 5,228 |
isocitrate lyase
| 5,196 |
1,586 |
In Vitro Bactericidal Activity of 4- and 5-Chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides against MRSA
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4321674/
SHA: f0e6cef57dbae030aea2f324e21e00945ac659cf
Authors: Zadrazilova, Iveta; Pospisilova, Sarka; Pauk, Karel; Imramovsky, Ales; Vinsova, Jarmila; Cizek, Alois; Jampilek, Josef
Date: 2015-01-15
DOI: 10.1155/2015/349534
License: cc-by
Abstract: A series of nine substituted 2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides was assessed as prospective bactericidal agents against three clinical isolates of methicillin-resistant Staphylococcus aureus (MRSA) and S. aureus ATCC 29213 as the reference and quality control strain. The minimum bactericidal concentration was determined by subculturing aliquots from MIC determination onto substance-free agar plates. The bactericidal kinetics of compounds 5-chloro-2-hydroxy-N-[(2S)-3-methyl-1-oxo-1-{[4-(trifluoromethyl)phenyl]amino}butan-2-yl]benzamide (1f), N-{(2S)-1-[(4-bromophenyl)amino]-3-methyl-1-oxobutan-2-yl}-4-chloro-2-hydroxybenzamide (1g), and 4-chloro-N-{(2S)-1-[(3,4-dichlorophenyl)amino]-3-methyl-1-oxobutan-2-yl}-2-hydroxybenzamide (1h) was established by time-kill assay with a final concentration of the compound equal to 1x, 2x, and 4x MIC; aliquots were removed at 0, 4, 6, 8, and 24 h time points. The most potent bactericidal agent was compound 1f exhibiting remarkable rapid concentration-dependent bactericidal effect even at 2x MIC at 4, 6, and 8 h (with a reduction in bacterial count ranging from 3.08 to 3.75 log(10) CFU/mL) and at 4x MIC at 4, 6, 8, and 24 h (5.30 log(10) CFU/mL reduction in bacterial count) after incubation against MRSA 63718. Reliable bactericidal effect against other strains was maintained at 4x MIC at 24 h.
Text: The antibiotic resistance of invasive pathogens has become one of the most challenging and persistent health problems [1] . Methicillin-resistant Staphylococcus aureus (MRSA) has become the most common clinically relevant multiresistant pathogen [2] causing both healthcare-associated and community-acquired bloodstream infections with mortality rates up to 40% [3] .
The prevalence of MRSA is increasing worldwide and, according to the latest information of the European Centre for Disease Prevention and Control from 2012 [4] , can be considered alarming in some European countries, especially in Portugal and Romania, where ≥50% of all S. aureus isolates from invasive infections were identified as MRSA in 2012 (although, e.g., in Romania the prevalence of MRSA was 25-50% in 2010), followed by Italy, Greece, and Poland with 25-50% isolates being MRSA in 2012 (for comparison, in Poland MRSA isolates constituted 10-25% from all S. aureus isolates in 2010).
The treatment failure of vancomycin, the therapeutic anti-MRSA agent of choice, due to the strains with elevated vancomycin minimum inhibitory concentration (MIC) values (i.e., the lowest concentration of an antimicrobial that will inhibit the visible growth of a microorganism) within the susceptible range was described previously [5, 6] . Thus, the emergence of MRSA (and vancomycin-resistant S. aureus in the recent years as well [7] ) makes the discovery of new molecular scaffolds a priority, and the current situation even necessitates the reengineering and repositioning of some old drug families to achieve adequate control of these bacteria [8] . However, for the treatment of S. aureus bloodstream infections, bactericidal antimicrobial agents are considered to be superior to bacteriostatic drugs [9] . This fact should be considered during the development of effective and safe treatment options for MRSA infections.
The history of clinical usage of salicylanilides (2-hydroxy-N-phenylbenzamides) dates back to the 1940s in therapy of tinea capitis, followed by the discovery of their anthelmintic properties in the mid 1950s [10] . Nowadays, salicylanilides (SALs) are a class of aromatic compounds possessing a wide range of interesting pharmacological activities, such as anthelmintic [11] , antibacterial [12, 13] , antimycobacterial [13] , antifungal [14] , and antiviral [15, 16] , among others. Despite being studied since the 1960s, the mechanism of action responsible for biological activities of these compounds has not been explained so far. SALs have been found to inhibit the two-component regulatory systems (TCS) of bacteria [17] . The latest studies specified them also as selective inhibitors of interleukin-12p40 production that plays a specific role in initiation, expansion, and control of cellular response to tuberculosis [18] . Furthermore, salicylanilides have been recognised as inhibitors of some bacterial enzymes, such as sortase A from S. aureus [19] , d-alanine-d-alanine ligase [20] , or transglycosylases from S. aureus (but not from M. tuberculosis) [12] . These enzymes participate in secretion of various proteins or in biosynthesis of bacterial cell wall. Recently, salicylanilides-like derivatives were described to inhibit two enzymes essential for mycobacteria: (i) methionine aminopeptidase, catalyzing a key step of the posttranslational modification of nascent proteins, and (ii) isocitrate lyase, which is essential for the metabolism of fatty acids [21] . Thus, SALs seem to be promising candidates for development of new antibacterial agents with a novel mechanism of action. Such new agents could be a solution to the resistance challenges.
This study is a follow-up paper to a recently published article [13] . The synthesis of the series of novel derivatives of salicylamides, 4-and 5-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides, called diamides due to their skeleton (for general structure see Table 1 ), was described previously [13, 22] , and their antimycobacterial and antibacterial activities against various bacterial species were reported [13] . As these compounds expressed very significant antibacterial activity with low MIC values against clinical isolates of MRSA as representatives of multidrugresistant bacteria, we decided to extend the knowledge about the antibacterial properties of these compounds against MRSA.
The aim of the current study was to assess the overall in vitro bactericidal activity of nine newly synthesized diamides in dependence on time and concentration against clinical isolates of MRSA as representatives of multidrug-resistant bacteria. To the best of our knowledge, this is the first study dealing with the evaluation of novel microbiological characteristics of SAL analogues and revealing their bactericidal effect.
The synthetic pathway of the series of novel diamides was described recently [13, 22] , and their structures (see Table 1 ) were confirmed by IR, NMR, and MS spectrometry, and the purity of the compounds was checked by CHN analysis [13, 22] . [27] ; and MRSA SA 3202 [27] (National Institute of Public Health, Prague, Czech Republic) both of human origin. Suspected colonies were confirmed by PCR; a 108 bp fragment specific for S. aureus was detected [28] . All isolates were tested for the presence of the mecA gene encoding methicillin resistance [29] . These three clinical isolates were classified as vancomycin-susceptible (but with higher MIC of vancomycin equal to 2 g/mL (VA2-MRSA) within the susceptible range for MRSA 63718) methicillinresistant S. aureus (VS-MRSA). For the MICs of vancomycin, see Table 1 . Vancomycin-susceptible methicillin-susceptible Staphylococcus aureus (VS-MSSA) ATCC 29213, obtained from the American Type Culture Collection, was used as the reference and quality control strain. The bacteria were stored at −80 ∘ C and were kept on blood agar plates (Columbia agar base with 5% ovine blood) between experiments. (MBCs) . The MBCs (i.e., the lowest concentrations of antibacterial agents required to kill a particular bacterium) were determined by subculturing aliquots (20 L) from wells with no visible bacterial growth and from control wells of MIC determination onto substance-free Mueller-Hinton agar (MHA) plates. The plates were incubated aerobically at 37 ∘ C for 24 h for colony count. The MBC was defined as the lowest concentration of substance, which produced ≥99.9% killing Table 1 : Chemical structures and in vitro MIC and MBC [ g/mL] values of tested 5-and 4-chloro-2-hydroxy-N-[1-oxo-1-(phenylamino)alkan-2-yl]benzamides (bactericidal effect of individual compounds against particular strains marked in bold). after 24 h of incubation as compared to the colony count of the starting inoculum [30] . To ensure reproducibility, each MBC assay was performed in at least triplicate on separate occasions.
N H O H N O OH 1 2 R 1 R 3 R 2 Comp. R 1 R 2 R 3 MIC [ g/mL] MBC [ g/mL] 1 2 3 4 1 2 3 4 1a 5-Cl 4-CH 3 (S)-CH 3 >256 >256 >256 >256 >256 >256 >256 >256 1b 5-Cl 4-CH 3 (S)-CH(CH 3 ) 2 >256 >256 32 32 >256 >256 128 >256 1c 5-Cl 4-CH 3 (S)-benzyl >256 >256 >256 >256 >256 >256 >256 >256 1d 5-Cl 4-CH 3 (R)-CH 2 -indolyl >256 >256 >256 >256 >256 >256 >256 >256 1e 5-Cl 4-OCH 3 (S)-CH(CH 3 ) 2 >256 >256 >256 >256 >256 >256 >256 >256 1f 5-Cl 4-CF 3 (S)-CH(CH 3 ) 2 4 2 2 2 4 4 8 4 1g 4-Cl 4-Br (S)-CH(CH 3 ) 2 8 4 4 4 1 6 8 8 8 1h 4-Cl 3,4-Cl (S)-CH(CH 3 ) 2 2 1 1 1 4 1 4 2 1i 4-Cl 3,4-Cl (S)-benzyl 1 1 0.5 0.5 8 1 8 1 AMP - - - >16 >16 >16 0.25 >16 >16 >16 0.25 CPX - - - >16 >16 >16 0.5 >16 >16 >16 0.5 VAN - - - 2 1 1 1 2 1 1 1
Time-kill assays were performed by the broth macrodilution method according to previously described methodology [30] with some modifications. Briefly, flasks containing sterile fresh Mueller-Hinton broth (MHB) with the appropriate antimicrobial agent were inoculated with the test organism in logarithmic growth phase to obtain the starting inoculum with the concentration of approximately 7.5 × 10 6 CFU/mL (actual inoculum concentrations ranged from 0.9 × 10 5 to 2.9 × 10 6 CFU/mL) and a final concentration of the antibiotic equal to 1x, 2x, and 4x MIC in 10 mL volume. For the determination of viable counts, aliquots were removed at 0, 4, 6, 8, and 24 h time points after inoculation, serially diluted in sterile phosphate buffered saline, and aliquots (20 L) were plated on MHA plates in duplicate. Colony counts were performed on plates yielding 6 to 60 colonies, and the mean was calculated. Antimicrobial carry-over was controlled by dilution and visual inspection of the distribution of colonies on the plates with observation of possible inhibition of growth at the site of the initial streaks. The plates were incubated at 37 ∘ C for 24 to 48 h, and the number of colonies was determined. To ensure reproducibility, each time-kill experiment was carried out in duplicate on separate occasions with results presented as the mean of all experiments. The growth control without the addition of antimicrobial agents and the control containing DMSO without any antimicrobial agent to exclude antibacterial activity of this solvent were included. Time-kill curves were constructed by plotting the log 10 CFU per millilitre versus time (over 24 h), and the change in bacterial concentration was determined. The results were analysed by evaluating the numbers of strains that yielded Δ(log 10 CFU/mL) values of −1 (corresponding to 90% killing), −2 (99% killing), and −3 (99.9% killing) at 4, 6, 8, and 24 h compared to counts at 0 h. Bactericidal activity was defined as a reduction of at least 99.9% (≥3 log 10 ) of the total count of CFU/mL in the original inoculum.
Diamides seem to be promising candidates for antibacterial agents with very strong anti-MRSA activity, as it was published recently [13] . In the present study the series of nine newly synthesized diamides was evaluated as prospective bactericidal agents against representatives of multidrugresistant bacteria, three clinical isolates of MRSA, and Staphylococcus aureus ATCC 29213 (methicillin-susceptible) as the reference and quality control strain. Since SALs and their analogues are known as compounds with bacteriostatic effect [31] , this is the first study where SAL-like compounds were considered as prospective bactericidal agents and the dependence of bactericidal effect of these compounds on time and concentration was evaluated. Thus, absolutely novel microbiological characteristics of these compounds were revealed in the present study.
Recently MIC values of diamides expressed as molar concentrations in mol/L were published [13] . To allow comparison with MBC values of the present study, MICs in g/mL were calculated and are recorded in Table 1 along with the activity of reference antibacterial drugs, ampicillin, ciprofloxacin, and vancomycin. Potential bactericidal activity of diamides was assessed using MBC assay [26] . MBC values of all tested compounds are recorded in Table 1 as well.
Based on the obtained results, all compounds assessed as active according to MIC values in our previous study (1f-i) showed low or moderate MBC values against all four strains. The MBC values of these compounds did not exceed the highest tested drug concentration and ranged from 1 to 16 g/mL. In all cases, there were comparable MBC values for the clinical isolates of MRSA and the S. aureus reference strain.
Bactericidal activity is defined as a ratio of MBC to MIC of ≤4 [32] . Table 1 bactericidal activity is expressed in bold.
As mentioned above, SALs are known to exhibit a bacteriostatic effect [31] , so it was very interesting to discover that diamides possess bactericidal activity. The amide bond (-CONH-) can cause interactions with a variety of enzymes [33] ; therefore the presence of two amide bonds could be responsible for the bactericidal effect of diamides against MRSA. The activity of SALs and their analogues results from multiple mechanisms, which are still under investigation; for example, it was found that SALs are capable of inhibiting transglycosylases in later stages of S. aureus (including MRSA) cell wall biosynthesis [12] . These enzymes catalyse the step prior to the transpeptidation in the peptidoglycan biosynthesis and are responsible for polymerization of lipid II, which occurs at the outer face of the membrane [12] . Since antibacterial agents targeting cell wall biosynthesis act as bactericidal agents [30, 34] , the failure in the cell wall biosynthesis due to the inhibition of transglycosylases could be responsible for bactericidal activity of diamides against MRSA.
Based on these findings, antibacterial active diamides with bactericidal effect against all four tested strains as prospective bactericidal agents were chosen for subsequent timekill curve studies to determine the real dependence of bactericidal effect on concentration over time.
1-oxobutan-2-yl}-2-hydroxybenzamide (1h) were tested in time-kill studies at 1x, 2x, and 4x MIC against all MRSA isolates and the S. aureus reference strain. The antibacterial effect of DMSO [35] used as the solvent of the tested compounds was excluded in this assay, as time-kill curves of this solvent were identical or very similar to those of the growth control. The extent of bacterial killing was estimated by the number of these strains showing a decrease ranging from 1 to 3 log 10 CFU/mL in viable cell count at different times after incubation. A summary of these data is presented in Table 2 . Based on these data it can be concluded that the bactericidal potency of tested diamides against all four strains decreased as follows: 1f > 1h > 1g. No bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) was observed at 1x MIC for any strain and time after incubation tested. At 4x MIC from the four strains, compounds 1f, 1 g, and 1h killed 2, 1, and 2 strains, respectively, at 8 h after incubation and 4, 2, and 2 strains, respectively, at 24 h after incubation.
The findings of time-kill studies for each of the four staphylococci strains at exposure to compounds 1f, 1g, and 1h are summarized in Table 3 . Bactericidal activity (i.e., ≥3 log 10 CFU/mL decrease) is expressed in bold.
For compound 1f rapid concentration-dependent antibacterial effect was recorded against clinical isolate of MRSA 63718. Time was not the predictive factor influencing the antibacterial activity because log 10 differences in CFU/mL from the starting inoculum were the same for 4x MIC (with the highest efficiency with a reduction in bacterial count of 5.30 log 10 CFU/mL) or very similar for 2x MIC (with a moderate regrowth after 24 h causing a loss of bactericidal activity) over 24 h. The bactericidal effect was maintained even at 2x MIC at 4 h after incubation for this strain (reduction of 3.08 log 10 CFU/mL). For the remaining strains, clinical isolates of MRSA SA 630, MRSA SA 3202, and S. aureus ATCC 29213, reliable bactericidal effect was recorded at 4x MIC at 24 h after incubation for all these strains with a reduction in bacterial count of 3.22, 3.30, and 3.65 log 10 CFU/mL, respectively.
For compound 1g bactericidal effect against MRSA 63718 was noticed at 2x MIC at 6 and 8 h after incubation and at 4x MIC at 4, 6, and 8 h after incubation with a reduction in bacterial count ranging from 3.10 to 3.58 log 10 CFU/mL. The most effective killing was achieved at 6 h for both concentrations. As in the case of compound 1f, a regrowth was observed after 24 h after incubation. For the remaining isolates of MRSA, SA 630 and SA 3202, bactericidal effect occurred only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.38 and 4.01 log 10 CFU/mL, respectively. The highest bactericidal effect was recorded for MRSA SA 3202 at 4x MIC at 24 h after incubation. A reduction consistent with bacteriostatic effect (0.03 to 2.37 log 10 CFU/mL) was observed at other concentrations over time for both isolates. No bactericidal effect was observed for the S. aureus reference strain; compound 1g demonstrated a pattern of bacteriostatic activity against this strain with a reduction in bacterial count ranging from 0.07 to 2.33 log 10 CFU/mL at 4x MIC over time. In other cases, a slight increase in bacterial counts (i.e., overgrowth) compared with the starting inoculum was observed with values ranging from 0.10 to 1.57 log 10 CFU/mL for this reference strain.
For compound 1h bactericidal effect against MRSA 63718 was maintained at 4x MIC at 6 and 8 h after incubation with a reduction in bacterial count of 3.54 and 3.31 log 10 CFU/mL, respectively. The same as for 1g, the most potent bactericidal effect was maintained at 6 h after incubation. Regrowth at 24 h after incubation causing a loss of bactericidal activity was recorded similarly as with previous compounds. The reason for regrowth of the test organism at 24 h in the experiment is unknown. Most probably, selection of resistant mutants is responsible for this phenomenon [30] ; degradation of the drug in the growth medium is not assumed, as regrowth was
Number of strains showing the following log 10 CFU/mL decrease a at the designated incubation time not observed for any other tested strain. For MRSA SA 630 concentration-dependent killing was recorded at 4x MIC at 6, 8, and 24 h after incubation with log 10 differences in CFU/mL from the starting inoculum being very similar over time (ranging from 3.18 to 3.39 log 10 CFU/mL). For MRSA SA 3202 reliable bactericidal effect was maintained only at 4x MIC at 24 h after incubation with a reduction in bacterial count of 3.02 log 10 CFU/mL. As for compound 1g, bacteriostatic activity against S. aureus reference strain was observed with a reduction in bacterial count ranging from 0.34 to 2.62 log 10 CFU/mL at 2x and 4x MIC. Overgrowth (values ranging from 0.04 to 1.43 log 10 CFU/mL) was recorded at 1x MIC for this strain. It is of note that in all staphylococci strains with similar MICs and MBCs for compounds 1g and 1h the responsiveness to antibacterial activity of these compounds varied with clinical strains of MRSA being effectively killed and the reference strain remaining unaffected at 4x MIC.
There is a discrepancy between bactericidal results of MBC assay compared with time-kill kinetics. This difference could be caused by comparing microtiter (MBC assay) to macrobroth (time-kill assay) dilutions [36] . Moreover, although time-kill assays are more labour intensive and time consuming than MBC assays, they are recognised to provide a greater degree of characterisation of the cell eradication potential of antibacterial agents [37] .
Concerning antibacterial effect, it is not generally important if the antibacterial agent is also bactericidal at higher concentrations, because the inhibition of bacterial proliferation usually achieves a therapeutic effect; the patient's immune system is capable of coping with the infection then [34] . However, bactericidal therapy could produce a better treatment result by rapid reduction of the bacterial load [38] . Moreover, in the case of an immune system disorder (e.g., immunosuppressive therapy, AIDS patients, etc.) bactericidal agents are unequivocally indicated. Considering steadily escalating numbers of immunocompromised patients with endocarditis, meningitis, or osteomyelitis in recent years, it is necessary to achieve bacterial killing and broaden the spectrum of antimicrobial agents with bactericidal active compounds [30] .
The clinical outcome of MRSA bacteraemia is significantly influenced by vancomycin MIC. Treatment failure exceeding 60% for S. aureus with vancomycin MIC of 4 g/mL resulted in the change of susceptibility breakpoint from 4 g/mL to 2 g/mL by the Clinical and Laboratory Standards Institute (CLSI) in 2006 [23] as well as by the US Food and Drug Administration (FDA) in 2008 [39] . It has been recommended that for infections caused by MRSA strains with elevated vancomycin MICs (2 g/mL), alternative therapy should be considered [40] . It is of note that based on time-kill assays in the present study, all tested diamides (particularly compound 1f exhibiting rapid bactericidal concentration-dependent effect even at 2x MIC) were most effective against isolate MRSA 63718, which is the strain with elevated vancomycin MIC of 2 g/mL. The activity against the remaining isolates with vancomycin MIC of 1 g/mL was lower.
Considering the emergence of decreasing vancomycin susceptibility of MRSA isolates and thus the therapeutic efficacy of vancomycin therapy, our aim was to determine the potential bactericidal role of novel antibacterial compounds against MRSA in vitro. Based on the obtained results, diamides can be suitable candidates for such novel bactericidal active compounds presenting a promising starting point for further investigations to ascertain real in vivo activity and the exact mechanism of action.
The present study is the first evidence of bactericidal effect of SAL analogues. Against other strains, reliable bactericidal effect was maintained at 4x MIC at 24 h after incubation. Considering the necessity to broaden the spectrum of bactericidal agents, diamides from the current study with a novel mechanism of action could present a very promising and interesting solution to this challenge for the future.
|
What was the purpose of this research?
| 5,229 |
to assess the overall in vitro bactericidal activity of nine newly synthesized diamides
| 6,207 |
1,587 |
Exploring the Innate Immunological Response of an Alternative Nonhuman Primate Model of Infectious Disease; the Common Marmoset
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4129158/
SHA: f4c43e4ae49ca69dbac32620bd0a73ecbb683b91
Authors: Nelson, M.; Loveday, M.
Date: 2014-07-22
DOI: 10.1155/2014/913632
License: cc-by
Abstract: The common marmoset (Callithrix jacchus) is increasingly being utilised as a nonhuman primate model for human disease, ranging from autoimmune to infectious disease. In order to fully exploit these models, meaningful comparison to the human host response is necessary. Commercially available reagents, primarily targeted to human cells, were utilised to assess the phenotype and activation status of key immune cell types and cytokines in naive and infected animals. Single cell suspensions of blood, spleen, and lung were examined. Generally, the phenotype of cells was comparable between humans and marmosets, with approximately 63% of all lymphocytes in the blood of marmosets being T cells, 25% B-cells, and 12% NK cells. The percentage of neutrophils in marmoset blood were more similar to human values than mouse values. Comparison of the activation status of cells following experimental systemic or inhalational infection exhibited different trends in different tissues, most obvious in cell types active in the innate immune response. This work significantly enhances the ability to understand the immune response in these animals and fortifies their use as models of infectious disease.
Text: The common marmoset (Callithrix jacchus), a New World monkey (NWM) species is a small, arboreal nonhuman primate (NHP), native to the Atlantic Coastal Forest in Northeast Brazil and parts of South East Brazil. In recent years the common marmoset has become more widely used in applied biomedical research, and an increasing body of evidence suggests the physiological and immunological responses to biological insults are similar between marmosets and humans [1] . In the field of infectious disease, the marmoset is primarily being investigated as an alternative NHP model to complement the more traditionally used Old World monkeys (OWM) (e.g., rhesus and cynomolgus macaques). Evolutionarily, both NWM and OWM sit within the simiiformes infraorder of the suborder Haplorhini of primates [2] . Marmosets sit within the family Callitrichidae of the Platyrrhini parvorder, while OWM sit within the Cercopithecidae family of the Catarrhini Parvorder. Marmosets therefore are separated from Old World monkeys by one ancestral step and are a lower order primate.
Marmosets have been used to model the infection syndrome caused by a number of public health pathogens including Lassa virus [3] , Hepatitis C virus [4] , Dengue virus [5] , Herpesvirus [6] , Junin virus [7] Rift Valley Fever [8] , and SARS [9] . Marmosets have also been used to model a number of biodefense pathogens including Eastern Equine Encephalitis virus [10] , Bacillus anthracis [11] , Francisella tularensis [12, 13] , Burkholderia pseudomallei [14] , Marburg haemorrhagic fever virus [15, 16] , Ebola haemorrhagic fever virus [16] , and Variola virus [17] . The utility of marmosets to assess medical countermeasures has also been demonstrated; a vaccine has been tested for Lassa fever [18] and the efficacy of ciprofloxacin and levofloxacin has been tested as postexposure therapies for anthrax and tularemia, respectively [19, 20] .
In order to exploit these models fully and to allow meaningful comparison with the human condition, the response of the immune system to infection/therapy needs to be 2 Journal of Immunology Research characterised and understood. Generally, NHPs have a close molecular, immunological, reproductive, and neurological similarity with humans making them ideal surrogates for humans and the study of infectious diseases. There is a high level of gene homology between humans and NHPs which underlies physiological and biochemical similarities. Similarities at the genetic level extend to the phenotypical level making NHPs well suited to modelling pathophysiological responses in man [21] . Immunologically, there is a high degree of homology between humans and marmosets [22] . The similarity of various immunological factors produced by humans and marmosets has been investigated at both the genetic and protein levels. There is at least 95% homology between human costimulatory molecules (e.g., CD80, CD86 etc.) and those of marmosets [23] . Also the immunoglobulin and T-cell receptor repertoire of humans and marmosets show at least 80% homology [24, 25] .
Currently, the availability of commercial reagents specifically designed for the marmoset is limited although a number of antibodies designed for use with human samples have been shown to cross-react with leucocytes from marmoset blood [26] [27] [28] . However, these reagents have not been exploited to investigate the immune response to infectious disease. To date, investigation of the immune response in marmosets has primarily been achieved using pathogen-specific antibodies to determine the serological response using ELISA such as in the smallpox, Dengue, Rift Valley Fever, and Herpes models [5, 6, 8, 17] or by immunohistochemistry to identify, for example, CD8+, CD3+, CD20+ cells, and IL-6 in the smallpox model [17] ; neutrophils and macrophages in the Herpes model [6] ; or CD3+ and CD20+ cells in the Lassa model [3] .
The work presented here focuses on understanding the immune profile of the naive marmoset as well as identifying and quantifying the immune response to infectious disease. The aim of this work is to determine key changes and identify correlates of infection or protection.
Healthy sexually mature common marmosets (C. jacchus) were obtained from the Dstl Porton Down breeding colony and housed in vasectomized male and female pairs. The Dstl colony was established during the 1970s and is a closed colony with a stable genotype. Animals included in these studies were mixed sex pairs, between 18 months and 5 years old and weighing between 320 g to 500 g. All animals were allowed free access to food and water as well as environmental enrichment. All animal studies were carried out in accordance with the UK Animals (Scientific Procedures) Act of 1986 and the Codes of Practice for the Housing and Care of Animals used in Scientific Procedures 1989. Animals were challenged with an intracellular pathogen by either the subcutaneous or inhalational route and were humanely killed at various time points after challenge. Prior to the infection study, animals were bled to determine baseline immunological parameters. Studies were performed to establish infection models in order to evaluate the efficacy of suitable therapies for transition ultimately to the clinic.
Populations. Blood and tissue samples were homogenised to provide single cell suspensions [12] . Red blood cells were lysed, and the mixed leucocyte population was washed and stained with various combinations of the following fluorescent antibody stains: CD3 (SP34-2), CD8 (LT8), CD11c (SHCL3), CD14 (M5E2), CD16 (3G8), CD20 (Bly1), CD45RA (5H9), CD54 (HCD54), CD56 (B159), CD69 (FN50), CD163 (GHI/61), and MCHII (L243) (BD Bioscience, Insight Bioscience, AbD serotec). Samples were fixed in 4% paraformaldehyde for 48 hrs at 4 ∘ C and analysed by flow cytometry (FACScanto II BD) within 72 hours of staining.
Levels of circulating cytokines and chemokines were also quantified in the blood of marmosets from the Dstl colony using human multiplex kits available commercially (BD cytokine flex beads and the Luminex system). These systems show significant cross-reactivity with the marmoset suggesting a high degree of conservation between the two species for IL-6, MIP-1 , MIP-1 , and MCP-1 [29] . However, for other cytokines that are pivotal in the innate response, TNF and IFN reagents were obtained from U-CyTech Biosciences and Mabtech AB, respectively, due to a lack of cross-reactivity observed within the kit obtained from BD [13] .
In order to fully characterise the immune response to infectious agent in the marmoset, single cell suspensions of lung and spleen tissue were also examined in conjunction with the traditionally used blood cells. These tissue homogenates are of particular interest in relation to target sites of infection: the lung as the site of initial infection following an inhalational challenge and the spleen as a representative organ following a parental challenge. Cell types targeted during this analysis include cells important in the innate response (e.g., neutrophils, macrophages, and NK cells) and the adaptive response (T and B cells) with a view to determine the response to infection and vaccination and to derive immune correlates of infection/protection. Dapi was included as a nuclear marker to ensure that the initial gating included only intact cells. Basic cell types in blood were easily identified by measuring size (forward) and granularity (side) scatter (Figure 1(a) ). Identification of cell types in tissue samples was more difficult as the scatter profiles are less clearly compartmentalized. The common leukocyte antigen (CD45) normally used to locate all leukocytes in human samples also worked well in marmoset blood but failed to provide relevant information in the tissue samples. Confirmation of neutrophil identification was done by nuclear morphology and macrophages were identified by their adherent nature in initial experiments (data not shown). Neutrophils were stained as CD11c dim CD14− and macrophages as CD11c + CD14+ regardless of tissue origin (Figure 1(b) ). Figure 1 shows the basic division of lymphocytes between T, B, and NK cells from a healthy blood sample.
Using this approach, the percentage of NK cells, B-cells, total T-cells, CD8+ T-cells, neutrophils, and monocytes was determined in the blood of naive marmosets (Figure 2 (a), Table 1 ); approximately 63% of all lymphocytes were T cells, 25% B cells, and 12% NK cells. The variability of the data is depicted in Figure 2 (a) with the greatest variability observed in the proportion of neutrophils. There were no obvious differences attributable to age or sex of the animals. This analysis was also applied to lung and spleen homogenates from naive marmosets (Figures 2(b) and 2(c) ). Greater variability was observed in the data relating to the identification of cell types in tissue samples, attributed to the inherent difficulties in identifying cell types in tissue homogenates by size and granularity and also the smaller cohort of animals. As expected, low numbers of neutrophils are found in naive spleen or lung tissue (8% both). Healthy mouse spleens typically have approximately 1-2% granulocytes [30] .
Understandably, there are few reports on the typical cell percentages expected in healthy human individuals for these tissues. However, it is reported that B cells are more prevalent in the spleens of humans at a ratio of 5 to 4 B to T cells than in the lungs which have a ratio of 1 to 8 B to T cells [34] . In marmoset data reported here, a ratio of 2 to 3 B to T cells in the spleen and 1 to 6 B to T-cells in the lungs was observed compared to a ratio of 3 to 2 B to T cells in mouse spleens [30] .
Upon comparison, the marmoset data is generally consistent with previously reported data which is only available for marmoset blood samples [27] and information available for human blood [32, 33] (Table 1 ). However, one report found the proportion of CD8+ T-cells was almost three times greater in marmosets than humans, 61% to 21% respectively [35] compared to the 30% observed in this study and the work previously reported by Brok et al. [27] . Brok's study involved a small number of animals (eight) and also used a different CD8+ clone to identify cells. Contrastingly, in mice, differences are observed in the proportion of both B cells and neutrophils [31] , although these differences are highly strain specific. C57BL/6J mice are reported to have 67% B cells and BALB/C mice 46%; both of which are consistently higher than the percentage found in marmosets and humans of approximately 25% (Table 1 ) [27, 31] . The proportion of neutrophils found in the blood of C57BL/6J mice at 13% is lower than the 35% found in marmosets and the 40-75% expected for healthy human blood. This is encouraging as neutrophils play a pivotal role in the innate response to infection [36] . A cross-species comparison suggests that monocytes comprise 3% of leukocytes ( Table 1) . Levels of circulating cytokines and chemokines (IL-6, IL-1 , MIP-1 , MCP-1, Rantes, TNF , and IFN ) were also quantified in the blood, lung, and spleen of naïve marmosets from the Dstl colony. None of these cytokines were detected in blood samples from uninfected animals; however low levels of MIP-1 , MCP-1, and Rantes were found in spleen and lung tissue.
Preliminary investigation of the immune response has supported the development of marmoset model of infection at Dstl. The levels of different cell types were measured at specific times after challenge with inhalational F. tularensis, B. pseudomallei, and Marburg virus [13] [14] [15] . Following challenge with F. tularensis, increasing levels of NK cells, neutrophils, T cells, and macrophages were observed, peaking at 48 hours after challenge before rapidly declining. This study also demonstrated the importance of investigating the immunological response in key target organs, as an increase in CD8+ T cells and T cells was observed in the spleen and lungs but not in the blood. Increasing levels of various cytokines, MCP-1, MIP-1 , MIP-1 , IL-6, and IL-1 , were observed in Table 1 : Comparison of the percentages of different cell types observed in the blood from healthy marmosets, mice, and humans.
Identification markers Marmoset (present data)
Marmoset [27] Mouse 4 [30, 31] Human Asian [32] Human Caucasian [33] Number the lungs, spleen, and blood as the disease progressed (TNF and IFN were not measured in this study). Following inhalational challenge of marmosets with B. pseudomallei, an increase in the number of neutrophils was observed in the blood at 36 hours after challenge, followed by a rapid decline that was associated with an influx of neutrophils into the lung at 46 hours after challenge. A subsequent decline in the number of neutrophils in the lung was associated with the increased number in the spleen of animals that exhibited severe disease and were humanely killed. There was a gradual increase in the number of macrophages in the spleen as the disease progressed with numbers of macrophages peaking in the blood and lungs at 36 hours after challenge. A rapid decline in the number of macrophages in the lungs and blood was observed by 46 hours after challenge.
The levels of various cell types and cytokines were also measured in the blood of animals following inhalational challenge with Marburg virus [15] . In these animals a general increase in the numbers of T cells, NK cells, macrophages IFN-, IL-1 , and MCP-1 was observed with time (TNF was not measured).
In order to gain more information from these acute bacterial infection models, we have sought out other markers from the literature. Primarily this was from marmoset models of autoimmune disorders such as rheumatoid arthritis and multiple sclerosis where the cross-reactivity of human antibodies was investigated, as well as the functionality of cells [37] [38] [39] [40] . More recent work at Dstl has reported further cross-reactivity between marmoset cells and human cytokines to induce activity in marmoset T cells [36, 41] . These studies, combined with increasing information available on the cross-reactivity of human antibodies to various NHPs (e.g., NIH NHP reagent resource, http://www.nhpreagents.org/NHP/default.aspx), has expanded the ability to assess activation markers for disease. Detection of the following cell surface markers with human antibodies was trialed: CD54 (ICAM-1) associated with cellular adhesion, inflammation, and leukocyte extravasation; CD69 the early activation marker; CD16 as a macrophage activation marker; CD163 the alternative macrophage activation marker; and MHC class II (HLA-DR). CD56 was originally included to identify NK cells; however, it was noted that its expression on T cells was upregulated during disease and that cells defined as CD3+ CD16− CD56+ have been shown to be functionally cytotoxic in marmosets [37, 42] .
These markers have been used to expand on our previously published work to determine changes in the activation status of basic cell types in response to an acute bacterial infection. Animals were challenged with bacteria at a comparable dose either by inhalation ( = 22) or by a systemic route ( = 12) and humanely killed once they had reached a humane endpoint (between day 4 and day 5 after challenge). Figure 3 illustrates the cellular activity in representative tissues following inhalational (Figures 3(b) and 3(e)) or systemic challenge (Figures 3(c) and 3(f)) and in naïve samples (Figures 3(a) and 3(d) ). Naïve T and NK cells appear to have similar resting activation states regardless of origin, whereas neutrophils and macrophages have differential expression of activation, for example, CD16. In response to disease, the proportions of the cell types appear to remain relativity constant; however, the activation markers provide more detailed information and show involvement of all the cell types explored. Extensive activation was to be expected considering that the samples were taken at the humane endpoint. There is also extensive variation between The response to infection within the lungs has similarities across disease routes in terms of neutrophil reduced expression of CD16 and CD54 and macrophage increased expression of CD16 and reduction in MHCII. Unexpectedly, the T and NK cells appear to be more actively involved in systemic disease, indicating that the disease develops a pneumonic element regardless of initial route of infection.
Levels of circulating cytokines and chemokines (IL-6, IL-1 , MIP-1 , MCP-1, Rantes, TNF , and IFN ) were also quantified in the lung and spleen samples. All of the cytokines (with the exception of Rantes) were expressed at high levels (ng/mg) in all samples, which was expected as the animals had succumbed to terminal disease.
The work presented here adds significant relevant information to the marmoset models of infection and to the understanding of the immune response in these animals. This work extends marmoset immunology from autoimmune disorders into the field of infectious diseases; this coupled with an increase in the information available on crossreactivity of human reagents to a variety of NHPs increases the utility/application of marmosets as models of human disease. In conclusion, the immune response in marmosets to infectious disease can be characterised in terms of the phenotype and activation status of all the major immune cells and key cytokine and chemokine expression. This can aid in the identification of correlates of infection or protection in medical countermeasures assessment studies. This information can also potentially be used for pivotal studies to support licensure of products under the FDA Animal Rule.
This, in conjunction with the small size of marmosets, their immune response to infection that is comparable to humans, and the ability to house more statistically relevant numbers within high containment, makes the marmoset an appropriate animal model for biodefense-related pathogens.
|
What is the conclusion of the study?
| 5,230 |
the marmoset an appropriate animal model for biodefense-related pathogens
| 19,511 |
1,587 |
Exploring the Innate Immunological Response of an Alternative Nonhuman Primate Model of Infectious Disease; the Common Marmoset
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4129158/
SHA: f4c43e4ae49ca69dbac32620bd0a73ecbb683b91
Authors: Nelson, M.; Loveday, M.
Date: 2014-07-22
DOI: 10.1155/2014/913632
License: cc-by
Abstract: The common marmoset (Callithrix jacchus) is increasingly being utilised as a nonhuman primate model for human disease, ranging from autoimmune to infectious disease. In order to fully exploit these models, meaningful comparison to the human host response is necessary. Commercially available reagents, primarily targeted to human cells, were utilised to assess the phenotype and activation status of key immune cell types and cytokines in naive and infected animals. Single cell suspensions of blood, spleen, and lung were examined. Generally, the phenotype of cells was comparable between humans and marmosets, with approximately 63% of all lymphocytes in the blood of marmosets being T cells, 25% B-cells, and 12% NK cells. The percentage of neutrophils in marmoset blood were more similar to human values than mouse values. Comparison of the activation status of cells following experimental systemic or inhalational infection exhibited different trends in different tissues, most obvious in cell types active in the innate immune response. This work significantly enhances the ability to understand the immune response in these animals and fortifies their use as models of infectious disease.
Text: The common marmoset (Callithrix jacchus), a New World monkey (NWM) species is a small, arboreal nonhuman primate (NHP), native to the Atlantic Coastal Forest in Northeast Brazil and parts of South East Brazil. In recent years the common marmoset has become more widely used in applied biomedical research, and an increasing body of evidence suggests the physiological and immunological responses to biological insults are similar between marmosets and humans [1] . In the field of infectious disease, the marmoset is primarily being investigated as an alternative NHP model to complement the more traditionally used Old World monkeys (OWM) (e.g., rhesus and cynomolgus macaques). Evolutionarily, both NWM and OWM sit within the simiiformes infraorder of the suborder Haplorhini of primates [2] . Marmosets sit within the family Callitrichidae of the Platyrrhini parvorder, while OWM sit within the Cercopithecidae family of the Catarrhini Parvorder. Marmosets therefore are separated from Old World monkeys by one ancestral step and are a lower order primate.
Marmosets have been used to model the infection syndrome caused by a number of public health pathogens including Lassa virus [3] , Hepatitis C virus [4] , Dengue virus [5] , Herpesvirus [6] , Junin virus [7] Rift Valley Fever [8] , and SARS [9] . Marmosets have also been used to model a number of biodefense pathogens including Eastern Equine Encephalitis virus [10] , Bacillus anthracis [11] , Francisella tularensis [12, 13] , Burkholderia pseudomallei [14] , Marburg haemorrhagic fever virus [15, 16] , Ebola haemorrhagic fever virus [16] , and Variola virus [17] . The utility of marmosets to assess medical countermeasures has also been demonstrated; a vaccine has been tested for Lassa fever [18] and the efficacy of ciprofloxacin and levofloxacin has been tested as postexposure therapies for anthrax and tularemia, respectively [19, 20] .
In order to exploit these models fully and to allow meaningful comparison with the human condition, the response of the immune system to infection/therapy needs to be 2 Journal of Immunology Research characterised and understood. Generally, NHPs have a close molecular, immunological, reproductive, and neurological similarity with humans making them ideal surrogates for humans and the study of infectious diseases. There is a high level of gene homology between humans and NHPs which underlies physiological and biochemical similarities. Similarities at the genetic level extend to the phenotypical level making NHPs well suited to modelling pathophysiological responses in man [21] . Immunologically, there is a high degree of homology between humans and marmosets [22] . The similarity of various immunological factors produced by humans and marmosets has been investigated at both the genetic and protein levels. There is at least 95% homology between human costimulatory molecules (e.g., CD80, CD86 etc.) and those of marmosets [23] . Also the immunoglobulin and T-cell receptor repertoire of humans and marmosets show at least 80% homology [24, 25] .
Currently, the availability of commercial reagents specifically designed for the marmoset is limited although a number of antibodies designed for use with human samples have been shown to cross-react with leucocytes from marmoset blood [26] [27] [28] . However, these reagents have not been exploited to investigate the immune response to infectious disease. To date, investigation of the immune response in marmosets has primarily been achieved using pathogen-specific antibodies to determine the serological response using ELISA such as in the smallpox, Dengue, Rift Valley Fever, and Herpes models [5, 6, 8, 17] or by immunohistochemistry to identify, for example, CD8+, CD3+, CD20+ cells, and IL-6 in the smallpox model [17] ; neutrophils and macrophages in the Herpes model [6] ; or CD3+ and CD20+ cells in the Lassa model [3] .
The work presented here focuses on understanding the immune profile of the naive marmoset as well as identifying and quantifying the immune response to infectious disease. The aim of this work is to determine key changes and identify correlates of infection or protection.
Healthy sexually mature common marmosets (C. jacchus) were obtained from the Dstl Porton Down breeding colony and housed in vasectomized male and female pairs. The Dstl colony was established during the 1970s and is a closed colony with a stable genotype. Animals included in these studies were mixed sex pairs, between 18 months and 5 years old and weighing between 320 g to 500 g. All animals were allowed free access to food and water as well as environmental enrichment. All animal studies were carried out in accordance with the UK Animals (Scientific Procedures) Act of 1986 and the Codes of Practice for the Housing and Care of Animals used in Scientific Procedures 1989. Animals were challenged with an intracellular pathogen by either the subcutaneous or inhalational route and were humanely killed at various time points after challenge. Prior to the infection study, animals were bled to determine baseline immunological parameters. Studies were performed to establish infection models in order to evaluate the efficacy of suitable therapies for transition ultimately to the clinic.
Populations. Blood and tissue samples were homogenised to provide single cell suspensions [12] . Red blood cells were lysed, and the mixed leucocyte population was washed and stained with various combinations of the following fluorescent antibody stains: CD3 (SP34-2), CD8 (LT8), CD11c (SHCL3), CD14 (M5E2), CD16 (3G8), CD20 (Bly1), CD45RA (5H9), CD54 (HCD54), CD56 (B159), CD69 (FN50), CD163 (GHI/61), and MCHII (L243) (BD Bioscience, Insight Bioscience, AbD serotec). Samples were fixed in 4% paraformaldehyde for 48 hrs at 4 ∘ C and analysed by flow cytometry (FACScanto II BD) within 72 hours of staining.
Levels of circulating cytokines and chemokines were also quantified in the blood of marmosets from the Dstl colony using human multiplex kits available commercially (BD cytokine flex beads and the Luminex system). These systems show significant cross-reactivity with the marmoset suggesting a high degree of conservation between the two species for IL-6, MIP-1 , MIP-1 , and MCP-1 [29] . However, for other cytokines that are pivotal in the innate response, TNF and IFN reagents were obtained from U-CyTech Biosciences and Mabtech AB, respectively, due to a lack of cross-reactivity observed within the kit obtained from BD [13] .
In order to fully characterise the immune response to infectious agent in the marmoset, single cell suspensions of lung and spleen tissue were also examined in conjunction with the traditionally used blood cells. These tissue homogenates are of particular interest in relation to target sites of infection: the lung as the site of initial infection following an inhalational challenge and the spleen as a representative organ following a parental challenge. Cell types targeted during this analysis include cells important in the innate response (e.g., neutrophils, macrophages, and NK cells) and the adaptive response (T and B cells) with a view to determine the response to infection and vaccination and to derive immune correlates of infection/protection. Dapi was included as a nuclear marker to ensure that the initial gating included only intact cells. Basic cell types in blood were easily identified by measuring size (forward) and granularity (side) scatter (Figure 1(a) ). Identification of cell types in tissue samples was more difficult as the scatter profiles are less clearly compartmentalized. The common leukocyte antigen (CD45) normally used to locate all leukocytes in human samples also worked well in marmoset blood but failed to provide relevant information in the tissue samples. Confirmation of neutrophil identification was done by nuclear morphology and macrophages were identified by their adherent nature in initial experiments (data not shown). Neutrophils were stained as CD11c dim CD14− and macrophages as CD11c + CD14+ regardless of tissue origin (Figure 1(b) ). Figure 1 shows the basic division of lymphocytes between T, B, and NK cells from a healthy blood sample.
Using this approach, the percentage of NK cells, B-cells, total T-cells, CD8+ T-cells, neutrophils, and monocytes was determined in the blood of naive marmosets (Figure 2 (a), Table 1 ); approximately 63% of all lymphocytes were T cells, 25% B cells, and 12% NK cells. The variability of the data is depicted in Figure 2 (a) with the greatest variability observed in the proportion of neutrophils. There were no obvious differences attributable to age or sex of the animals. This analysis was also applied to lung and spleen homogenates from naive marmosets (Figures 2(b) and 2(c) ). Greater variability was observed in the data relating to the identification of cell types in tissue samples, attributed to the inherent difficulties in identifying cell types in tissue homogenates by size and granularity and also the smaller cohort of animals. As expected, low numbers of neutrophils are found in naive spleen or lung tissue (8% both). Healthy mouse spleens typically have approximately 1-2% granulocytes [30] .
Understandably, there are few reports on the typical cell percentages expected in healthy human individuals for these tissues. However, it is reported that B cells are more prevalent in the spleens of humans at a ratio of 5 to 4 B to T cells than in the lungs which have a ratio of 1 to 8 B to T cells [34] . In marmoset data reported here, a ratio of 2 to 3 B to T cells in the spleen and 1 to 6 B to T-cells in the lungs was observed compared to a ratio of 3 to 2 B to T cells in mouse spleens [30] .
Upon comparison, the marmoset data is generally consistent with previously reported data which is only available for marmoset blood samples [27] and information available for human blood [32, 33] (Table 1 ). However, one report found the proportion of CD8+ T-cells was almost three times greater in marmosets than humans, 61% to 21% respectively [35] compared to the 30% observed in this study and the work previously reported by Brok et al. [27] . Brok's study involved a small number of animals (eight) and also used a different CD8+ clone to identify cells. Contrastingly, in mice, differences are observed in the proportion of both B cells and neutrophils [31] , although these differences are highly strain specific. C57BL/6J mice are reported to have 67% B cells and BALB/C mice 46%; both of which are consistently higher than the percentage found in marmosets and humans of approximately 25% (Table 1 ) [27, 31] . The proportion of neutrophils found in the blood of C57BL/6J mice at 13% is lower than the 35% found in marmosets and the 40-75% expected for healthy human blood. This is encouraging as neutrophils play a pivotal role in the innate response to infection [36] . A cross-species comparison suggests that monocytes comprise 3% of leukocytes ( Table 1) . Levels of circulating cytokines and chemokines (IL-6, IL-1 , MIP-1 , MCP-1, Rantes, TNF , and IFN ) were also quantified in the blood, lung, and spleen of naïve marmosets from the Dstl colony. None of these cytokines were detected in blood samples from uninfected animals; however low levels of MIP-1 , MCP-1, and Rantes were found in spleen and lung tissue.
Preliminary investigation of the immune response has supported the development of marmoset model of infection at Dstl. The levels of different cell types were measured at specific times after challenge with inhalational F. tularensis, B. pseudomallei, and Marburg virus [13] [14] [15] . Following challenge with F. tularensis, increasing levels of NK cells, neutrophils, T cells, and macrophages were observed, peaking at 48 hours after challenge before rapidly declining. This study also demonstrated the importance of investigating the immunological response in key target organs, as an increase in CD8+ T cells and T cells was observed in the spleen and lungs but not in the blood. Increasing levels of various cytokines, MCP-1, MIP-1 , MIP-1 , IL-6, and IL-1 , were observed in Table 1 : Comparison of the percentages of different cell types observed in the blood from healthy marmosets, mice, and humans.
Identification markers Marmoset (present data)
Marmoset [27] Mouse 4 [30, 31] Human Asian [32] Human Caucasian [33] Number the lungs, spleen, and blood as the disease progressed (TNF and IFN were not measured in this study). Following inhalational challenge of marmosets with B. pseudomallei, an increase in the number of neutrophils was observed in the blood at 36 hours after challenge, followed by a rapid decline that was associated with an influx of neutrophils into the lung at 46 hours after challenge. A subsequent decline in the number of neutrophils in the lung was associated with the increased number in the spleen of animals that exhibited severe disease and were humanely killed. There was a gradual increase in the number of macrophages in the spleen as the disease progressed with numbers of macrophages peaking in the blood and lungs at 36 hours after challenge. A rapid decline in the number of macrophages in the lungs and blood was observed by 46 hours after challenge.
The levels of various cell types and cytokines were also measured in the blood of animals following inhalational challenge with Marburg virus [15] . In these animals a general increase in the numbers of T cells, NK cells, macrophages IFN-, IL-1 , and MCP-1 was observed with time (TNF was not measured).
In order to gain more information from these acute bacterial infection models, we have sought out other markers from the literature. Primarily this was from marmoset models of autoimmune disorders such as rheumatoid arthritis and multiple sclerosis where the cross-reactivity of human antibodies was investigated, as well as the functionality of cells [37] [38] [39] [40] . More recent work at Dstl has reported further cross-reactivity between marmoset cells and human cytokines to induce activity in marmoset T cells [36, 41] . These studies, combined with increasing information available on the cross-reactivity of human antibodies to various NHPs (e.g., NIH NHP reagent resource, http://www.nhpreagents.org/NHP/default.aspx), has expanded the ability to assess activation markers for disease. Detection of the following cell surface markers with human antibodies was trialed: CD54 (ICAM-1) associated with cellular adhesion, inflammation, and leukocyte extravasation; CD69 the early activation marker; CD16 as a macrophage activation marker; CD163 the alternative macrophage activation marker; and MHC class II (HLA-DR). CD56 was originally included to identify NK cells; however, it was noted that its expression on T cells was upregulated during disease and that cells defined as CD3+ CD16− CD56+ have been shown to be functionally cytotoxic in marmosets [37, 42] .
These markers have been used to expand on our previously published work to determine changes in the activation status of basic cell types in response to an acute bacterial infection. Animals were challenged with bacteria at a comparable dose either by inhalation ( = 22) or by a systemic route ( = 12) and humanely killed once they had reached a humane endpoint (between day 4 and day 5 after challenge). Figure 3 illustrates the cellular activity in representative tissues following inhalational (Figures 3(b) and 3(e)) or systemic challenge (Figures 3(c) and 3(f)) and in naïve samples (Figures 3(a) and 3(d) ). Naïve T and NK cells appear to have similar resting activation states regardless of origin, whereas neutrophils and macrophages have differential expression of activation, for example, CD16. In response to disease, the proportions of the cell types appear to remain relativity constant; however, the activation markers provide more detailed information and show involvement of all the cell types explored. Extensive activation was to be expected considering that the samples were taken at the humane endpoint. There is also extensive variation between The response to infection within the lungs has similarities across disease routes in terms of neutrophil reduced expression of CD16 and CD54 and macrophage increased expression of CD16 and reduction in MHCII. Unexpectedly, the T and NK cells appear to be more actively involved in systemic disease, indicating that the disease develops a pneumonic element regardless of initial route of infection.
Levels of circulating cytokines and chemokines (IL-6, IL-1 , MIP-1 , MCP-1, Rantes, TNF , and IFN ) were also quantified in the lung and spleen samples. All of the cytokines (with the exception of Rantes) were expressed at high levels (ng/mg) in all samples, which was expected as the animals had succumbed to terminal disease.
The work presented here adds significant relevant information to the marmoset models of infection and to the understanding of the immune response in these animals. This work extends marmoset immunology from autoimmune disorders into the field of infectious diseases; this coupled with an increase in the information available on crossreactivity of human reagents to a variety of NHPs increases the utility/application of marmosets as models of human disease. In conclusion, the immune response in marmosets to infectious disease can be characterised in terms of the phenotype and activation status of all the major immune cells and key cytokine and chemokine expression. This can aid in the identification of correlates of infection or protection in medical countermeasures assessment studies. This information can also potentially be used for pivotal studies to support licensure of products under the FDA Animal Rule.
This, in conjunction with the small size of marmosets, their immune response to infection that is comparable to humans, and the ability to house more statistically relevant numbers within high containment, makes the marmoset an appropriate animal model for biodefense-related pathogens.
|
Why makes the marmoset an appropriate animal model for pathogen research?
| 5,231 |
the small size of marmosets, their immune response to infection that is comparable to humans, and the ability to house more statistically relevant numbers within high containment
| 19,325 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
How can the efficacy of DAAs be diminished?
| 3,895 |
the presence of resistance-associated substitutions
| 655 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
Was is the response rate of the Hepatitis C virus to direct-acting antiviral treatments?
| 3,896 |
up to 98%
| 2,043 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
How do nonnucleoside NS5B polymerase inhibitors work?
| 3,897 |
inhibit polymerase activity by allosteric mechanisms
| 2,755 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
How many patients were studied?
| 3,898 |
31
| 4,443 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
Was written consent obtained?
| 3,899 |
was obtained
| 4,871 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
How much of the RNA template was in the reverse transcription reaction mixture?
| 3,900 |
5 μl
| 5,473 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
How many RASs to NS5A inhibitors were identified?
| 3,901 |
2 strains out of 25 (8%)
| 11,338 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
Why is the substitution E62D important in drug resistance?
| 3,902 |
confers a higher level of resistance than the one achieved by the RAS alone
| 11,896 |
1,592 |
Pretreatment Hepatitis C Virus NS5A/NS5B Resistance-Associated Substitutions in Genotype 1 Uruguayan Infected Patients
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6112080/
SHA: f01ad3545245b4f884b48aa2b69c9deb942c3e77
Authors: Aldunate, Fabián; Echeverría, Natalia; Chiodi, Daniela; López, Pablo; Sánchez-Cicerón, Adriana; Fajardo, Alvaro; Soñora, Martín; Cristina, Juan; Hernández, Nelia; Moreno, Pilar
Date: 2018-08-14
DOI: 10.1155/2018/2514901
License: cc-by
Abstract: Hepatitis C Virus (HCV) infection treatment has dramatically changed with the advent of direct-acting antiviral agents (DAAs). However, the efficacy of DAAs can be attenuated by the presence of resistance-associated substitutions (RASs) before and after treatment. Indeed, RASs detected in DAA treatment-naïve HCV-infected patients could be useful for clinical management and outcome prediction. Although the frequency of naturally occurring HCV NS5A and NS5B RASs has been addressed in many countries, there are only a few reports on their prevalence in the South American region. The aim of this study was to investigate the presence of RASs to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients infected with chronic hepatitis C and compare them with reports from other South American countries. Here, we found that naturally occurring substitutions conferring resistance to NS5A and NS5B inhibitors were present in 8% and 19.2%, respectively, of treatment-naïve HCV genotype 1 infected patients. Importantly, the baseline substitutions in NS5A and NS5B herein identified differ from the studies previously reported in Brazil. Furthermore, Uruguayan strains subtype 1a clustered within all major world clades, showing that HCV variants currently circulating in this country are characterized by a remarkable genetic diversity.
Text: Hepatitis C Virus (HCV) infection treatment has dramatically improved thanks to the introduction of direct-acting antiviral agents (DAAs). These antivirals have significantly increased response rates (up to 98%) and greatly reduced treatment duration [1] . Currently available DAAs are classified into four categories given their molecular targets in the HCV replication cycle: (1) NS3/4A protease inhibitors (PIs) bind to the active site of the NS3/4A protease; (2) NS5A inhibitors interact with domain 1 of the NS5A dimer, although the exact mechanism of NS5A inhibition remains to be fully elucidated; (3) nucleos(t)ide analog NS5B polymerase inhibitors are incorporated into the nascent RNA chain resulting in chain termination by compromising the binding of the incoming nucleotide; (4) nonnucleoside NS5B polymerase inhibitors interact with either the thumb 1, thumb 2, palm 1, or palm 2 domain of NS5B and inhibit polymerase activity by allosteric mechanisms [2] [3] [4] . However, the extreme mutation and high replication rates of HCV, together with the immune system pressure, lead to a remarkable genetic variability that can compromise the high response rates to DAAs due to the preexistence of resistanceassociated substitutions (RASs) [5, 6] .
Each drug or class of DAA is characterized by specific resistance profiles. The likelihood that a DAA will select for and allow outgrowth of viral populations carrying RASs depends on the DAA's genetic barrier to resistance (the number and type of mutations needed to generate an amino acid substitution that confers resistance), the viral fitness (replicative capacity) of the resistant variant, and viral genotypes and subtypes [7, 8] .
The prevalence of RASs in treatment-naïve patients has been broadly reported worldwide [9] [10] [11] [12] [13] [14] [15] [16] . However, apart from Brazil and Argentina, this issue has not been fully addressed in other South American countries yet [9, [17] [18] [19] . The lack of information in relation to preexisting baseline RASs, added to the high cost of these new drugs, are the major limiting factors for the broad implementation of these new therapies in Uruguay as well as in other Latin American countries (low-or lower-middle income) [20] .
In this study, we explored the presence of resistance variants to NS5A and NS5B inhibitors in a DAA treatment naïve cohort of Uruguayan patients chronically infected with hepatitis C. Here, we aimed to contribute to the knowledge of the circulation of HCV resistant variants in the South American region.
Samples. Serum samples were obtained from 31 patients with serological markers for HCV, which were recruited between 2015 and 2017 at the Gastroenterology Clinic from Hospital de Clínicas, Montevideo, Uruguay. HCV infection was confirmed by Abbott realtime HCV (Abbott Molecular Inc., Des Plaines, USA). Patients selected for this study were both chronically infected with HCV genotype 1 and DAA treatment-naïve at the time of blood extraction. Written informed consent was obtained from all patients. The studies have been performed according to the World Medical Association Declaration of Helsinki and approved by the appropriate institutional board (Hospital de Clínicas ethical committee).
2.2. RNA Extraction, cDNA Synthesis, and NS5A and NS5B Amplification. Viral RNA was extracted from 140 μl of serum using the QIAamp Viral RNA mini kit (QIAgen, Hilden, Germany) according to the manufacturer's protocol. The viral RNA was heated at 65°C for 5 min and used as a template for a reverse transcription reaction. The reverse transcription reaction mixture contained 5 μl of the RNA template, 1 μl of random hexamer 100 ng/μl (Invitrogen Life Technologies, Carlsbad, CA, USA), 1 μl of dNTP mix (10 mM each), 4 μl of 5X first-strand buffer, 2 μl of 0.1 M DTT, 1 μl of SuperScript II reverse transcriptase (200 U/μl) (Invitrogen Life Technologies, Carlsbad, CA, USA), and 1 μl (40 U/μl) RNaseOUT (Invitrogen Life Technologies, Carlsbad, CA, USA). The reverse transcription was performed at 42°C for 50 min, and then the reverse transcriptase enzyme was inactivated at 70°C for 15 min. PCR amplification of NS5A and NS5B genome regions was performed using primers and conditions previously described [10] . Amplicons were purified using the Illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Science, Buckinghamshire, UK) according to the manufacturer's protocol.
2.3. NS5A and NS5B Sequencing. The purified product was then sequenced using the same sets of primers used for PCR amplification. Bidirectional Sanger sequencing was performed by Macrogen Korea (http://www.macrogen.com).
2.4. NS5A and NS5B Genotype Determination. HCV NS5A and NS5B consensus sequences obtained from Uruguayan patients were aligned with sequences from HCV representing all genotypes and main subtypes isolated in different geographic regions of the world. These sequences were obtained from Los Alamos HCV sequence database and from the NIAID Virus Pathogen Database and Analysis Resource (ViPR) [21, 22] . For strains included in these studies, see Supplementary Material Table S1 . Sequences were aligned using the CLUSTAL W software [23] . Once aligned, the best evolutionary model that described our sequence data was assessed using ModelGenerator program [24] . Using the GTR + G + I model (General time reversible + gamma + invariant sites), maximum likelihood phylogenetic trees were constructed for both NS5A and NS5B using the MEGA 5.0 software [25] . For NS5A, 953 nucleotides (positions 6367 to 7319, relative to HCV 1a reference strain, H77 NC_004102) were included in the phylogenetic analysis, whereas for NS5B, only 361 nucleotides corresponding to the Okamoto region (positions 8265 to 8625, relative to strain H77 NC_004102) were included. As a measure of the robustness of each node, we employed the bootstrapping method (1000 pseudoreplicates).
For NS5A 1a Uruguayan sequences (n = 20), a second alignment and maximum likelihood phylogenetic tree was generated in order to analyze HCV evolutionary relationships between Uruguayan, Brazilian, and worldwide strains. For non-Uruguayan strains included in this analysis, see Supplementary Material Table S2. 2.5. NS5A and NS5B Sequence Analysis. In order to properly identify substitution changes in NS5A and NS5B regions from HCV strains circulating in Uruguayan patients, we generated world consensus sequences for 1a and 1b subtypes using a wide range of NS5A and NS5B sequences from HCV strains isolated worldwide. For this purpose, NS5A gene sequences corresponding to subtypes 1a (n = 160) and 1b (n = 88) were retrieved from Los Alamos HCV sequence database and from the NIAID ViPR [21, 22] . Likewise, datasets of 150 and 124 NS5B sequences were generated for subtypes 1a and 1b, respectively. Using Seqman program, implemented in DNAStar 5.01 package (DNASTAR, Madison, USA), a world consensus nucleotide sequences were generated for each gene and subtype. Each Uruguayan sequence was subsequently aligned to the corresponding reference sequences, and then in silico translated. The amino acid sequences obtained were compared in order to explore the presence of RASs as well as the presence of polymorphisms at a RAS position (RAPs) in Uruguayan HCV strains. RAPs are defined as any change from reference sequence for a specific genotype at a position associated with NS5A resistance [26] .
To study the genetic variability of NS5A and NS5B regions of HCV strains circulating in Uruguayan patients, sequences of these regions (accession numbers MH070029-MH070090) were aligned with corresponding sequences from 59 HCV strains isolated elsewhere, representing all genotypes and main subtypes (for strains included in these analyses, see Supplementary Material Table S1 ). Therefore, maximum likelihood phylogenetic trees were constructed. The results of these studies are shown in Figure 1 All strains in the phylogenies were assigned according to their genotype, and each cluster was supported by very high bootstrap values for both analyzed regions. Strains isolated from Uruguayan patients (n = 31) were assigned to genotype 1, 20 of which corresponded to subtype 1a and 11 to subtype 1b. The results of NS5A (Figure 1 (a)) and NS5B (Figure 1
Genotype 1b phylogenetic analyses were concordant for both genomic regions in all 31 sequences, suggesting no recombination events between these regions.
To further analyze the evolutionary relationships between the Uruguayan strains and those circulating in Brazil and elsewhere, a second maximum likelihood phylogenetic tree of HCV-1a sequences of NS5A partial region was built ( Figure 2 ). As was previously described, two distinct 1a clades (clades 1 and 2) were observed. Brazilian sequences clustered in a large group of related sequences inside clade 1 [9] . Whereas NS5A Uruguayan strains (in red) did not cluster in a particular clade, rather, they grouped dispersedly within all major world clades.
With the purpose of studying the amino acid (AA) substitutions along the NS5A protein, Uruguayan HCV AA sequences were aligned with NS5A world consensus sequences (residues 23 to 354 relative to NS5A protein sequence). AA substitutions at positions previously found to be potentially associated with resistance to NS5A inhibitors, as well as polymorphisms at a RAS position, were identified. These results are summarized in Table 1 .
RASs to NS5A inhibitors (L31M and L31V) were identified in 2 strains out of 25 (8%) fully sequenced samples. RAPs were found in 3 strains (subtype 1a): 2 exhibited the substitution H58P and 1 the substitution K24Q. Although these substitutions were not reported as resistant, some changes at these positions were previously described as RASs in subtype 1a, namely H58D and K24R [27, 28] . Finally, substitution E62D was found in one subtype 1a strain. This change is considered as a secondary substitution because, although it does not confer resistance by itself, when combined with a known RAS it does. In fact, it confers a higher level of resistance than the one achieved by the RAS alone [26] . In addition, several polymorphisms that have not been previously reported to be associated with a resistant phenotype were also detected (see Supplementary Material Table S3 ).
In order to study substitutions along NS5B protein, Uruguayan HCV AA sequences were aligned to the NS5B world consensus sequences. Almost full-length AA sequences were obtained in 26 out of 31 analyzed strains. 23 sequences span residues 36 to 539 whereas the remaining 3 span residues 36 to 557 of NS5B protein.
This issue limited our studies, since many of the described RASs are observed as of residue 553.
Importantly, RASs to NS5B inhibitors ( Table 2) were observed in 5 strains out of 26 sequenced samples (19.2%). C451R was found in two isolates while A421V was found in only one. In 2 of the 3 strains for which we were able to obtain longer sequences, RASs S556G (subtype 1a) and Q556R (subtype 1b) were observed.
Finally, we found two RAPs: A421V (in 2 subtype 1b strains) and A553G (in 1 subtype 1a strain). Although A421V has been associated with resistance to beclabuvir (BCV) in patients infected with HCV subtype 1a, this resistant phenotype has not been proven in strains subtype 1b [29] . In position 553, the substitution reported as resistant was A553T [8] .
As was the case for NS5A, different polymorphisms not previously associated with a resistant phenotype were also detected in NS5B (see Supplementary Material Table S4 ).
The advent of DAAs therapies constitutes one of the major breakthroughs in HCV infected patients management. However, these new treatment options are far from being universally available, in particular for HCV infected patients relying on Latin American public healthcare systems. The main limiting factors for worldwide access to DAAs in our region concern the high cost, the inadequate management of public healthcare systems, the limited access of low-income or uninsured populations to healthcare providers, and the lack of accurate epidemiological information [20, [30] [31] [32] . In Uruguay, these therapies became recently available, and although some have been approved for their use by the public health authorities (Viekira pak and sofosbuvir/ledipasvir therapies), they are not currently financially covered, except in specific cases. Despite the high rates of viral response achieved with DAA-based treatments, still 1 to10% of the patients fails to eliminate infection, and in these cases, baseline and emergent resistance variants turn out to be key factors contributing to treatment failure [5, 17, 33] .
Unfortunately, we are currently unable to properly assess the number of HCV infected people in Uruguay and even more to figure out the frequency and type of RASs circulating. These facts could compromise the effectiveness of these new therapies in our country.
We have previously reported that naturally occurring substitutions conferring resistance to NS3 inhibitors exist in a significant proportion of Uruguayan patients infected with HCV genotype 1, and we showed that this frequency seemed to be higher than in other South American countries (Brazil and Argentina) [34] . The present study describes the prevalence of baseline NS5A and NS5B RASs in HCV genotype 1 infected DAA-naïve patients in a Uruguayan cohort. The presence of substitutions conferring resistance to NS5A inhibitors has been widely reported both in therapynaïve and in relapser patients from Europe [10, 33, [35] [36] [37] [38] , USA [37, 39, 40] , and Asia [41] [42] [43] . However, NS5A sequences from South America are poorly analyzed yet [9, 44] . Recent studies have revealed that the mean prevalence of NS5A genotype 1 baseline RASs to different inhibitors ranges from 6% to 16% using population sequencing or deep sequencing [27, 37, 45, 46] . Importantly, the prevalence and type of baseline NS5A RASs varies slightly by geographic regions. For instance, L31M was found in 2.2% of genotype 1a infected patients in Europe, in 4.1% of those in Oceania, and strikingly in no patient from the USA [27] . For this reason, we believe that there is a need to contribute data from our region, for which we still do not have enough information, apart from Brazil [9, 44] . The results of this study indicate the presence of DAA NS5A RASs in 2 HCV strains (8% of the patients enrolled in this study), with baseline RASs detected at position 31 (see Table 1 ). L31M substitution confers resistance to daclatasvir (DCV), ledipasvir (LDV), and elbasvir (EBV) in both 1a and 1b subtypes [5, 6, 8, 28, 47, 48] , whereas substitution L31V does it to DCV in subtypes 1a and 1b, to LDV in subtype 1b, and to EBV in subtype 1a [5, 6, 28] . Given that both L31V and L31M are clinically relevant RASs, their detection at baseline may influence the choice of first-line treatment regimens [28] .
The substitutions H58P and K24Q found in two patients are considered as resistance-associated polymorphisms (RAPs). The RASs characterized at these positions were H58D and K24G/N/R [5, 6, 27, 28, 49, 50] . The substitution H58P was found as a baseline RAP in relapsers to LDV (HARVONI prescription, https://www.gilead.com/-/ media/files/pdfs/medicines/liver-disease/harvoni/harvoni_pi. pdf?la=en). However, it is sometimes regarded as a RAS [10, 51] , despite conferring only 1.2 fold change in resistance in in vitro studies using the 1a replicon system [39] .
We did not find M28T/V, Q30R/H, or Y93H substitutions as there were previously reported in Brazil and worldwide [9, 27, 44] . The amino acid substitution E62H was found in one Uruguayan patient. Although this change does not confer resistance by itself but in combination with Q30R, it generates a high resistance level to DCV [52] .
The presence of baseline NS5A RASs impacts treatment outcome in some patient groups by affecting SVR rates. The detection of NS5A preexistent RASs may play a relevant role in the choice of first-line treatment regimens or in the simplification/shortening of recommended regimens, in order to bring SVR rates close to the highest achievable [27, 38, 41, 53] , in particular in countries such as Uruguay, where only two different DAA-containing treatment regimens are approved for their use.
Regarding NS5B gene, global analysis (with the exception of South America [17, 19] ) revealed that NS5B DAA resistance substitutions are infrequent [14] . Our study showed the presence of NS5B inhibitors RASs in 5 out of 26 analyzed HCV infected Uruguayan patients naïve to treatment (19.2%). Substitutions found in this work were A421V and S556G associated in subtype 1a with resistance to BCV and dasabuvir (DSV), respectively [8, 28, 29, 54, 55] , and Q556R associated with resistance to DSV both in genotype 1a and 1b [12, 28] . Substitution C451R, observed in two Uruguayan patients, was reported previously in patients who failed to clear the infection after treatment with OBV/PTV/r + DSV ± RBV. In these cases, it appeared in combination with G558R (Trial Coral I-Cohort 2: http:// www.hcv-trials.com/showStudy.asp?Study=86). RAPs in positions 421 and 553 (A421V in two subtype 1b isolates and A553G in one subtype 1b isolate) were also found. Although A421V has been associated with resistance to BCV in patients with subtype 1a, this phenotype has not been proven in strains of subtype 1b [29] . In position 553, the substitutions reported as resistant are A553T in subtype 1a [8] and A553V in subtype 1b [54] , conferring resistance to DSV.
In contrast to our results, Noble and coworkers (2016) reported the presence of V321A, A421G, M414V, Y448H, L159F, and C316N in Brazilian isolates [17] , yet none of these mutations were found in this study, probably due to the diversity found between Uruguayan and Brazilian strains ( Figure 2 ). Nevertheless, substitution A421V was found in Brazil [17] , Argentina [19] , and Uruguay. The RAS S282T was detected neither in Brazilian reports nor in this current work (Uruguay) [17, 18, 56] . Our findings further confirm and complement previous studies which evidenced a low prevalence of this substitution in vivo, probably due to its low replicative fitness [14, 18, 57] . Despite our results, it is worth mentioning that the presence of baseline NS5B RASs conferring resistance to nucleotide or nonnucleoside NS5B inhibitors has not been shown to have any impact on virologic responses thus far [53, 58] .
These results show both diversity in the baseline polymorphisms found in different Latin American countries and in the evolutionary relationships of Uruguayan isolates ( Figure 2 ). This fact could be linked not only to the isolates' geographic region and viral intrinsic characteristics but also to the genetic background of the host. It is worth mentioning that we live in a vast continent inhabited by populations with different genotypic characteristics that might, depending on the situation, require different approaches to treatment. Indeed, we have recently found that allele and genotype frequencies at IL28B locus of Uruguayan individuals closely resemble those of an admixed population rather than a uniformly European-descendant one [59] . Altogether, we believe that it could be important to carry out studies throughout the South American region in order to establish the prevalence of RASs in NS5A and NS5B in different countries. In fact, this will aid in understanding that not every treatment regimen might be adequate for every patient and country. The data we presented here might guide not only physicians in making therapeutic decisions but also public health authorities in approving more diverse treatment combinations. These treatment formulations would cover most of the circulating strains in our region, a region with an extremely diverse genetic background population.
To our knowledge, the present study revealed for the first time the presence of RASs in the NS5A and NS5B regions of HCV genotype 1 Uruguayan strains from patients who have not been previously treated with DAAs and is one of the few South American countries to report on this matter. It is currently unclear if preexisting viral variants with reduced susceptibility to DAAs are clinically relevant for the prediction of virologic treatment failure. However, individualized DAA therapy based on baseline resistance analysis may be beneficial for optimizing treatment efficacy in patients with HCV genotype 1 infection and risk factors for treatment failure. Therefore, the potential role of baseline resistance testing remains an area of critical research and clinical questions.
The data used to support the findings of this study are included within the article.
The authors declare that they have no conflicts of interest.
Fabián Aldunate and Natalia Echeverría contributed equally to this work.
Supplementary Material Table S1 : hepatitis C Virus NS5A and NS5B sequences used as representatives of each genotype to perform the phylogenetic analysis. Their corresponding genotype, country of isolation, and GenBank accession number are indicated. Supplementary Material Table S2 : hepatitis C Virus NS5A subtype 1a sequences used to reveal evolutionary relationships between Uruguayan strains and others isolated elsewhere. Their corresponding country of isolation and GenBank accession number are indicated. Supplementary Material Table S3 : amino acid substitutions in NS5A protein not previously associated with resistance to NS5A inhibitors. Supplementary Material Table S4 : amino acid substitutions in NS5B protein not previously associated with resistance to polymerase inhibitors. (Supplementary Materials)
|
What are the key factors preventing the elimination of HCV infection in some patients?
| 3,903 |
baseline and emergent resistance variants
| 14,415 |
1,598 |
Which Kind of Provider’s Operation Volumes Matters? Associations between CABG Surgical Site Infection Risk and Hospital and Surgeon Operation Volumes among Medical Centers in Taiwan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4459823/
SHA: f3cbc0503581249a834895fc94cd3bae24714a0d
Authors: Yu, Tsung-Hsien; Tung, Yu-Chi; Chung, Kuo-Piao
Date: 2015-06-08
DOI: 10.1371/journal.pone.0129178
License: cc-by
Abstract: BACKGROUND: Volume-infection relationships have been examined for high-risk surgical procedures, but the conclusions remain controversial. The inconsistency might be due to inaccurate identification of cases of infection and different methods of categorizing service volumes. This study takes coronary artery bypass graft (CABG) surgical site infections (SSIs) as an example to examine whether a relationship exists between operation volumes and SSIs, when different SSIs case identification, definitions and categorization methods of operation volumes were implemented. METHODS: A population-based cross-sectional multilevel study was conducted. A total of 7,007 patients who received CABG surgery between 2006 and 2008 from19 medical centers in Taiwan were recruited. SSIs associated with CABG surgery were identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9 CM) codes and a Classification and Regression Trees (CART) model. Two definitions of surgeon and hospital operation volumes were used: (1) the cumulative CABG operation volumes within the study period; and (2) the cumulative CABG operation volumes in the previous one year before each CABG surgery. Operation volumes were further treated in three different ways: (1) a continuous variable; (2) a categorical variable based on the quartile; and (3) a data-driven categorical variable based on k-means clustering algorithm. Furthermore, subgroup analysis for comorbidities was also conducted. RESULTS: This study showed that hospital volumes were not significantly associated with SSIs, no matter which definitions or categorization methods of operation volume, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon’s volumes varied. Most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons. CONCLUSION: Surgeon volumes were more important than hospital volumes in exploring the relationship between CABG operation volumes and SSIs in Taiwan. However, the relationships were not robust. Definitions and categorization methods of operation volume and correct identification of SSIs are important issues for future research.
Text: data, which should use hierarchical models, may result in biased estimation of the variation and also lead to incorrect conclusions.
SSIs following coronary artery bypass graft (CABG) procedures place a heavy burden on patients and healthcare systems. The total length of stay and expenditure for patients with SSIs after CABG surgery is significantly longer and higher than those without SSIs. [20, 21] In 2008, the Centers for Medicare & Medicaid of the United States of America implemented the "Never Event" policy, where hospitals would no longer receive higher payments for the additional costs associated with treating patients for certain healthcare-acquired infections, including those related to CABG.
In view of the accuracy of SSIs identification and the heterogeneity of definition and categorization methods, no existing studies have used different infection case identification nor definitions and categorization methods of operation volume simultaneously to explore the relationship between operation volumes and infection. The current study takes CABG SSIs as an example to examine whether a relationship exists between operation volumes and SSIs, given different SSI cases identification, operation volume definitions and categorization methods.
This retrospective and cross-sectional study adopted a multilevel design to examine the relationships between provider volumes and SSIs after adjusting for patient-, surgeon-, and hospital-level covariates.
We used data from the Taiwan National Health Insurance Research Database (NHIRD) from 2005 and 2008. The NHIRD, published by the Taiwan National Health Research Institute, includes all the original claims data and registration files for beneficiaries enrolled under the National Health Insurance (NHI) program. The database covers the 23 million Taiwanese enrollees (approximately 98% of the population) in the NHI program. It is a de-identified secondary database containing patient-level demographic and administrative information; however, treatment items are aggregated and without time-related and clinical information. The data is released for research purposes.
The protocol for the study was approved by the Institutional Review Board of the National Taiwan University Hospital (protocol #201001027R). The dataset we used in this study was secondary data; all information was de-identified by data owners.
In this study, we adopted the ICD-9-CM SSI codes (hereafter referred to as the ICD-9-CM based model) and the Classification and Regression Trees (CART) model, which was developed in our previous work [11] to identify SSI cases. As we mentioned above, the ICD-9-CM SSI codes were the most popular tool to identify the SSI cases in claims data. In the ICD-9-CM based model, SSI cases were divided into two categories: index hospitalization events and post-discharge events (i.e., SSIs that occurred within 1 year after discharge and required readmission to a hospital and/ or the use of ambulatory services). Following Wu et al [13] , this study adopted the secondary ICD-9-CM diagnosis codes for index hospitalization events (ICD-9-CM code: 996.03, 996.61, 996.72, and 998.5), and the primary and secondary diagnosis codes for post-discharge events (ICD-9-CM code: 038.0-038. 4 ) as the criteria for SSI identification, in order to avoid cases in which infection existed prior to hospitalization. If a case had an index hospitalization event or a post-discharge event, then he/ she will be identified as SSIs by the ICD-9-CM based model. In the CART model, we adopted the type of antibiotics, dose of cefazolin, length of stay, and number of vessels obstructed (as a proxy indicator of duration of operation) as the parameters to identify the SSIs, according to our previous findings. [11] In our previous work, we used the 2005-2008 National Health Insurance claims data and healthcare-associated infection surveillance data from two medical centers for model development and model verification. Infection cases based on surveillance were identified by infection control personnel if the patient met the Taiwan CDC's criteria, which are the same as those adopted in the U.S. CDC. They manually review medical records of all patients at risk for the specified healthcare-associated infection.
The classification algorithms, the multivariable regression model, and the data mining model were adopted to develop alternative models based on surrogate indicators to identify cases of CABG SSIs and to compare the performance among these models and the ICD-9-CMbased model. For the classification algorithms, researchers build up several criteria, and if a case satisfies (or exceeds) a specific number of criteria, then it will be identified as a case of infection. For the multivariable regression model, researchers usually calculated a risk score by the logistic regression model, and the optimal cutoff point was determined according to the resulting receiver operating characteristic curve.
Concerning the data mining approach, which is widely used for predicting and classifying objects, the characteristics are: automatic discovery of patterns, prediction of likely outcomes, creation of actionable information, and focus on large data sets and databases. The classification and regression tree (CART) model, which is the most popular approach as applied in our work, and the growing, stopping, and pruning of the tree were determined by Gini improvement measures. [22, 23] After referring to the literature and conferring with infectious disease specialists, we adopted the following seven parameters: type of antibiotic, doses of antibiotic, doses of cefazolin, use of second-line antibiotics, length of stay, and number of vessels obstructed. Additionally, cross-validation was also employed, where data from one medical center was used for model development, and another one was used for model validation.
The results of our previous work revealed that the CART model offered better performance than that of the other identification models or the ICD-9-CM based model, especially in the positive predictive value (>70%), which was only found to be 20% in the ICD-9-CM based model. (Table 1 ) The findings also implied that the CART was a decidedly better tool for identifying cases of SSI in the Taiwan National Health Insurance database. Therefore, this study also adopted the CART model for identifying CABG SSIs.
To ensure homogeneity, current study analyzed 7,007 patients from 19 medical centers in Taiwan who underwent CABG surgery (ICD-9-CM procedure codes 36.1x-36.2x) between 2006 and 2008. CABG patients under the age of 18 years or over 85 years were excluded in this study. A total of 302 cases were identified as SSIs by ICD-9-CM based model, and a total of 107 cases were identified as SSIs by CART model.
In this study, we used the following two definitions to define operation volumes: (1) the cumulative operation volumes by each surgeon and hospital within the study period, which was the most common definition in the literature; and (2) following Yasunaga et al.'s study, [24] cumulative operation volumes by each surgeon and hospital in the previous one year for each surgery. However, our data was skewed, which did not follow a normal distribution. Therefore, we conducted the log transformations on operation volumes.
The current work treated operation volumes in three different ways: (1) a continuous variable; (2) a categorical variable based on the first and the third quartile as cutoff points (the most common method to categorize service/ operation volumes) [25] [26] [27] [28] ; and (3) a data-driven categorical variable based on k-means clustering algorithm. This study categorized surgeon and hospital volumes into low, medium, and high volume groups by quartile method and kmeans clustering algorithm.
In the quartile method, the cut-off value (transformed by logarithm) of the first quartile (<25%) for hospital volumes was 5.65, and the third quartile (>75%) was 6.43. In terms of surgeon volumes, the first quartile was 4.38, and the third was 5.35, when we used the cumulative operation volumes within the study period as the definition. While the definition changed, first quartile (<25%) for hospital volumes was 4.66, and the third quartile (>75%) was 5.31. In terms of surgeon volumes, the first quartile was 3.40, and the third was 4.32.
K-means clustering is an unsupervised machine-learning algorithm introduced by MacQueen in 1960s. This method is not only a simple and very reliable method in categorization/ classification, but is also recognized as one of the top 10 algorithms in data mining. [29] This method has often been applied in many fields. [30] [31] [32] Yu and his colleagues even applied it to define the quality of CABG care, and to explore the relationship among patient's income status, the level of quality of care, and inpatient mortality. [33] The main idea of this method is to partition observed data points into k non-overlapping clusters by minimizing the within-group sum of squares. Each point is assigned to the mean of its cluster using the Euclidian distance. Firstly, k cluster centers were randomly generated. Previous studies usually divided surgeons and hospitals into low-, medium-, and high-volume groups; therefore, we also predetermined the surgeon and hospital service volumes into 3 groups (k = 3). Then, participants were assigned to the cluster with the shortest distance to these cluster centers. Finally, the cluster centers were recomputed using the new cluster assignment and these steps would be iterated until convergence was achieved. [34] The cut-off values of hospital volumes were 5.21 and 5.69, and for surgeon's volumes were 2.40 and 4.38 respectively, when cumulative operation volumes within the study period was used as the definition. Likewise, when cumulative operation volumes before each surgery was used as definition, the cut-off values were 4.11 and 4.89 for hospital volumes, and 2.64 and 3.91 for surgeon's volumes. All cutoff values were transformed by logarithm. The results of k-means clustering are demonstrated in Figs 1-4. As the results show, the operation volumes were divided into three groups separately. In addition to surgeon and hospital volumes and SSI, we collected patient-, surgeon-, and hospital-level data. Firstly, patient-level variables included age, gender, length of ICU stay, number of vessels obstructed that were involved in the surgical operation, and the presence of important underlying diseases (e.g. diabetes mellitus, chronic obstructive pulmonary disease (COPD), heart failure, renal failure and renal insufficiency, which were associated with SSI).
[13] Secondly, the surgeon-level variables included age and gender. Thirdly, the hospital-level variables included hospital ownership and geographic location.
All statistical analyses of volume-infection relationship were performed using SAS (version 9.2, SAS Institution Inc., Cary, NC, USA). In statistical testing, a two-sided p value 0.05 was considered statistically significant. The distributional properties of continuous variables were expressed by mean ± standard deviation (SD), whereas categorical variables were presented by frequency and percentage. In univariate analysis, the potential three-level predictors of SSI were examined using chi-square test or two-sample t-test as appropriate. Next, to account for the correlations within surgeon (level-2) and hospital (level-3), multivariate analysis was conducted by fitting mixed-effects logistic regression models to each patient's data for estimating the effects of three-level predictors on the probability of post-operational SSI. Furthermore, subgroup analysis for comorbidities was also conducted. Table 2 shows that there were 7,007 patients with CABG performed by 199 surgeons in 19 hospitals during 2006-2008 in Taiwan. The majority of patients were male (77.5%), and the mean age of patients was 65.3 years. The average ICU stay was 6.05 days, the mean level of number of vessels obstructed was around 1.6, while 51.8% of patients had diabetes mellitus, 33.3% had heart failure, 14.1% had renal failure and renal insufficiency, and 22.0% had COPD. Three hundred and two patients (4.31%) were identified as having the ICD-9-CM SSI codes. However, identification by the CART model only revealed 107 infection cases, and 94 cases were identified in both models. Most cases received CABG surgery by male surgeons, with a mean age of 45.0 years, and the surgeon's average operation volumes within the study period was 151.64, while the average operation volumes before surgery was 52.18. More than half of the cases were performed with CABG in not-for-profit hospitals, and the hospitals' average operation volumes within the study period was 473.60, while the average operation volumes before each surgery was 158.79. Moreover, most of patients received their surgeries by high-volume surgeons and hospitals, when k-means algorithm was used for categorization, regardless of which definition of operation volumes were used. Table 3 shows the results of multilevel mixed-effect models, with the SSIs being identified by ICD-9-CM codes, and the operation volumes defined as the cumulative volumes within the study period. The results of Model 1 (continuous) reveal that the surgeon's volumes were negatively associated with SSIs, while hospital's volumes were not associated with surgical site infection SSIs. Model 2 (quartile) suggests that low-volume surgeons had higher SSI risk (OR = 2.220, p-value = 0.022) than high-volume surgeons. There were also no associations between hospital's operation volumes and SSIs. Model 3 (k-means) shows that the association did not exist between hospital's/ surgeon's volumes and SSIs. Table 4 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volumes within the study period. Model 1 again indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results revealed low-volume surgeons had higher risk (OR = 1.691, p = 0.002) than high-volume surgeons. Table 5 displays the results of multilevel mixed-effect models, in which the SSIs were identified by ICD-9-CM codes, but the operation volumes were defined as the cumulative volume in the previous one year for each surgery. Model 1 also indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.642, p = 0.040) than high-volume surgeons. Table 6 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volume in previous one year for each surgery. In Model 1, different to the above findings, there was no association between hospital's/ surgeon's volumes and SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.163, p = 0.020) than high-volume surgeons.
We further examined the associations of surgeon and hospital volumes with SSIs in stratification analyses by underlying diseases. When the operation volumes were defined as the cumulative operation volume within the study period, no relationships existed between hospital/ surgeon operation volumes and SSIs. (Table 7 ) However, when the operation volumes were defined as the cumulative operation volumes in the previous one year for each surgery, the results suggested that there was a negative association between surgeon volumes and SSIs in the diabetes group, except that the volumes were treated as continuous variable and the infection cases were identified by ICD-9 codes. In terms of hospital operation volumes, the association did not exist. (Table 8 )
No studies have evaluated how different service/ operation volumes definitions and categorization methods affect volume-infection relationships. Moreover, several studies have pointed out the inappropriateness of identifying infection cases using the ICD-9-CM codes in claims data. Given these reasons, this study adopted two approaches to identifying SSIs, two definitions of operation volumes, and three methods for categorizing operation volumes to examine the relationships between operation volumes and SSIs. Our findings showed that the relationships between hospital volumes and SSIs did not exist, no matter which definitions, categorization mehods, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon volumes and SSIs were not robust in our data. It might be affected by different definitions and categorization methods of operation volumes, and also by different SSI cases identification approaches. In summary, most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons, and they also showed the risks were similar between medium-volume and high-volume surgeons. However, why did surgeon volume relate to SSIs, but hospital volume did not? Except for those issues we were concerned about in this study, there are some disagreements in the literature. Such as "Does provider volume really represent quality of care?" [12, 35] Or "Is provider volume the only one predictor for outcome of care?" [35, 36] These issues are worthy of further discussion, but are out of the scope of this study.
Service/ operation volumes are treated as a proxy indicator for experiences; previous studies used it to examine whether practice makes perfect or not. But, except for provider's experiences, SSIs are also impacted by many factors, such as environmental and clinical factors. Wu et al once used Taiwan 2001 NHI claims data to explore the relationship between provider CABG operation volumes and SSIs. [13] They found that hospital volumes had a greater effect than surgeon volumes and claimed that this may imply that hospital teamwork is more important than individual surgeon. However, our findings demonstrated that there was no relationship between hospital volumes and SSIs. Wu et al. adopted the cumulative operation volumes within the study period as the definition, and identified SSIs by ICD-9-CM codes. Except, there were two differences between our work and Wu et al., which were the length and year of the data; our data was longer and more updated than theirs. Moreover, it is worth noting that there was an outbreak of severe acute respiratory syndrome (SARS) in Taiwan in 2003, after which the hospital infection control system in Taiwan was reviewed and re-designed. Wu et al data was before SARS, so these efforts may also have improved the level of SSIs control in hospitals, leading to different findings in this study.
In addition, although most models revealed that there were negative relationships between surgeon's volumes and surgical site infection, the relationships were not robust. The results varied between different definitions and categorization method of operation volumes, and between SSIs identification approaches. Researchers need to consider how to identify SSIs correctly, how to choose optimal cut-off values, and how to decide on which definition is appropriate.
Finally, the results of stratification analyses showed that low-volume surgeon had higher risk than high-volume surgeon in the diabetes mellitus group, when the cumulative operation in the previous one year before surgery was used as definition. A large number of studies have indicated diabetes mellitus is associated with a higher risk of SSIs, [37] [38] [39] and the findings of this study suggest that CABG patients with diabetes mellitus should be cared for by experienced surgeons.
A multilevel analysis was applied to manage the nested factors, and two definitions of operation volume along with three different operation volume categorization methods were adopted to examine the relationship between volume and SSIs under two kinds of SSIs identification approaches. Nevertheless, the study suffered from several major limitations. First, the accuracy of SSIs identification was still an issue. Although the performance of the CART model to identify CABG SSIs was better than ICD-9-CM codes in Taiwan NHI claims data, it did not reach the perfect scenario. The accuracy of SSIs identification was still a challenge in our work. The second limitation relates to unmeasured variables, such as length of stay before operation, infection condition, hair removal, clinical information (e.g. blood glucose level, causative microorganism), time-related information (e.g. the duration of operation), the environment, surgical skills, use of post-operative drains, number of operations involved, and surgical site and wound care, etc. [40] Furthermore, information about type (elective or urgent) and incision site for surgery was not available in the Taiwan NHI claims data.
In conclusion, the findings of this study suggest that different definitions and categorization methods of operation volumes, and different SSIs identification approaches might lead to different findings, although surgeon volumes were more important than hospital volumes in exploring the relationships between CABG operation volumes and SSIs in Taiwan, but they were still not robust. Definitions and categorization methods of operation volumes, and correct identification of SSIs are important issues for future research.
|
What is the purpose of this research study?
| 5,247 |
to examine whether a relationship exists between operation volumes and SSIs
| 793 |
1,598 |
Which Kind of Provider’s Operation Volumes Matters? Associations between CABG Surgical Site Infection Risk and Hospital and Surgeon Operation Volumes among Medical Centers in Taiwan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4459823/
SHA: f3cbc0503581249a834895fc94cd3bae24714a0d
Authors: Yu, Tsung-Hsien; Tung, Yu-Chi; Chung, Kuo-Piao
Date: 2015-06-08
DOI: 10.1371/journal.pone.0129178
License: cc-by
Abstract: BACKGROUND: Volume-infection relationships have been examined for high-risk surgical procedures, but the conclusions remain controversial. The inconsistency might be due to inaccurate identification of cases of infection and different methods of categorizing service volumes. This study takes coronary artery bypass graft (CABG) surgical site infections (SSIs) as an example to examine whether a relationship exists between operation volumes and SSIs, when different SSIs case identification, definitions and categorization methods of operation volumes were implemented. METHODS: A population-based cross-sectional multilevel study was conducted. A total of 7,007 patients who received CABG surgery between 2006 and 2008 from19 medical centers in Taiwan were recruited. SSIs associated with CABG surgery were identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9 CM) codes and a Classification and Regression Trees (CART) model. Two definitions of surgeon and hospital operation volumes were used: (1) the cumulative CABG operation volumes within the study period; and (2) the cumulative CABG operation volumes in the previous one year before each CABG surgery. Operation volumes were further treated in three different ways: (1) a continuous variable; (2) a categorical variable based on the quartile; and (3) a data-driven categorical variable based on k-means clustering algorithm. Furthermore, subgroup analysis for comorbidities was also conducted. RESULTS: This study showed that hospital volumes were not significantly associated with SSIs, no matter which definitions or categorization methods of operation volume, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon’s volumes varied. Most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons. CONCLUSION: Surgeon volumes were more important than hospital volumes in exploring the relationship between CABG operation volumes and SSIs in Taiwan. However, the relationships were not robust. Definitions and categorization methods of operation volume and correct identification of SSIs are important issues for future research.
Text: data, which should use hierarchical models, may result in biased estimation of the variation and also lead to incorrect conclusions.
SSIs following coronary artery bypass graft (CABG) procedures place a heavy burden on patients and healthcare systems. The total length of stay and expenditure for patients with SSIs after CABG surgery is significantly longer and higher than those without SSIs. [20, 21] In 2008, the Centers for Medicare & Medicaid of the United States of America implemented the "Never Event" policy, where hospitals would no longer receive higher payments for the additional costs associated with treating patients for certain healthcare-acquired infections, including those related to CABG.
In view of the accuracy of SSIs identification and the heterogeneity of definition and categorization methods, no existing studies have used different infection case identification nor definitions and categorization methods of operation volume simultaneously to explore the relationship between operation volumes and infection. The current study takes CABG SSIs as an example to examine whether a relationship exists between operation volumes and SSIs, given different SSI cases identification, operation volume definitions and categorization methods.
This retrospective and cross-sectional study adopted a multilevel design to examine the relationships between provider volumes and SSIs after adjusting for patient-, surgeon-, and hospital-level covariates.
We used data from the Taiwan National Health Insurance Research Database (NHIRD) from 2005 and 2008. The NHIRD, published by the Taiwan National Health Research Institute, includes all the original claims data and registration files for beneficiaries enrolled under the National Health Insurance (NHI) program. The database covers the 23 million Taiwanese enrollees (approximately 98% of the population) in the NHI program. It is a de-identified secondary database containing patient-level demographic and administrative information; however, treatment items are aggregated and without time-related and clinical information. The data is released for research purposes.
The protocol for the study was approved by the Institutional Review Board of the National Taiwan University Hospital (protocol #201001027R). The dataset we used in this study was secondary data; all information was de-identified by data owners.
In this study, we adopted the ICD-9-CM SSI codes (hereafter referred to as the ICD-9-CM based model) and the Classification and Regression Trees (CART) model, which was developed in our previous work [11] to identify SSI cases. As we mentioned above, the ICD-9-CM SSI codes were the most popular tool to identify the SSI cases in claims data. In the ICD-9-CM based model, SSI cases were divided into two categories: index hospitalization events and post-discharge events (i.e., SSIs that occurred within 1 year after discharge and required readmission to a hospital and/ or the use of ambulatory services). Following Wu et al [13] , this study adopted the secondary ICD-9-CM diagnosis codes for index hospitalization events (ICD-9-CM code: 996.03, 996.61, 996.72, and 998.5), and the primary and secondary diagnosis codes for post-discharge events (ICD-9-CM code: 038.0-038. 4 ) as the criteria for SSI identification, in order to avoid cases in which infection existed prior to hospitalization. If a case had an index hospitalization event or a post-discharge event, then he/ she will be identified as SSIs by the ICD-9-CM based model. In the CART model, we adopted the type of antibiotics, dose of cefazolin, length of stay, and number of vessels obstructed (as a proxy indicator of duration of operation) as the parameters to identify the SSIs, according to our previous findings. [11] In our previous work, we used the 2005-2008 National Health Insurance claims data and healthcare-associated infection surveillance data from two medical centers for model development and model verification. Infection cases based on surveillance were identified by infection control personnel if the patient met the Taiwan CDC's criteria, which are the same as those adopted in the U.S. CDC. They manually review medical records of all patients at risk for the specified healthcare-associated infection.
The classification algorithms, the multivariable regression model, and the data mining model were adopted to develop alternative models based on surrogate indicators to identify cases of CABG SSIs and to compare the performance among these models and the ICD-9-CMbased model. For the classification algorithms, researchers build up several criteria, and if a case satisfies (or exceeds) a specific number of criteria, then it will be identified as a case of infection. For the multivariable regression model, researchers usually calculated a risk score by the logistic regression model, and the optimal cutoff point was determined according to the resulting receiver operating characteristic curve.
Concerning the data mining approach, which is widely used for predicting and classifying objects, the characteristics are: automatic discovery of patterns, prediction of likely outcomes, creation of actionable information, and focus on large data sets and databases. The classification and regression tree (CART) model, which is the most popular approach as applied in our work, and the growing, stopping, and pruning of the tree were determined by Gini improvement measures. [22, 23] After referring to the literature and conferring with infectious disease specialists, we adopted the following seven parameters: type of antibiotic, doses of antibiotic, doses of cefazolin, use of second-line antibiotics, length of stay, and number of vessels obstructed. Additionally, cross-validation was also employed, where data from one medical center was used for model development, and another one was used for model validation.
The results of our previous work revealed that the CART model offered better performance than that of the other identification models or the ICD-9-CM based model, especially in the positive predictive value (>70%), which was only found to be 20% in the ICD-9-CM based model. (Table 1 ) The findings also implied that the CART was a decidedly better tool for identifying cases of SSI in the Taiwan National Health Insurance database. Therefore, this study also adopted the CART model for identifying CABG SSIs.
To ensure homogeneity, current study analyzed 7,007 patients from 19 medical centers in Taiwan who underwent CABG surgery (ICD-9-CM procedure codes 36.1x-36.2x) between 2006 and 2008. CABG patients under the age of 18 years or over 85 years were excluded in this study. A total of 302 cases were identified as SSIs by ICD-9-CM based model, and a total of 107 cases were identified as SSIs by CART model.
In this study, we used the following two definitions to define operation volumes: (1) the cumulative operation volumes by each surgeon and hospital within the study period, which was the most common definition in the literature; and (2) following Yasunaga et al.'s study, [24] cumulative operation volumes by each surgeon and hospital in the previous one year for each surgery. However, our data was skewed, which did not follow a normal distribution. Therefore, we conducted the log transformations on operation volumes.
The current work treated operation volumes in three different ways: (1) a continuous variable; (2) a categorical variable based on the first and the third quartile as cutoff points (the most common method to categorize service/ operation volumes) [25] [26] [27] [28] ; and (3) a data-driven categorical variable based on k-means clustering algorithm. This study categorized surgeon and hospital volumes into low, medium, and high volume groups by quartile method and kmeans clustering algorithm.
In the quartile method, the cut-off value (transformed by logarithm) of the first quartile (<25%) for hospital volumes was 5.65, and the third quartile (>75%) was 6.43. In terms of surgeon volumes, the first quartile was 4.38, and the third was 5.35, when we used the cumulative operation volumes within the study period as the definition. While the definition changed, first quartile (<25%) for hospital volumes was 4.66, and the third quartile (>75%) was 5.31. In terms of surgeon volumes, the first quartile was 3.40, and the third was 4.32.
K-means clustering is an unsupervised machine-learning algorithm introduced by MacQueen in 1960s. This method is not only a simple and very reliable method in categorization/ classification, but is also recognized as one of the top 10 algorithms in data mining. [29] This method has often been applied in many fields. [30] [31] [32] Yu and his colleagues even applied it to define the quality of CABG care, and to explore the relationship among patient's income status, the level of quality of care, and inpatient mortality. [33] The main idea of this method is to partition observed data points into k non-overlapping clusters by minimizing the within-group sum of squares. Each point is assigned to the mean of its cluster using the Euclidian distance. Firstly, k cluster centers were randomly generated. Previous studies usually divided surgeons and hospitals into low-, medium-, and high-volume groups; therefore, we also predetermined the surgeon and hospital service volumes into 3 groups (k = 3). Then, participants were assigned to the cluster with the shortest distance to these cluster centers. Finally, the cluster centers were recomputed using the new cluster assignment and these steps would be iterated until convergence was achieved. [34] The cut-off values of hospital volumes were 5.21 and 5.69, and for surgeon's volumes were 2.40 and 4.38 respectively, when cumulative operation volumes within the study period was used as the definition. Likewise, when cumulative operation volumes before each surgery was used as definition, the cut-off values were 4.11 and 4.89 for hospital volumes, and 2.64 and 3.91 for surgeon's volumes. All cutoff values were transformed by logarithm. The results of k-means clustering are demonstrated in Figs 1-4. As the results show, the operation volumes were divided into three groups separately. In addition to surgeon and hospital volumes and SSI, we collected patient-, surgeon-, and hospital-level data. Firstly, patient-level variables included age, gender, length of ICU stay, number of vessels obstructed that were involved in the surgical operation, and the presence of important underlying diseases (e.g. diabetes mellitus, chronic obstructive pulmonary disease (COPD), heart failure, renal failure and renal insufficiency, which were associated with SSI).
[13] Secondly, the surgeon-level variables included age and gender. Thirdly, the hospital-level variables included hospital ownership and geographic location.
All statistical analyses of volume-infection relationship were performed using SAS (version 9.2, SAS Institution Inc., Cary, NC, USA). In statistical testing, a two-sided p value 0.05 was considered statistically significant. The distributional properties of continuous variables were expressed by mean ± standard deviation (SD), whereas categorical variables were presented by frequency and percentage. In univariate analysis, the potential three-level predictors of SSI were examined using chi-square test or two-sample t-test as appropriate. Next, to account for the correlations within surgeon (level-2) and hospital (level-3), multivariate analysis was conducted by fitting mixed-effects logistic regression models to each patient's data for estimating the effects of three-level predictors on the probability of post-operational SSI. Furthermore, subgroup analysis for comorbidities was also conducted. Table 2 shows that there were 7,007 patients with CABG performed by 199 surgeons in 19 hospitals during 2006-2008 in Taiwan. The majority of patients were male (77.5%), and the mean age of patients was 65.3 years. The average ICU stay was 6.05 days, the mean level of number of vessels obstructed was around 1.6, while 51.8% of patients had diabetes mellitus, 33.3% had heart failure, 14.1% had renal failure and renal insufficiency, and 22.0% had COPD. Three hundred and two patients (4.31%) were identified as having the ICD-9-CM SSI codes. However, identification by the CART model only revealed 107 infection cases, and 94 cases were identified in both models. Most cases received CABG surgery by male surgeons, with a mean age of 45.0 years, and the surgeon's average operation volumes within the study period was 151.64, while the average operation volumes before surgery was 52.18. More than half of the cases were performed with CABG in not-for-profit hospitals, and the hospitals' average operation volumes within the study period was 473.60, while the average operation volumes before each surgery was 158.79. Moreover, most of patients received their surgeries by high-volume surgeons and hospitals, when k-means algorithm was used for categorization, regardless of which definition of operation volumes were used. Table 3 shows the results of multilevel mixed-effect models, with the SSIs being identified by ICD-9-CM codes, and the operation volumes defined as the cumulative volumes within the study period. The results of Model 1 (continuous) reveal that the surgeon's volumes were negatively associated with SSIs, while hospital's volumes were not associated with surgical site infection SSIs. Model 2 (quartile) suggests that low-volume surgeons had higher SSI risk (OR = 2.220, p-value = 0.022) than high-volume surgeons. There were also no associations between hospital's operation volumes and SSIs. Model 3 (k-means) shows that the association did not exist between hospital's/ surgeon's volumes and SSIs. Table 4 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volumes within the study period. Model 1 again indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results revealed low-volume surgeons had higher risk (OR = 1.691, p = 0.002) than high-volume surgeons. Table 5 displays the results of multilevel mixed-effect models, in which the SSIs were identified by ICD-9-CM codes, but the operation volumes were defined as the cumulative volume in the previous one year for each surgery. Model 1 also indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.642, p = 0.040) than high-volume surgeons. Table 6 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volume in previous one year for each surgery. In Model 1, different to the above findings, there was no association between hospital's/ surgeon's volumes and SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.163, p = 0.020) than high-volume surgeons.
We further examined the associations of surgeon and hospital volumes with SSIs in stratification analyses by underlying diseases. When the operation volumes were defined as the cumulative operation volume within the study period, no relationships existed between hospital/ surgeon operation volumes and SSIs. (Table 7 ) However, when the operation volumes were defined as the cumulative operation volumes in the previous one year for each surgery, the results suggested that there was a negative association between surgeon volumes and SSIs in the diabetes group, except that the volumes were treated as continuous variable and the infection cases were identified by ICD-9 codes. In terms of hospital operation volumes, the association did not exist. (Table 8 )
No studies have evaluated how different service/ operation volumes definitions and categorization methods affect volume-infection relationships. Moreover, several studies have pointed out the inappropriateness of identifying infection cases using the ICD-9-CM codes in claims data. Given these reasons, this study adopted two approaches to identifying SSIs, two definitions of operation volumes, and three methods for categorizing operation volumes to examine the relationships between operation volumes and SSIs. Our findings showed that the relationships between hospital volumes and SSIs did not exist, no matter which definitions, categorization mehods, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon volumes and SSIs were not robust in our data. It might be affected by different definitions and categorization methods of operation volumes, and also by different SSI cases identification approaches. In summary, most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons, and they also showed the risks were similar between medium-volume and high-volume surgeons. However, why did surgeon volume relate to SSIs, but hospital volume did not? Except for those issues we were concerned about in this study, there are some disagreements in the literature. Such as "Does provider volume really represent quality of care?" [12, 35] Or "Is provider volume the only one predictor for outcome of care?" [35, 36] These issues are worthy of further discussion, but are out of the scope of this study.
Service/ operation volumes are treated as a proxy indicator for experiences; previous studies used it to examine whether practice makes perfect or not. But, except for provider's experiences, SSIs are also impacted by many factors, such as environmental and clinical factors. Wu et al once used Taiwan 2001 NHI claims data to explore the relationship between provider CABG operation volumes and SSIs. [13] They found that hospital volumes had a greater effect than surgeon volumes and claimed that this may imply that hospital teamwork is more important than individual surgeon. However, our findings demonstrated that there was no relationship between hospital volumes and SSIs. Wu et al. adopted the cumulative operation volumes within the study period as the definition, and identified SSIs by ICD-9-CM codes. Except, there were two differences between our work and Wu et al., which were the length and year of the data; our data was longer and more updated than theirs. Moreover, it is worth noting that there was an outbreak of severe acute respiratory syndrome (SARS) in Taiwan in 2003, after which the hospital infection control system in Taiwan was reviewed and re-designed. Wu et al data was before SARS, so these efforts may also have improved the level of SSIs control in hospitals, leading to different findings in this study.
In addition, although most models revealed that there were negative relationships between surgeon's volumes and surgical site infection, the relationships were not robust. The results varied between different definitions and categorization method of operation volumes, and between SSIs identification approaches. Researchers need to consider how to identify SSIs correctly, how to choose optimal cut-off values, and how to decide on which definition is appropriate.
Finally, the results of stratification analyses showed that low-volume surgeon had higher risk than high-volume surgeon in the diabetes mellitus group, when the cumulative operation in the previous one year before surgery was used as definition. A large number of studies have indicated diabetes mellitus is associated with a higher risk of SSIs, [37] [38] [39] and the findings of this study suggest that CABG patients with diabetes mellitus should be cared for by experienced surgeons.
A multilevel analysis was applied to manage the nested factors, and two definitions of operation volume along with three different operation volume categorization methods were adopted to examine the relationship between volume and SSIs under two kinds of SSIs identification approaches. Nevertheless, the study suffered from several major limitations. First, the accuracy of SSIs identification was still an issue. Although the performance of the CART model to identify CABG SSIs was better than ICD-9-CM codes in Taiwan NHI claims data, it did not reach the perfect scenario. The accuracy of SSIs identification was still a challenge in our work. The second limitation relates to unmeasured variables, such as length of stay before operation, infection condition, hair removal, clinical information (e.g. blood glucose level, causative microorganism), time-related information (e.g. the duration of operation), the environment, surgical skills, use of post-operative drains, number of operations involved, and surgical site and wound care, etc. [40] Furthermore, information about type (elective or urgent) and incision site for surgery was not available in the Taiwan NHI claims data.
In conclusion, the findings of this study suggest that different definitions and categorization methods of operation volumes, and different SSIs identification approaches might lead to different findings, although surgeon volumes were more important than hospital volumes in exploring the relationships between CABG operation volumes and SSIs in Taiwan, but they were still not robust. Definitions and categorization methods of operation volumes, and correct identification of SSIs are important issues for future research.
|
Why are SSIs important to the overall burden on the healthcare system?
| 5,248 |
The total length of stay and expenditure for patients with SSIs after CABG surgery is significantly longer and higher than those without SSIs
| 2,903 |
1,598 |
Which Kind of Provider’s Operation Volumes Matters? Associations between CABG Surgical Site Infection Risk and Hospital and Surgeon Operation Volumes among Medical Centers in Taiwan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4459823/
SHA: f3cbc0503581249a834895fc94cd3bae24714a0d
Authors: Yu, Tsung-Hsien; Tung, Yu-Chi; Chung, Kuo-Piao
Date: 2015-06-08
DOI: 10.1371/journal.pone.0129178
License: cc-by
Abstract: BACKGROUND: Volume-infection relationships have been examined for high-risk surgical procedures, but the conclusions remain controversial. The inconsistency might be due to inaccurate identification of cases of infection and different methods of categorizing service volumes. This study takes coronary artery bypass graft (CABG) surgical site infections (SSIs) as an example to examine whether a relationship exists between operation volumes and SSIs, when different SSIs case identification, definitions and categorization methods of operation volumes were implemented. METHODS: A population-based cross-sectional multilevel study was conducted. A total of 7,007 patients who received CABG surgery between 2006 and 2008 from19 medical centers in Taiwan were recruited. SSIs associated with CABG surgery were identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9 CM) codes and a Classification and Regression Trees (CART) model. Two definitions of surgeon and hospital operation volumes were used: (1) the cumulative CABG operation volumes within the study period; and (2) the cumulative CABG operation volumes in the previous one year before each CABG surgery. Operation volumes were further treated in three different ways: (1) a continuous variable; (2) a categorical variable based on the quartile; and (3) a data-driven categorical variable based on k-means clustering algorithm. Furthermore, subgroup analysis for comorbidities was also conducted. RESULTS: This study showed that hospital volumes were not significantly associated with SSIs, no matter which definitions or categorization methods of operation volume, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon’s volumes varied. Most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons. CONCLUSION: Surgeon volumes were more important than hospital volumes in exploring the relationship between CABG operation volumes and SSIs in Taiwan. However, the relationships were not robust. Definitions and categorization methods of operation volume and correct identification of SSIs are important issues for future research.
Text: data, which should use hierarchical models, may result in biased estimation of the variation and also lead to incorrect conclusions.
SSIs following coronary artery bypass graft (CABG) procedures place a heavy burden on patients and healthcare systems. The total length of stay and expenditure for patients with SSIs after CABG surgery is significantly longer and higher than those without SSIs. [20, 21] In 2008, the Centers for Medicare & Medicaid of the United States of America implemented the "Never Event" policy, where hospitals would no longer receive higher payments for the additional costs associated with treating patients for certain healthcare-acquired infections, including those related to CABG.
In view of the accuracy of SSIs identification and the heterogeneity of definition and categorization methods, no existing studies have used different infection case identification nor definitions and categorization methods of operation volume simultaneously to explore the relationship between operation volumes and infection. The current study takes CABG SSIs as an example to examine whether a relationship exists between operation volumes and SSIs, given different SSI cases identification, operation volume definitions and categorization methods.
This retrospective and cross-sectional study adopted a multilevel design to examine the relationships between provider volumes and SSIs after adjusting for patient-, surgeon-, and hospital-level covariates.
We used data from the Taiwan National Health Insurance Research Database (NHIRD) from 2005 and 2008. The NHIRD, published by the Taiwan National Health Research Institute, includes all the original claims data and registration files for beneficiaries enrolled under the National Health Insurance (NHI) program. The database covers the 23 million Taiwanese enrollees (approximately 98% of the population) in the NHI program. It is a de-identified secondary database containing patient-level demographic and administrative information; however, treatment items are aggregated and without time-related and clinical information. The data is released for research purposes.
The protocol for the study was approved by the Institutional Review Board of the National Taiwan University Hospital (protocol #201001027R). The dataset we used in this study was secondary data; all information was de-identified by data owners.
In this study, we adopted the ICD-9-CM SSI codes (hereafter referred to as the ICD-9-CM based model) and the Classification and Regression Trees (CART) model, which was developed in our previous work [11] to identify SSI cases. As we mentioned above, the ICD-9-CM SSI codes were the most popular tool to identify the SSI cases in claims data. In the ICD-9-CM based model, SSI cases were divided into two categories: index hospitalization events and post-discharge events (i.e., SSIs that occurred within 1 year after discharge and required readmission to a hospital and/ or the use of ambulatory services). Following Wu et al [13] , this study adopted the secondary ICD-9-CM diagnosis codes for index hospitalization events (ICD-9-CM code: 996.03, 996.61, 996.72, and 998.5), and the primary and secondary diagnosis codes for post-discharge events (ICD-9-CM code: 038.0-038. 4 ) as the criteria for SSI identification, in order to avoid cases in which infection existed prior to hospitalization. If a case had an index hospitalization event or a post-discharge event, then he/ she will be identified as SSIs by the ICD-9-CM based model. In the CART model, we adopted the type of antibiotics, dose of cefazolin, length of stay, and number of vessels obstructed (as a proxy indicator of duration of operation) as the parameters to identify the SSIs, according to our previous findings. [11] In our previous work, we used the 2005-2008 National Health Insurance claims data and healthcare-associated infection surveillance data from two medical centers for model development and model verification. Infection cases based on surveillance were identified by infection control personnel if the patient met the Taiwan CDC's criteria, which are the same as those adopted in the U.S. CDC. They manually review medical records of all patients at risk for the specified healthcare-associated infection.
The classification algorithms, the multivariable regression model, and the data mining model were adopted to develop alternative models based on surrogate indicators to identify cases of CABG SSIs and to compare the performance among these models and the ICD-9-CMbased model. For the classification algorithms, researchers build up several criteria, and if a case satisfies (or exceeds) a specific number of criteria, then it will be identified as a case of infection. For the multivariable regression model, researchers usually calculated a risk score by the logistic regression model, and the optimal cutoff point was determined according to the resulting receiver operating characteristic curve.
Concerning the data mining approach, which is widely used for predicting and classifying objects, the characteristics are: automatic discovery of patterns, prediction of likely outcomes, creation of actionable information, and focus on large data sets and databases. The classification and regression tree (CART) model, which is the most popular approach as applied in our work, and the growing, stopping, and pruning of the tree were determined by Gini improvement measures. [22, 23] After referring to the literature and conferring with infectious disease specialists, we adopted the following seven parameters: type of antibiotic, doses of antibiotic, doses of cefazolin, use of second-line antibiotics, length of stay, and number of vessels obstructed. Additionally, cross-validation was also employed, where data from one medical center was used for model development, and another one was used for model validation.
The results of our previous work revealed that the CART model offered better performance than that of the other identification models or the ICD-9-CM based model, especially in the positive predictive value (>70%), which was only found to be 20% in the ICD-9-CM based model. (Table 1 ) The findings also implied that the CART was a decidedly better tool for identifying cases of SSI in the Taiwan National Health Insurance database. Therefore, this study also adopted the CART model for identifying CABG SSIs.
To ensure homogeneity, current study analyzed 7,007 patients from 19 medical centers in Taiwan who underwent CABG surgery (ICD-9-CM procedure codes 36.1x-36.2x) between 2006 and 2008. CABG patients under the age of 18 years or over 85 years were excluded in this study. A total of 302 cases were identified as SSIs by ICD-9-CM based model, and a total of 107 cases were identified as SSIs by CART model.
In this study, we used the following two definitions to define operation volumes: (1) the cumulative operation volumes by each surgeon and hospital within the study period, which was the most common definition in the literature; and (2) following Yasunaga et al.'s study, [24] cumulative operation volumes by each surgeon and hospital in the previous one year for each surgery. However, our data was skewed, which did not follow a normal distribution. Therefore, we conducted the log transformations on operation volumes.
The current work treated operation volumes in three different ways: (1) a continuous variable; (2) a categorical variable based on the first and the third quartile as cutoff points (the most common method to categorize service/ operation volumes) [25] [26] [27] [28] ; and (3) a data-driven categorical variable based on k-means clustering algorithm. This study categorized surgeon and hospital volumes into low, medium, and high volume groups by quartile method and kmeans clustering algorithm.
In the quartile method, the cut-off value (transformed by logarithm) of the first quartile (<25%) for hospital volumes was 5.65, and the third quartile (>75%) was 6.43. In terms of surgeon volumes, the first quartile was 4.38, and the third was 5.35, when we used the cumulative operation volumes within the study period as the definition. While the definition changed, first quartile (<25%) for hospital volumes was 4.66, and the third quartile (>75%) was 5.31. In terms of surgeon volumes, the first quartile was 3.40, and the third was 4.32.
K-means clustering is an unsupervised machine-learning algorithm introduced by MacQueen in 1960s. This method is not only a simple and very reliable method in categorization/ classification, but is also recognized as one of the top 10 algorithms in data mining. [29] This method has often been applied in many fields. [30] [31] [32] Yu and his colleagues even applied it to define the quality of CABG care, and to explore the relationship among patient's income status, the level of quality of care, and inpatient mortality. [33] The main idea of this method is to partition observed data points into k non-overlapping clusters by minimizing the within-group sum of squares. Each point is assigned to the mean of its cluster using the Euclidian distance. Firstly, k cluster centers were randomly generated. Previous studies usually divided surgeons and hospitals into low-, medium-, and high-volume groups; therefore, we also predetermined the surgeon and hospital service volumes into 3 groups (k = 3). Then, participants were assigned to the cluster with the shortest distance to these cluster centers. Finally, the cluster centers were recomputed using the new cluster assignment and these steps would be iterated until convergence was achieved. [34] The cut-off values of hospital volumes were 5.21 and 5.69, and for surgeon's volumes were 2.40 and 4.38 respectively, when cumulative operation volumes within the study period was used as the definition. Likewise, when cumulative operation volumes before each surgery was used as definition, the cut-off values were 4.11 and 4.89 for hospital volumes, and 2.64 and 3.91 for surgeon's volumes. All cutoff values were transformed by logarithm. The results of k-means clustering are demonstrated in Figs 1-4. As the results show, the operation volumes were divided into three groups separately. In addition to surgeon and hospital volumes and SSI, we collected patient-, surgeon-, and hospital-level data. Firstly, patient-level variables included age, gender, length of ICU stay, number of vessels obstructed that were involved in the surgical operation, and the presence of important underlying diseases (e.g. diabetes mellitus, chronic obstructive pulmonary disease (COPD), heart failure, renal failure and renal insufficiency, which were associated with SSI).
[13] Secondly, the surgeon-level variables included age and gender. Thirdly, the hospital-level variables included hospital ownership and geographic location.
All statistical analyses of volume-infection relationship were performed using SAS (version 9.2, SAS Institution Inc., Cary, NC, USA). In statistical testing, a two-sided p value 0.05 was considered statistically significant. The distributional properties of continuous variables were expressed by mean ± standard deviation (SD), whereas categorical variables were presented by frequency and percentage. In univariate analysis, the potential three-level predictors of SSI were examined using chi-square test or two-sample t-test as appropriate. Next, to account for the correlations within surgeon (level-2) and hospital (level-3), multivariate analysis was conducted by fitting mixed-effects logistic regression models to each patient's data for estimating the effects of three-level predictors on the probability of post-operational SSI. Furthermore, subgroup analysis for comorbidities was also conducted. Table 2 shows that there were 7,007 patients with CABG performed by 199 surgeons in 19 hospitals during 2006-2008 in Taiwan. The majority of patients were male (77.5%), and the mean age of patients was 65.3 years. The average ICU stay was 6.05 days, the mean level of number of vessels obstructed was around 1.6, while 51.8% of patients had diabetes mellitus, 33.3% had heart failure, 14.1% had renal failure and renal insufficiency, and 22.0% had COPD. Three hundred and two patients (4.31%) were identified as having the ICD-9-CM SSI codes. However, identification by the CART model only revealed 107 infection cases, and 94 cases were identified in both models. Most cases received CABG surgery by male surgeons, with a mean age of 45.0 years, and the surgeon's average operation volumes within the study period was 151.64, while the average operation volumes before surgery was 52.18. More than half of the cases were performed with CABG in not-for-profit hospitals, and the hospitals' average operation volumes within the study period was 473.60, while the average operation volumes before each surgery was 158.79. Moreover, most of patients received their surgeries by high-volume surgeons and hospitals, when k-means algorithm was used for categorization, regardless of which definition of operation volumes were used. Table 3 shows the results of multilevel mixed-effect models, with the SSIs being identified by ICD-9-CM codes, and the operation volumes defined as the cumulative volumes within the study period. The results of Model 1 (continuous) reveal that the surgeon's volumes were negatively associated with SSIs, while hospital's volumes were not associated with surgical site infection SSIs. Model 2 (quartile) suggests that low-volume surgeons had higher SSI risk (OR = 2.220, p-value = 0.022) than high-volume surgeons. There were also no associations between hospital's operation volumes and SSIs. Model 3 (k-means) shows that the association did not exist between hospital's/ surgeon's volumes and SSIs. Table 4 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volumes within the study period. Model 1 again indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results revealed low-volume surgeons had higher risk (OR = 1.691, p = 0.002) than high-volume surgeons. Table 5 displays the results of multilevel mixed-effect models, in which the SSIs were identified by ICD-9-CM codes, but the operation volumes were defined as the cumulative volume in the previous one year for each surgery. Model 1 also indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.642, p = 0.040) than high-volume surgeons. Table 6 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volume in previous one year for each surgery. In Model 1, different to the above findings, there was no association between hospital's/ surgeon's volumes and SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.163, p = 0.020) than high-volume surgeons.
We further examined the associations of surgeon and hospital volumes with SSIs in stratification analyses by underlying diseases. When the operation volumes were defined as the cumulative operation volume within the study period, no relationships existed between hospital/ surgeon operation volumes and SSIs. (Table 7 ) However, when the operation volumes were defined as the cumulative operation volumes in the previous one year for each surgery, the results suggested that there was a negative association between surgeon volumes and SSIs in the diabetes group, except that the volumes were treated as continuous variable and the infection cases were identified by ICD-9 codes. In terms of hospital operation volumes, the association did not exist. (Table 8 )
No studies have evaluated how different service/ operation volumes definitions and categorization methods affect volume-infection relationships. Moreover, several studies have pointed out the inappropriateness of identifying infection cases using the ICD-9-CM codes in claims data. Given these reasons, this study adopted two approaches to identifying SSIs, two definitions of operation volumes, and three methods for categorizing operation volumes to examine the relationships between operation volumes and SSIs. Our findings showed that the relationships between hospital volumes and SSIs did not exist, no matter which definitions, categorization mehods, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon volumes and SSIs were not robust in our data. It might be affected by different definitions and categorization methods of operation volumes, and also by different SSI cases identification approaches. In summary, most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons, and they also showed the risks were similar between medium-volume and high-volume surgeons. However, why did surgeon volume relate to SSIs, but hospital volume did not? Except for those issues we were concerned about in this study, there are some disagreements in the literature. Such as "Does provider volume really represent quality of care?" [12, 35] Or "Is provider volume the only one predictor for outcome of care?" [35, 36] These issues are worthy of further discussion, but are out of the scope of this study.
Service/ operation volumes are treated as a proxy indicator for experiences; previous studies used it to examine whether practice makes perfect or not. But, except for provider's experiences, SSIs are also impacted by many factors, such as environmental and clinical factors. Wu et al once used Taiwan 2001 NHI claims data to explore the relationship between provider CABG operation volumes and SSIs. [13] They found that hospital volumes had a greater effect than surgeon volumes and claimed that this may imply that hospital teamwork is more important than individual surgeon. However, our findings demonstrated that there was no relationship between hospital volumes and SSIs. Wu et al. adopted the cumulative operation volumes within the study period as the definition, and identified SSIs by ICD-9-CM codes. Except, there were two differences between our work and Wu et al., which were the length and year of the data; our data was longer and more updated than theirs. Moreover, it is worth noting that there was an outbreak of severe acute respiratory syndrome (SARS) in Taiwan in 2003, after which the hospital infection control system in Taiwan was reviewed and re-designed. Wu et al data was before SARS, so these efforts may also have improved the level of SSIs control in hospitals, leading to different findings in this study.
In addition, although most models revealed that there were negative relationships between surgeon's volumes and surgical site infection, the relationships were not robust. The results varied between different definitions and categorization method of operation volumes, and between SSIs identification approaches. Researchers need to consider how to identify SSIs correctly, how to choose optimal cut-off values, and how to decide on which definition is appropriate.
Finally, the results of stratification analyses showed that low-volume surgeon had higher risk than high-volume surgeon in the diabetes mellitus group, when the cumulative operation in the previous one year before surgery was used as definition. A large number of studies have indicated diabetes mellitus is associated with a higher risk of SSIs, [37] [38] [39] and the findings of this study suggest that CABG patients with diabetes mellitus should be cared for by experienced surgeons.
A multilevel analysis was applied to manage the nested factors, and two definitions of operation volume along with three different operation volume categorization methods were adopted to examine the relationship between volume and SSIs under two kinds of SSIs identification approaches. Nevertheless, the study suffered from several major limitations. First, the accuracy of SSIs identification was still an issue. Although the performance of the CART model to identify CABG SSIs was better than ICD-9-CM codes in Taiwan NHI claims data, it did not reach the perfect scenario. The accuracy of SSIs identification was still a challenge in our work. The second limitation relates to unmeasured variables, such as length of stay before operation, infection condition, hair removal, clinical information (e.g. blood glucose level, causative microorganism), time-related information (e.g. the duration of operation), the environment, surgical skills, use of post-operative drains, number of operations involved, and surgical site and wound care, etc. [40] Furthermore, information about type (elective or urgent) and incision site for surgery was not available in the Taiwan NHI claims data.
In conclusion, the findings of this study suggest that different definitions and categorization methods of operation volumes, and different SSIs identification approaches might lead to different findings, although surgeon volumes were more important than hospital volumes in exploring the relationships between CABG operation volumes and SSIs in Taiwan, but they were still not robust. Definitions and categorization methods of operation volumes, and correct identification of SSIs are important issues for future research.
|
What is the "Never Event" policy?
| 5,249 |
hospitals would no longer receive higher payments for the additional costs associated with treating patients for certain healthcare-acquired infections
| 3,176 |
1,598 |
Which Kind of Provider’s Operation Volumes Matters? Associations between CABG Surgical Site Infection Risk and Hospital and Surgeon Operation Volumes among Medical Centers in Taiwan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4459823/
SHA: f3cbc0503581249a834895fc94cd3bae24714a0d
Authors: Yu, Tsung-Hsien; Tung, Yu-Chi; Chung, Kuo-Piao
Date: 2015-06-08
DOI: 10.1371/journal.pone.0129178
License: cc-by
Abstract: BACKGROUND: Volume-infection relationships have been examined for high-risk surgical procedures, but the conclusions remain controversial. The inconsistency might be due to inaccurate identification of cases of infection and different methods of categorizing service volumes. This study takes coronary artery bypass graft (CABG) surgical site infections (SSIs) as an example to examine whether a relationship exists between operation volumes and SSIs, when different SSIs case identification, definitions and categorization methods of operation volumes were implemented. METHODS: A population-based cross-sectional multilevel study was conducted. A total of 7,007 patients who received CABG surgery between 2006 and 2008 from19 medical centers in Taiwan were recruited. SSIs associated with CABG surgery were identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9 CM) codes and a Classification and Regression Trees (CART) model. Two definitions of surgeon and hospital operation volumes were used: (1) the cumulative CABG operation volumes within the study period; and (2) the cumulative CABG operation volumes in the previous one year before each CABG surgery. Operation volumes were further treated in three different ways: (1) a continuous variable; (2) a categorical variable based on the quartile; and (3) a data-driven categorical variable based on k-means clustering algorithm. Furthermore, subgroup analysis for comorbidities was also conducted. RESULTS: This study showed that hospital volumes were not significantly associated with SSIs, no matter which definitions or categorization methods of operation volume, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon’s volumes varied. Most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons. CONCLUSION: Surgeon volumes were more important than hospital volumes in exploring the relationship between CABG operation volumes and SSIs in Taiwan. However, the relationships were not robust. Definitions and categorization methods of operation volume and correct identification of SSIs are important issues for future research.
Text: data, which should use hierarchical models, may result in biased estimation of the variation and also lead to incorrect conclusions.
SSIs following coronary artery bypass graft (CABG) procedures place a heavy burden on patients and healthcare systems. The total length of stay and expenditure for patients with SSIs after CABG surgery is significantly longer and higher than those without SSIs. [20, 21] In 2008, the Centers for Medicare & Medicaid of the United States of America implemented the "Never Event" policy, where hospitals would no longer receive higher payments for the additional costs associated with treating patients for certain healthcare-acquired infections, including those related to CABG.
In view of the accuracy of SSIs identification and the heterogeneity of definition and categorization methods, no existing studies have used different infection case identification nor definitions and categorization methods of operation volume simultaneously to explore the relationship between operation volumes and infection. The current study takes CABG SSIs as an example to examine whether a relationship exists between operation volumes and SSIs, given different SSI cases identification, operation volume definitions and categorization methods.
This retrospective and cross-sectional study adopted a multilevel design to examine the relationships between provider volumes and SSIs after adjusting for patient-, surgeon-, and hospital-level covariates.
We used data from the Taiwan National Health Insurance Research Database (NHIRD) from 2005 and 2008. The NHIRD, published by the Taiwan National Health Research Institute, includes all the original claims data and registration files for beneficiaries enrolled under the National Health Insurance (NHI) program. The database covers the 23 million Taiwanese enrollees (approximately 98% of the population) in the NHI program. It is a de-identified secondary database containing patient-level demographic and administrative information; however, treatment items are aggregated and without time-related and clinical information. The data is released for research purposes.
The protocol for the study was approved by the Institutional Review Board of the National Taiwan University Hospital (protocol #201001027R). The dataset we used in this study was secondary data; all information was de-identified by data owners.
In this study, we adopted the ICD-9-CM SSI codes (hereafter referred to as the ICD-9-CM based model) and the Classification and Regression Trees (CART) model, which was developed in our previous work [11] to identify SSI cases. As we mentioned above, the ICD-9-CM SSI codes were the most popular tool to identify the SSI cases in claims data. In the ICD-9-CM based model, SSI cases were divided into two categories: index hospitalization events and post-discharge events (i.e., SSIs that occurred within 1 year after discharge and required readmission to a hospital and/ or the use of ambulatory services). Following Wu et al [13] , this study adopted the secondary ICD-9-CM diagnosis codes for index hospitalization events (ICD-9-CM code: 996.03, 996.61, 996.72, and 998.5), and the primary and secondary diagnosis codes for post-discharge events (ICD-9-CM code: 038.0-038. 4 ) as the criteria for SSI identification, in order to avoid cases in which infection existed prior to hospitalization. If a case had an index hospitalization event or a post-discharge event, then he/ she will be identified as SSIs by the ICD-9-CM based model. In the CART model, we adopted the type of antibiotics, dose of cefazolin, length of stay, and number of vessels obstructed (as a proxy indicator of duration of operation) as the parameters to identify the SSIs, according to our previous findings. [11] In our previous work, we used the 2005-2008 National Health Insurance claims data and healthcare-associated infection surveillance data from two medical centers for model development and model verification. Infection cases based on surveillance were identified by infection control personnel if the patient met the Taiwan CDC's criteria, which are the same as those adopted in the U.S. CDC. They manually review medical records of all patients at risk for the specified healthcare-associated infection.
The classification algorithms, the multivariable regression model, and the data mining model were adopted to develop alternative models based on surrogate indicators to identify cases of CABG SSIs and to compare the performance among these models and the ICD-9-CMbased model. For the classification algorithms, researchers build up several criteria, and if a case satisfies (or exceeds) a specific number of criteria, then it will be identified as a case of infection. For the multivariable regression model, researchers usually calculated a risk score by the logistic regression model, and the optimal cutoff point was determined according to the resulting receiver operating characteristic curve.
Concerning the data mining approach, which is widely used for predicting and classifying objects, the characteristics are: automatic discovery of patterns, prediction of likely outcomes, creation of actionable information, and focus on large data sets and databases. The classification and regression tree (CART) model, which is the most popular approach as applied in our work, and the growing, stopping, and pruning of the tree were determined by Gini improvement measures. [22, 23] After referring to the literature and conferring with infectious disease specialists, we adopted the following seven parameters: type of antibiotic, doses of antibiotic, doses of cefazolin, use of second-line antibiotics, length of stay, and number of vessels obstructed. Additionally, cross-validation was also employed, where data from one medical center was used for model development, and another one was used for model validation.
The results of our previous work revealed that the CART model offered better performance than that of the other identification models or the ICD-9-CM based model, especially in the positive predictive value (>70%), which was only found to be 20% in the ICD-9-CM based model. (Table 1 ) The findings also implied that the CART was a decidedly better tool for identifying cases of SSI in the Taiwan National Health Insurance database. Therefore, this study also adopted the CART model for identifying CABG SSIs.
To ensure homogeneity, current study analyzed 7,007 patients from 19 medical centers in Taiwan who underwent CABG surgery (ICD-9-CM procedure codes 36.1x-36.2x) between 2006 and 2008. CABG patients under the age of 18 years or over 85 years were excluded in this study. A total of 302 cases were identified as SSIs by ICD-9-CM based model, and a total of 107 cases were identified as SSIs by CART model.
In this study, we used the following two definitions to define operation volumes: (1) the cumulative operation volumes by each surgeon and hospital within the study period, which was the most common definition in the literature; and (2) following Yasunaga et al.'s study, [24] cumulative operation volumes by each surgeon and hospital in the previous one year for each surgery. However, our data was skewed, which did not follow a normal distribution. Therefore, we conducted the log transformations on operation volumes.
The current work treated operation volumes in three different ways: (1) a continuous variable; (2) a categorical variable based on the first and the third quartile as cutoff points (the most common method to categorize service/ operation volumes) [25] [26] [27] [28] ; and (3) a data-driven categorical variable based on k-means clustering algorithm. This study categorized surgeon and hospital volumes into low, medium, and high volume groups by quartile method and kmeans clustering algorithm.
In the quartile method, the cut-off value (transformed by logarithm) of the first quartile (<25%) for hospital volumes was 5.65, and the third quartile (>75%) was 6.43. In terms of surgeon volumes, the first quartile was 4.38, and the third was 5.35, when we used the cumulative operation volumes within the study period as the definition. While the definition changed, first quartile (<25%) for hospital volumes was 4.66, and the third quartile (>75%) was 5.31. In terms of surgeon volumes, the first quartile was 3.40, and the third was 4.32.
K-means clustering is an unsupervised machine-learning algorithm introduced by MacQueen in 1960s. This method is not only a simple and very reliable method in categorization/ classification, but is also recognized as one of the top 10 algorithms in data mining. [29] This method has often been applied in many fields. [30] [31] [32] Yu and his colleagues even applied it to define the quality of CABG care, and to explore the relationship among patient's income status, the level of quality of care, and inpatient mortality. [33] The main idea of this method is to partition observed data points into k non-overlapping clusters by minimizing the within-group sum of squares. Each point is assigned to the mean of its cluster using the Euclidian distance. Firstly, k cluster centers were randomly generated. Previous studies usually divided surgeons and hospitals into low-, medium-, and high-volume groups; therefore, we also predetermined the surgeon and hospital service volumes into 3 groups (k = 3). Then, participants were assigned to the cluster with the shortest distance to these cluster centers. Finally, the cluster centers were recomputed using the new cluster assignment and these steps would be iterated until convergence was achieved. [34] The cut-off values of hospital volumes were 5.21 and 5.69, and for surgeon's volumes were 2.40 and 4.38 respectively, when cumulative operation volumes within the study period was used as the definition. Likewise, when cumulative operation volumes before each surgery was used as definition, the cut-off values were 4.11 and 4.89 for hospital volumes, and 2.64 and 3.91 for surgeon's volumes. All cutoff values were transformed by logarithm. The results of k-means clustering are demonstrated in Figs 1-4. As the results show, the operation volumes were divided into three groups separately. In addition to surgeon and hospital volumes and SSI, we collected patient-, surgeon-, and hospital-level data. Firstly, patient-level variables included age, gender, length of ICU stay, number of vessels obstructed that were involved in the surgical operation, and the presence of important underlying diseases (e.g. diabetes mellitus, chronic obstructive pulmonary disease (COPD), heart failure, renal failure and renal insufficiency, which were associated with SSI).
[13] Secondly, the surgeon-level variables included age and gender. Thirdly, the hospital-level variables included hospital ownership and geographic location.
All statistical analyses of volume-infection relationship were performed using SAS (version 9.2, SAS Institution Inc., Cary, NC, USA). In statistical testing, a two-sided p value 0.05 was considered statistically significant. The distributional properties of continuous variables were expressed by mean ± standard deviation (SD), whereas categorical variables were presented by frequency and percentage. In univariate analysis, the potential three-level predictors of SSI were examined using chi-square test or two-sample t-test as appropriate. Next, to account for the correlations within surgeon (level-2) and hospital (level-3), multivariate analysis was conducted by fitting mixed-effects logistic regression models to each patient's data for estimating the effects of three-level predictors on the probability of post-operational SSI. Furthermore, subgroup analysis for comorbidities was also conducted. Table 2 shows that there were 7,007 patients with CABG performed by 199 surgeons in 19 hospitals during 2006-2008 in Taiwan. The majority of patients were male (77.5%), and the mean age of patients was 65.3 years. The average ICU stay was 6.05 days, the mean level of number of vessels obstructed was around 1.6, while 51.8% of patients had diabetes mellitus, 33.3% had heart failure, 14.1% had renal failure and renal insufficiency, and 22.0% had COPD. Three hundred and two patients (4.31%) were identified as having the ICD-9-CM SSI codes. However, identification by the CART model only revealed 107 infection cases, and 94 cases were identified in both models. Most cases received CABG surgery by male surgeons, with a mean age of 45.0 years, and the surgeon's average operation volumes within the study period was 151.64, while the average operation volumes before surgery was 52.18. More than half of the cases were performed with CABG in not-for-profit hospitals, and the hospitals' average operation volumes within the study period was 473.60, while the average operation volumes before each surgery was 158.79. Moreover, most of patients received their surgeries by high-volume surgeons and hospitals, when k-means algorithm was used for categorization, regardless of which definition of operation volumes were used. Table 3 shows the results of multilevel mixed-effect models, with the SSIs being identified by ICD-9-CM codes, and the operation volumes defined as the cumulative volumes within the study period. The results of Model 1 (continuous) reveal that the surgeon's volumes were negatively associated with SSIs, while hospital's volumes were not associated with surgical site infection SSIs. Model 2 (quartile) suggests that low-volume surgeons had higher SSI risk (OR = 2.220, p-value = 0.022) than high-volume surgeons. There were also no associations between hospital's operation volumes and SSIs. Model 3 (k-means) shows that the association did not exist between hospital's/ surgeon's volumes and SSIs. Table 4 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volumes within the study period. Model 1 again indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results revealed low-volume surgeons had higher risk (OR = 1.691, p = 0.002) than high-volume surgeons. Table 5 displays the results of multilevel mixed-effect models, in which the SSIs were identified by ICD-9-CM codes, but the operation volumes were defined as the cumulative volume in the previous one year for each surgery. Model 1 also indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.642, p = 0.040) than high-volume surgeons. Table 6 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volume in previous one year for each surgery. In Model 1, different to the above findings, there was no association between hospital's/ surgeon's volumes and SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.163, p = 0.020) than high-volume surgeons.
We further examined the associations of surgeon and hospital volumes with SSIs in stratification analyses by underlying diseases. When the operation volumes were defined as the cumulative operation volume within the study period, no relationships existed between hospital/ surgeon operation volumes and SSIs. (Table 7 ) However, when the operation volumes were defined as the cumulative operation volumes in the previous one year for each surgery, the results suggested that there was a negative association between surgeon volumes and SSIs in the diabetes group, except that the volumes were treated as continuous variable and the infection cases were identified by ICD-9 codes. In terms of hospital operation volumes, the association did not exist. (Table 8 )
No studies have evaluated how different service/ operation volumes definitions and categorization methods affect volume-infection relationships. Moreover, several studies have pointed out the inappropriateness of identifying infection cases using the ICD-9-CM codes in claims data. Given these reasons, this study adopted two approaches to identifying SSIs, two definitions of operation volumes, and three methods for categorizing operation volumes to examine the relationships between operation volumes and SSIs. Our findings showed that the relationships between hospital volumes and SSIs did not exist, no matter which definitions, categorization mehods, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon volumes and SSIs were not robust in our data. It might be affected by different definitions and categorization methods of operation volumes, and also by different SSI cases identification approaches. In summary, most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons, and they also showed the risks were similar between medium-volume and high-volume surgeons. However, why did surgeon volume relate to SSIs, but hospital volume did not? Except for those issues we were concerned about in this study, there are some disagreements in the literature. Such as "Does provider volume really represent quality of care?" [12, 35] Or "Is provider volume the only one predictor for outcome of care?" [35, 36] These issues are worthy of further discussion, but are out of the scope of this study.
Service/ operation volumes are treated as a proxy indicator for experiences; previous studies used it to examine whether practice makes perfect or not. But, except for provider's experiences, SSIs are also impacted by many factors, such as environmental and clinical factors. Wu et al once used Taiwan 2001 NHI claims data to explore the relationship between provider CABG operation volumes and SSIs. [13] They found that hospital volumes had a greater effect than surgeon volumes and claimed that this may imply that hospital teamwork is more important than individual surgeon. However, our findings demonstrated that there was no relationship between hospital volumes and SSIs. Wu et al. adopted the cumulative operation volumes within the study period as the definition, and identified SSIs by ICD-9-CM codes. Except, there were two differences between our work and Wu et al., which were the length and year of the data; our data was longer and more updated than theirs. Moreover, it is worth noting that there was an outbreak of severe acute respiratory syndrome (SARS) in Taiwan in 2003, after which the hospital infection control system in Taiwan was reviewed and re-designed. Wu et al data was before SARS, so these efforts may also have improved the level of SSIs control in hospitals, leading to different findings in this study.
In addition, although most models revealed that there were negative relationships between surgeon's volumes and surgical site infection, the relationships were not robust. The results varied between different definitions and categorization method of operation volumes, and between SSIs identification approaches. Researchers need to consider how to identify SSIs correctly, how to choose optimal cut-off values, and how to decide on which definition is appropriate.
Finally, the results of stratification analyses showed that low-volume surgeon had higher risk than high-volume surgeon in the diabetes mellitus group, when the cumulative operation in the previous one year before surgery was used as definition. A large number of studies have indicated diabetes mellitus is associated with a higher risk of SSIs, [37] [38] [39] and the findings of this study suggest that CABG patients with diabetes mellitus should be cared for by experienced surgeons.
A multilevel analysis was applied to manage the nested factors, and two definitions of operation volume along with three different operation volume categorization methods were adopted to examine the relationship between volume and SSIs under two kinds of SSIs identification approaches. Nevertheless, the study suffered from several major limitations. First, the accuracy of SSIs identification was still an issue. Although the performance of the CART model to identify CABG SSIs was better than ICD-9-CM codes in Taiwan NHI claims data, it did not reach the perfect scenario. The accuracy of SSIs identification was still a challenge in our work. The second limitation relates to unmeasured variables, such as length of stay before operation, infection condition, hair removal, clinical information (e.g. blood glucose level, causative microorganism), time-related information (e.g. the duration of operation), the environment, surgical skills, use of post-operative drains, number of operations involved, and surgical site and wound care, etc. [40] Furthermore, information about type (elective or urgent) and incision site for surgery was not available in the Taiwan NHI claims data.
In conclusion, the findings of this study suggest that different definitions and categorization methods of operation volumes, and different SSIs identification approaches might lead to different findings, although surgeon volumes were more important than hospital volumes in exploring the relationships between CABG operation volumes and SSIs in Taiwan, but they were still not robust. Definitions and categorization methods of operation volumes, and correct identification of SSIs are important issues for future research.
|
Patients from how many medical centers were studied?
| 5,250 |
19
| 9,132 |
1,598 |
Which Kind of Provider’s Operation Volumes Matters? Associations between CABG Surgical Site Infection Risk and Hospital and Surgeon Operation Volumes among Medical Centers in Taiwan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4459823/
SHA: f3cbc0503581249a834895fc94cd3bae24714a0d
Authors: Yu, Tsung-Hsien; Tung, Yu-Chi; Chung, Kuo-Piao
Date: 2015-06-08
DOI: 10.1371/journal.pone.0129178
License: cc-by
Abstract: BACKGROUND: Volume-infection relationships have been examined for high-risk surgical procedures, but the conclusions remain controversial. The inconsistency might be due to inaccurate identification of cases of infection and different methods of categorizing service volumes. This study takes coronary artery bypass graft (CABG) surgical site infections (SSIs) as an example to examine whether a relationship exists between operation volumes and SSIs, when different SSIs case identification, definitions and categorization methods of operation volumes were implemented. METHODS: A population-based cross-sectional multilevel study was conducted. A total of 7,007 patients who received CABG surgery between 2006 and 2008 from19 medical centers in Taiwan were recruited. SSIs associated with CABG surgery were identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9 CM) codes and a Classification and Regression Trees (CART) model. Two definitions of surgeon and hospital operation volumes were used: (1) the cumulative CABG operation volumes within the study period; and (2) the cumulative CABG operation volumes in the previous one year before each CABG surgery. Operation volumes were further treated in three different ways: (1) a continuous variable; (2) a categorical variable based on the quartile; and (3) a data-driven categorical variable based on k-means clustering algorithm. Furthermore, subgroup analysis for comorbidities was also conducted. RESULTS: This study showed that hospital volumes were not significantly associated with SSIs, no matter which definitions or categorization methods of operation volume, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon’s volumes varied. Most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons. CONCLUSION: Surgeon volumes were more important than hospital volumes in exploring the relationship between CABG operation volumes and SSIs in Taiwan. However, the relationships were not robust. Definitions and categorization methods of operation volume and correct identification of SSIs are important issues for future research.
Text: data, which should use hierarchical models, may result in biased estimation of the variation and also lead to incorrect conclusions.
SSIs following coronary artery bypass graft (CABG) procedures place a heavy burden on patients and healthcare systems. The total length of stay and expenditure for patients with SSIs after CABG surgery is significantly longer and higher than those without SSIs. [20, 21] In 2008, the Centers for Medicare & Medicaid of the United States of America implemented the "Never Event" policy, where hospitals would no longer receive higher payments for the additional costs associated with treating patients for certain healthcare-acquired infections, including those related to CABG.
In view of the accuracy of SSIs identification and the heterogeneity of definition and categorization methods, no existing studies have used different infection case identification nor definitions and categorization methods of operation volume simultaneously to explore the relationship between operation volumes and infection. The current study takes CABG SSIs as an example to examine whether a relationship exists between operation volumes and SSIs, given different SSI cases identification, operation volume definitions and categorization methods.
This retrospective and cross-sectional study adopted a multilevel design to examine the relationships between provider volumes and SSIs after adjusting for patient-, surgeon-, and hospital-level covariates.
We used data from the Taiwan National Health Insurance Research Database (NHIRD) from 2005 and 2008. The NHIRD, published by the Taiwan National Health Research Institute, includes all the original claims data and registration files for beneficiaries enrolled under the National Health Insurance (NHI) program. The database covers the 23 million Taiwanese enrollees (approximately 98% of the population) in the NHI program. It is a de-identified secondary database containing patient-level demographic and administrative information; however, treatment items are aggregated and without time-related and clinical information. The data is released for research purposes.
The protocol for the study was approved by the Institutional Review Board of the National Taiwan University Hospital (protocol #201001027R). The dataset we used in this study was secondary data; all information was de-identified by data owners.
In this study, we adopted the ICD-9-CM SSI codes (hereafter referred to as the ICD-9-CM based model) and the Classification and Regression Trees (CART) model, which was developed in our previous work [11] to identify SSI cases. As we mentioned above, the ICD-9-CM SSI codes were the most popular tool to identify the SSI cases in claims data. In the ICD-9-CM based model, SSI cases were divided into two categories: index hospitalization events and post-discharge events (i.e., SSIs that occurred within 1 year after discharge and required readmission to a hospital and/ or the use of ambulatory services). Following Wu et al [13] , this study adopted the secondary ICD-9-CM diagnosis codes for index hospitalization events (ICD-9-CM code: 996.03, 996.61, 996.72, and 998.5), and the primary and secondary diagnosis codes for post-discharge events (ICD-9-CM code: 038.0-038. 4 ) as the criteria for SSI identification, in order to avoid cases in which infection existed prior to hospitalization. If a case had an index hospitalization event or a post-discharge event, then he/ she will be identified as SSIs by the ICD-9-CM based model. In the CART model, we adopted the type of antibiotics, dose of cefazolin, length of stay, and number of vessels obstructed (as a proxy indicator of duration of operation) as the parameters to identify the SSIs, according to our previous findings. [11] In our previous work, we used the 2005-2008 National Health Insurance claims data and healthcare-associated infection surveillance data from two medical centers for model development and model verification. Infection cases based on surveillance were identified by infection control personnel if the patient met the Taiwan CDC's criteria, which are the same as those adopted in the U.S. CDC. They manually review medical records of all patients at risk for the specified healthcare-associated infection.
The classification algorithms, the multivariable regression model, and the data mining model were adopted to develop alternative models based on surrogate indicators to identify cases of CABG SSIs and to compare the performance among these models and the ICD-9-CMbased model. For the classification algorithms, researchers build up several criteria, and if a case satisfies (or exceeds) a specific number of criteria, then it will be identified as a case of infection. For the multivariable regression model, researchers usually calculated a risk score by the logistic regression model, and the optimal cutoff point was determined according to the resulting receiver operating characteristic curve.
Concerning the data mining approach, which is widely used for predicting and classifying objects, the characteristics are: automatic discovery of patterns, prediction of likely outcomes, creation of actionable information, and focus on large data sets and databases. The classification and regression tree (CART) model, which is the most popular approach as applied in our work, and the growing, stopping, and pruning of the tree were determined by Gini improvement measures. [22, 23] After referring to the literature and conferring with infectious disease specialists, we adopted the following seven parameters: type of antibiotic, doses of antibiotic, doses of cefazolin, use of second-line antibiotics, length of stay, and number of vessels obstructed. Additionally, cross-validation was also employed, where data from one medical center was used for model development, and another one was used for model validation.
The results of our previous work revealed that the CART model offered better performance than that of the other identification models or the ICD-9-CM based model, especially in the positive predictive value (>70%), which was only found to be 20% in the ICD-9-CM based model. (Table 1 ) The findings also implied that the CART was a decidedly better tool for identifying cases of SSI in the Taiwan National Health Insurance database. Therefore, this study also adopted the CART model for identifying CABG SSIs.
To ensure homogeneity, current study analyzed 7,007 patients from 19 medical centers in Taiwan who underwent CABG surgery (ICD-9-CM procedure codes 36.1x-36.2x) between 2006 and 2008. CABG patients under the age of 18 years or over 85 years were excluded in this study. A total of 302 cases were identified as SSIs by ICD-9-CM based model, and a total of 107 cases were identified as SSIs by CART model.
In this study, we used the following two definitions to define operation volumes: (1) the cumulative operation volumes by each surgeon and hospital within the study period, which was the most common definition in the literature; and (2) following Yasunaga et al.'s study, [24] cumulative operation volumes by each surgeon and hospital in the previous one year for each surgery. However, our data was skewed, which did not follow a normal distribution. Therefore, we conducted the log transformations on operation volumes.
The current work treated operation volumes in three different ways: (1) a continuous variable; (2) a categorical variable based on the first and the third quartile as cutoff points (the most common method to categorize service/ operation volumes) [25] [26] [27] [28] ; and (3) a data-driven categorical variable based on k-means clustering algorithm. This study categorized surgeon and hospital volumes into low, medium, and high volume groups by quartile method and kmeans clustering algorithm.
In the quartile method, the cut-off value (transformed by logarithm) of the first quartile (<25%) for hospital volumes was 5.65, and the third quartile (>75%) was 6.43. In terms of surgeon volumes, the first quartile was 4.38, and the third was 5.35, when we used the cumulative operation volumes within the study period as the definition. While the definition changed, first quartile (<25%) for hospital volumes was 4.66, and the third quartile (>75%) was 5.31. In terms of surgeon volumes, the first quartile was 3.40, and the third was 4.32.
K-means clustering is an unsupervised machine-learning algorithm introduced by MacQueen in 1960s. This method is not only a simple and very reliable method in categorization/ classification, but is also recognized as one of the top 10 algorithms in data mining. [29] This method has often been applied in many fields. [30] [31] [32] Yu and his colleagues even applied it to define the quality of CABG care, and to explore the relationship among patient's income status, the level of quality of care, and inpatient mortality. [33] The main idea of this method is to partition observed data points into k non-overlapping clusters by minimizing the within-group sum of squares. Each point is assigned to the mean of its cluster using the Euclidian distance. Firstly, k cluster centers were randomly generated. Previous studies usually divided surgeons and hospitals into low-, medium-, and high-volume groups; therefore, we also predetermined the surgeon and hospital service volumes into 3 groups (k = 3). Then, participants were assigned to the cluster with the shortest distance to these cluster centers. Finally, the cluster centers were recomputed using the new cluster assignment and these steps would be iterated until convergence was achieved. [34] The cut-off values of hospital volumes were 5.21 and 5.69, and for surgeon's volumes were 2.40 and 4.38 respectively, when cumulative operation volumes within the study period was used as the definition. Likewise, when cumulative operation volumes before each surgery was used as definition, the cut-off values were 4.11 and 4.89 for hospital volumes, and 2.64 and 3.91 for surgeon's volumes. All cutoff values were transformed by logarithm. The results of k-means clustering are demonstrated in Figs 1-4. As the results show, the operation volumes were divided into three groups separately. In addition to surgeon and hospital volumes and SSI, we collected patient-, surgeon-, and hospital-level data. Firstly, patient-level variables included age, gender, length of ICU stay, number of vessels obstructed that were involved in the surgical operation, and the presence of important underlying diseases (e.g. diabetes mellitus, chronic obstructive pulmonary disease (COPD), heart failure, renal failure and renal insufficiency, which were associated with SSI).
[13] Secondly, the surgeon-level variables included age and gender. Thirdly, the hospital-level variables included hospital ownership and geographic location.
All statistical analyses of volume-infection relationship were performed using SAS (version 9.2, SAS Institution Inc., Cary, NC, USA). In statistical testing, a two-sided p value 0.05 was considered statistically significant. The distributional properties of continuous variables were expressed by mean ± standard deviation (SD), whereas categorical variables were presented by frequency and percentage. In univariate analysis, the potential three-level predictors of SSI were examined using chi-square test or two-sample t-test as appropriate. Next, to account for the correlations within surgeon (level-2) and hospital (level-3), multivariate analysis was conducted by fitting mixed-effects logistic regression models to each patient's data for estimating the effects of three-level predictors on the probability of post-operational SSI. Furthermore, subgroup analysis for comorbidities was also conducted. Table 2 shows that there were 7,007 patients with CABG performed by 199 surgeons in 19 hospitals during 2006-2008 in Taiwan. The majority of patients were male (77.5%), and the mean age of patients was 65.3 years. The average ICU stay was 6.05 days, the mean level of number of vessels obstructed was around 1.6, while 51.8% of patients had diabetes mellitus, 33.3% had heart failure, 14.1% had renal failure and renal insufficiency, and 22.0% had COPD. Three hundred and two patients (4.31%) were identified as having the ICD-9-CM SSI codes. However, identification by the CART model only revealed 107 infection cases, and 94 cases were identified in both models. Most cases received CABG surgery by male surgeons, with a mean age of 45.0 years, and the surgeon's average operation volumes within the study period was 151.64, while the average operation volumes before surgery was 52.18. More than half of the cases were performed with CABG in not-for-profit hospitals, and the hospitals' average operation volumes within the study period was 473.60, while the average operation volumes before each surgery was 158.79. Moreover, most of patients received their surgeries by high-volume surgeons and hospitals, when k-means algorithm was used for categorization, regardless of which definition of operation volumes were used. Table 3 shows the results of multilevel mixed-effect models, with the SSIs being identified by ICD-9-CM codes, and the operation volumes defined as the cumulative volumes within the study period. The results of Model 1 (continuous) reveal that the surgeon's volumes were negatively associated with SSIs, while hospital's volumes were not associated with surgical site infection SSIs. Model 2 (quartile) suggests that low-volume surgeons had higher SSI risk (OR = 2.220, p-value = 0.022) than high-volume surgeons. There were also no associations between hospital's operation volumes and SSIs. Model 3 (k-means) shows that the association did not exist between hospital's/ surgeon's volumes and SSIs. Table 4 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volumes within the study period. Model 1 again indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results revealed low-volume surgeons had higher risk (OR = 1.691, p = 0.002) than high-volume surgeons. Table 5 displays the results of multilevel mixed-effect models, in which the SSIs were identified by ICD-9-CM codes, but the operation volumes were defined as the cumulative volume in the previous one year for each surgery. Model 1 also indicated a negative association between surgeon's volumes and SSIs, and hospital's volumes were not found to be associated with SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.642, p = 0.040) than high-volume surgeons. Table 6 displays the results of multilevel mixed-effect models, in which the SSIs were identified by the CART model, and the operation volumes were also defined as the cumulative volume in previous one year for each surgery. In Model 1, different to the above findings, there was no association between hospital's/ surgeon's volumes and SSIs. In Model 2, the results showed that the relationship between hospital's/ surgeon's volumes and SSIs did not exist. In Model 3, results also revealed low-volume surgeons had higher risk (OR = 1.163, p = 0.020) than high-volume surgeons.
We further examined the associations of surgeon and hospital volumes with SSIs in stratification analyses by underlying diseases. When the operation volumes were defined as the cumulative operation volume within the study period, no relationships existed between hospital/ surgeon operation volumes and SSIs. (Table 7 ) However, when the operation volumes were defined as the cumulative operation volumes in the previous one year for each surgery, the results suggested that there was a negative association between surgeon volumes and SSIs in the diabetes group, except that the volumes were treated as continuous variable and the infection cases were identified by ICD-9 codes. In terms of hospital operation volumes, the association did not exist. (Table 8 )
No studies have evaluated how different service/ operation volumes definitions and categorization methods affect volume-infection relationships. Moreover, several studies have pointed out the inappropriateness of identifying infection cases using the ICD-9-CM codes in claims data. Given these reasons, this study adopted two approaches to identifying SSIs, two definitions of operation volumes, and three methods for categorizing operation volumes to examine the relationships between operation volumes and SSIs. Our findings showed that the relationships between hospital volumes and SSIs did not exist, no matter which definitions, categorization mehods, or SSIs case identification approaches were used. On the contrary, the relationships between surgeon volumes and SSIs were not robust in our data. It might be affected by different definitions and categorization methods of operation volumes, and also by different SSI cases identification approaches. In summary, most of the models demonstrated that the low-volume surgeons had higher risk than high-volume surgeons, and they also showed the risks were similar between medium-volume and high-volume surgeons. However, why did surgeon volume relate to SSIs, but hospital volume did not? Except for those issues we were concerned about in this study, there are some disagreements in the literature. Such as "Does provider volume really represent quality of care?" [12, 35] Or "Is provider volume the only one predictor for outcome of care?" [35, 36] These issues are worthy of further discussion, but are out of the scope of this study.
Service/ operation volumes are treated as a proxy indicator for experiences; previous studies used it to examine whether practice makes perfect or not. But, except for provider's experiences, SSIs are also impacted by many factors, such as environmental and clinical factors. Wu et al once used Taiwan 2001 NHI claims data to explore the relationship between provider CABG operation volumes and SSIs. [13] They found that hospital volumes had a greater effect than surgeon volumes and claimed that this may imply that hospital teamwork is more important than individual surgeon. However, our findings demonstrated that there was no relationship between hospital volumes and SSIs. Wu et al. adopted the cumulative operation volumes within the study period as the definition, and identified SSIs by ICD-9-CM codes. Except, there were two differences between our work and Wu et al., which were the length and year of the data; our data was longer and more updated than theirs. Moreover, it is worth noting that there was an outbreak of severe acute respiratory syndrome (SARS) in Taiwan in 2003, after which the hospital infection control system in Taiwan was reviewed and re-designed. Wu et al data was before SARS, so these efforts may also have improved the level of SSIs control in hospitals, leading to different findings in this study.
In addition, although most models revealed that there were negative relationships between surgeon's volumes and surgical site infection, the relationships were not robust. The results varied between different definitions and categorization method of operation volumes, and between SSIs identification approaches. Researchers need to consider how to identify SSIs correctly, how to choose optimal cut-off values, and how to decide on which definition is appropriate.
Finally, the results of stratification analyses showed that low-volume surgeon had higher risk than high-volume surgeon in the diabetes mellitus group, when the cumulative operation in the previous one year before surgery was used as definition. A large number of studies have indicated diabetes mellitus is associated with a higher risk of SSIs, [37] [38] [39] and the findings of this study suggest that CABG patients with diabetes mellitus should be cared for by experienced surgeons.
A multilevel analysis was applied to manage the nested factors, and two definitions of operation volume along with three different operation volume categorization methods were adopted to examine the relationship between volume and SSIs under two kinds of SSIs identification approaches. Nevertheless, the study suffered from several major limitations. First, the accuracy of SSIs identification was still an issue. Although the performance of the CART model to identify CABG SSIs was better than ICD-9-CM codes in Taiwan NHI claims data, it did not reach the perfect scenario. The accuracy of SSIs identification was still a challenge in our work. The second limitation relates to unmeasured variables, such as length of stay before operation, infection condition, hair removal, clinical information (e.g. blood glucose level, causative microorganism), time-related information (e.g. the duration of operation), the environment, surgical skills, use of post-operative drains, number of operations involved, and surgical site and wound care, etc. [40] Furthermore, information about type (elective or urgent) and incision site for surgery was not available in the Taiwan NHI claims data.
In conclusion, the findings of this study suggest that different definitions and categorization methods of operation volumes, and different SSIs identification approaches might lead to different findings, although surgeon volumes were more important than hospital volumes in exploring the relationships between CABG operation volumes and SSIs in Taiwan, but they were still not robust. Definitions and categorization methods of operation volumes, and correct identification of SSIs are important issues for future research.
|
Which patients were excluded from the study?
| 5,251 |
CABG patients under the age of 18 years or over 85 years
| 9,250 |
1,600 |
The influenza pandemic preparedness planning tool InfluSim
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1832202/
SHA: f3f471d10a36a7a28e9050c10bd4dfd680cba17b
Authors: Eichner, Martin; Schwehm, Markus; Duerr, Hans-Peter; Brockmann, Stefan O
Date: 2007-03-13
DOI: 10.1186/1471-2334-7-17
License: cc-by
Abstract: BACKGROUND: Planning public health responses against pandemic influenza relies on predictive models by which the impact of different intervention strategies can be evaluated. Research has to date rather focused on producing predictions for certain localities or under specific conditions, than on designing a publicly available planning tool which can be applied by public health administrations. Here, we provide such a tool which is reproducible by an explicitly formulated structure and designed to operate with an optimal combination of the competing requirements of precision, realism and generality. RESULTS: InfluSim is a deterministic compartment model based on a system of over 1,000 differential equations which extend the classic SEIR model by clinical and demographic parameters relevant for pandemic preparedness planning. It allows for producing time courses and cumulative numbers of influenza cases, outpatient visits, applied antiviral treatment doses, hospitalizations, deaths and work days lost due to sickness, all of which may be associated with economic aspects. The software is programmed in Java, operates platform independent and can be executed on regular desktop computers. CONCLUSION: InfluSim is an online available software which efficiently assists public health planners in designing optimal interventions against pandemic influenza. It can reproduce the infection dynamics of pandemic influenza like complex computer simulations while offering at the same time reproducibility, higher computational performance and better operability.
Text: Preparedness against pandemic influenza has become a high priority public health issue and many countries that have pandemic preparedness plans [1] . For the design of such plans, mathematical models and computer simulations play an essential role because they allow to predict and compare the effects of different intervention strategies [2] . The outstanding significance of the tools for purposes of intervention optimization is limited by the fact that they cannot maximize realism, generality and precision at the same time [3] . Public health planners, on the other hand, wish to have an optimal combination of these properties, because they need to formulate intervention strategies which can be generalized into recommendations, but are sufficiently realistic and precise to satisfy public health requirements.
Published influenza models which came into application, are represented by two extremes: generalized but oversimplified models without dynamic structure which are publicly available (e.g. [4] ), and complex computer simulations which are specifically adjusted to real conditions and/or are not publicly available (e.g. [5, 6] ). The complexity of the latter simulations, however, is not necessary for a reliable description of infection dynamics in large populations [7] . A minimum requirement for a pandemic influenza planning tool is a dynamic modelling structure which allows investigation of time-dependent variables like incidence, height of the epidemic peak, antiviral availability etc. The tool should, on the other hand, be adjustable to local conditions to adequately support the pandemic preparedness plans of different countries which involve considerably different assumptions (Table 1) .
Here we describe a publicly available influenza pandemic preparedness planning tool [8] which is designed to meet the requirements in preparedness planning. It is based on an explicitly formulated dynamic system which allows addressing time-dependent factors. It is sufficiently flexible to evaluate the impact of most candidate interventions and to consider local conditions like demographic and economic factors, contact patterns or constraints within the public health system. In subsequent papers we will also provide examples and applications of this model for various interventions, like antiviral treatment and social distancing measures.
The model is based on a system of 1,081 differential equations which extend the classic SEIR model. Demographic parameters reflect the situation in Germany in 2005, but can be adjusted to other countries. Epidemiologic and clinic values were taken from the literature (see Tables 1, 2 , 3, 4, 5, 6 and the sources quoted there). Pre-set values can be varied by sliders and input fields to make different assumptions on the transmissibility and clinical severity of a new pandemic strain, to change the costs connected to medical treatment or work loss, or to simply apply the simulation to different demographic settings. Model properties can be summarized as follows. The mathematical formulation of this model is presented in detail in the online supporting material. The corresponding source code, programmed in Java, and further information can be downloaded from [8] .
According to the German National Pandemic Preparedness Plan [9] , the total population is divided in age classes, each of which is subdivided into individuals of low and high risk ( Table 2) . Transmission between these age classes is based on a contact matrix (Table 3) which is scaled such that the model with standard parameter values yields a given basic reproduction number R0. Values for the R0 associated with an influenza strain with pandemic potential are suggested to lie between 2 and 3 [10] . This value is higher than the effective reproduction number which has been estimated to be slightly lower than 2 [11, 12] . As a standard parameter, we use R0 = 2.5 which means that cases infect on average 2.5 individuals if everybody is susceptible and if no interventions are performed.
Susceptible individuals who become infected, incubate the infection, then become fully contagious and finally develop protective immunity (Table 4) . A fraction of cases remains asymptomatic; others become moderately sick or clinically ill (i.e. they need medical help). Depending on the combination of age and risk group, a fraction of the clinically ill cases needs to be hospitalized, and an agedependent fraction of hospitalized cases may die from the disease ( Table 5 ). This partitioning of the cases into four categories allows combining the realistic description of the transmission dynamics with an easy calculation of the resources consumed during an outbreak. The degree and duration of contagiousness of a patient depend on the course of the disease; the latter furthermore depends on the age of the patient (Table 5) . Passing through the incubation and contagious period is modelled in several stages which allows for realistic distributions of the sojourn times ( Table 4 ). The last two stages of the incubation period are used as early infectious period during which the patient can already spread the disease. Infectiousness is highest after onset of symptoms and thereafter declines geometrically (Table 6 ). Clinically ill patients seek medical help on average one day after onset of symptoms. Very sick patients are advised to withdraw to their home until their disease is over, whereas extremely sick patients need to be hospitalized and may die from the disease (Table 4) . After the end of their contagious period, clinically ill patients go through a convalescent period before they can resume their ordinary life and go back to work (Table 4) .
We provide some examples of model output of InfluSim [8] , version 2.0, by means of four sensitivity analyses; further investigations will be presented elsewhere. Figure 1 shows the graphical user interface of the software which is divided into input and output windows. The user may set new values in the input fields or move sliders to almost simultaneously obtain new results for the course of an epidemic in a given population. Figures 2A and 2B show pandemic waves which result from varying the basic reproduction number from 1.5 to 4.0. Using the standard parameter values as given in Tables 2, 3 , 4, 5, 6 and omitting all interventions in a town of 100,000 inhabitants results in a pandemic wave which lasts for about ten weeks (Figure 2A , with R 0 = 2.5). The peak of the pandemic wave is reached after six to seven weeks, with a daily incidence of up to 2,340 influenza patients seeking medical help, with up to 280 hospital beds occupied by influenza cases and with up to 14,000 out of 60,000 working adults unable to go to work because of illness or convalescence. These results depend on the assumptions concerning the yet unknown contagiousness and pathogenicity of the virus. Figures 2C and 2D show how the shape of the curves depends on the course of contagiousness: the pandemic wave proceeds relative slowly if the contagiousness does not change during the infectious period (x 50 = 50%), but proceeds quickly if the contagiousness is highest after onset of symptoms and decreases thereafter (x 50 > 50%).
The influenza pandemic preparedness planning tool InfluSim stands between simple spreadsheet models and sophisticated stochastic computer simulations. It describes a pandemic wave within a homogeneously mixing population like a town or city, but surprisingly produces the same dynamics as individual-based simulations which explicitly consider geographic spread through the US (cf. [6] and [5] with Figure 2 using R 0 = 2). Similar observations were made with a simple deterministic compartmental model [7] . Stochastic models are known to behave quasi-deterministically when the simulated population becomes very large.
A further reason for the congruence of complex stochastic and simple deterministic models must lie in the incredi-bly quick way in which pandemic influenza spreads geographically. Unless being controlled at the place of origin [12, 13] , a pandemic starting in a far-off country will lead to multiple introductions [14] into the large industrialized nations where it can be expected to quickly spread to neighbouring towns and to rural areas. The large populations which have to be considered susceptible to a pandemic virus and the quick geographic spread tend to diminish the differences between the results of sophisticated individual-based and simple deterministic models.
However, a deterministic model like InfluSim cannot reliably represent effects originating from stochasticity, from effects in small populations, or from heterogeneities. Examples are: (i) a geographically limited spread and fairly effective control measures can imply that the epidemic affects only a small population and thus, may be strongly influenced by stochastic events [15] [16] [17] ; (ii) transmission which predominantly occurs in households or hospitals, or which is driven by other substantial features of the contact network is not in agreement with the assumption of homogeneous mixing in the deterministic model cannot reliably predict the spread of infection [18] [19] [20] [21] [22] [23] . In particular, (iii) super-spreading events can substantially change the course of an epidemic compared to the deterministic prediction [24] [25] [26] [27] . Apart from such factors, the predictability of intervention success is generally subject to uncertainties in the choice of parameter values, Assumed scenarios and outcomes of pandemic preparedness plans. * Gross attack rate (i.e. clinically ill and moderately ill cases). A population of N = 100,000 inhabitants of Germany is subdivided according to age a and risk category r. We assume that all age groups are fully susceptible at begin of the outbreak. A fraction of F a = 6% of all children (age < 20 years) are regarded as being under high risk (r = r 1 ) after an influenza infection whereby the remaining 94% are under low risk (r = r 2 ). The high risk fractions of working adults (ages 20-59) and elderly (ages 60+) are F a = 14% and F a = 47%, respectively. Source: [9] demanding additional efforts like Bayesian approaches [28] to evaluate the reliability of predictions [29] .
Pandemic preparedness plans must consider constraints and capacities of locally operating public health systems. The time-dependent solutions of InfluSim allow assessing peak values of the relevant variables, such as outpatients, hospitalizations and deaths. Various interventions may be combined to find optimal ways to reduce the total number of cases, to lower the peak values or to delay the peak, hoping that at least part of the population may benefit from a newly developed vaccine.
Special care was taken when implementing a variety of pharmaceutical and non-pharmaceutical interventions which will be discussed in subsequent papers. Despite its comprehensible structure, the model does not suffer from over-simplifications common to usual compartment models. Instead of implicitly using exponentially distributed sojourn times, we have implemented realistically distributed delays. For example, the model considers that individuals may transmit infection before onset of symptoms, and that some cases may remain asymptomatic, but still infecting others. Such features have serious implications for the success of targeted control measures.
InfluSim is freely accessible, runs on a regular desktop computer and produces results within a second after changing parameter values. The user-friendly interface and the ease at which results can be generated make this program a useful public health planning tool. Although we have taken care of providing a bug-free program, including the source code, the user is encouraged to treat results with due caution, to test it, and to participate in bug-reports and discussions on the open-source platform [30] which also provides regular updates of InfluSim.
The author(s) declare that they have no competing interests.
ME developed the model, MS designed the software, HPD wrote the manuscript and SOB formulated the public The who-acquires-infection-from-whom matrix shows the frequency of contacts (per week per person) between different age classes.
Source: [38] . Distribution of sojourn times (the last two stages of the latent period are used as early infectious period with an average duration of D L = 0.5 days). Sources: A [11] , B [39, 40] , C assumed, D [41] health requirements of the software. All authors read and approved the final manuscript.
Susceptible individuals S a, r are infected at a rate λ a (t) which depends on their age a and on time t. Infected individuals, E a, r , incubate the infection for a mean duration D E . To obtain a realistic distribution of this duration, the incubation period is modelled in n stages so that progression from one stage to the next one occurs at rate δ = n/D E .
The last l incubation stages are regarded as early infectious period during which patients may already spread the infection (this accounts for an average time of lD E /n for the "early infectious period" which is about half a day for the standard set of parameters). After passing through the last incubation stage, infected individuals become fully contagious and a fraction of them develops clinical symptoms. The course of disease depends on the age a of the infected individual and on the risk category r to which he or she belongs: a fraction c a, r (A) becomes asymptomatic (A a ), a fraction c a, r (M) becomes moderately sick (M a ), a fraction c a, r (V) becomes very sick (V a ) and the remaining fraction c a, r (X) becomes extremely sick (X a ) and need hospitalization (i.e., c a, r (A) + c a, r (M) + c a, r (V) + c a, r (X) = 1 for each combination of a and r). ) . A fraction f V (t) of all severe and a fraction f X (t) of all extremely severe cases who visit the doctor within D T days after onset of symptoms are offered antiviral treatment, given that its supply has not yet been exhausted. As our model does not explicitly consider the age of the disease (which would demand partial differential equations), we use the contagious stages to measure time since onset and allow for treatment up to stage m a, T Sources: Contagiousness of asymptomatic cases: [11] ; degree of contagiousness during the early infectious period and equality of the contagiousness of moderately and severely sick cases: assumed. Independent of age a and risk group r, a fraction c a, r (A) = 33% of infections result in asymptomatic cases, a fraction c a, r (M) = 33.5% become moderately sick and the remaining fraction develops severe disease. An age-and risk-dependent fraction h a, r of untreated patients with severe disease needs hospitalization. An age-dependent fraction d a of hospitalized cases dies. Sources: fraction of asymptomatic cases: [11] ; 50% of symptomatic cases see a doctor: [9] ; hospitalizations per severe case: [9] ; case fatality of hospitalized, but untreated patients calculated from [4] .
(see below for details). This imposes some variability to the maximum time until which treatment can be given, which may even improve the realism of the model with respect to real-life scenarios. Antiviral treatment reduces the patients' contagiousness by f I percent and it reduces hospitalization and death by f H percent. Extremely sick patients, whose hospitalization is prevented by treatment, are sent home and join the group of treated very sick patients(W a, T ). The remaining duration of disease and contagiousness of treated cases is reduced by f D percent so that their rate of progressing from one stage to the next has To obtain a realistic distribution of this sojourn time, convalescence is modelled in j stages so that progression from one stage to the next occurs at rate ρ = j/D C . Fully recovered patients who have passed through their last stage of convalescence join the group of healthy immunes I; working adults will go back to work. Further interventions, describing the reduction of contacts, will be discussed after the presentation of the differential equations.
InfluSim user interface Figure 1 InfluSim user interface. x 50 = 95% means that 95% of the cumulative contagiousness is concentrated during the first half of the contagious period, see Table 6 ). D: Cumulative number of deaths for values of x 50 as in C. All other parameters as listed in Tables 2-6 . Hospitalized, but untreated cases
Contact matrix For the mixing of the age classes, we employ a whoacquires-infection-from whom matrix which gives the relative frequency of contacts of infective individuals of age a i with other people of age a s . In this paper, we assume bi-directional contacts (e.g. children have the same total number of contacts with adults as adults with children). Multiplication of this matrix with an appropriate constant scaling factor κ (see below)
results in the matrix of crude contact rates .
In the absence of interventions, we have to multiply these contact rates with the contagiousness factors b L , b A , b M and b V to obtain the effective contact rates:
during the early infectious period, of asymptomatic cases, of moderately sick cases, of (untreated) very sick cases.
To assess the effect of day care centre and school closing on the transmission of an infectious disease, we have to first make an assumption on what fraction r sch of the contacts among healthy children who are in the same age class occurs in day care centres and schools. The contact rates between very sick or hospitalized children (who do not attend day care centre or school) and other children need, therefore, be reduced to (contact rate between healthy and very sick children in the same age class, i.e. a i = a s ).
As very sick children have to be taken care of by adults at home or in hospital, their contact rate to adults increases by a factor F HC (contact rate between very sick children of age a i and adults of age a s ).
Contacts between very sick children and other children in a higher or lower age class remain unchanged: (contact rate between healthy children of age a s and very sick children of a different age a i ).
Closing day care centres and schools at time t will not necessarily prevent all the contacts that would have happened with other children. During the closing of schools and day care centres, the contact rates between susceptible children of age a s and infected children of age a i who are in their late incubation period ( ), who are asymptomatic ( ), or who are moderately sick ( ) are reduced by the factor r sch if the children are in the same age class:
where 1 sch (t) is a function which indicates when schools and day care centres are opened or closed:
,..., While day care centres and schools are closed, children (age a i ) need adult supervision at home. Their contact with susceptible adults (age a s ) increases by the "child care factor" F CC :
Child care at home also increases the exposure of healthy children (age a s ) to contagious adults (age a i ):
Cancelling mass gathering events effects only the contacts of adults who are healthy enough to attend such events. Assuming that such an intervention at time t reduces contacts by a fraction r mass , we get for all contacts between susceptible adults of age a s and infectious adults of age a i the following contact rates:
where 1 mass (t) is a function which indicates when mass gathering events are possible or when they are closed:
As contacts with adults who are too sick to attend such mass gathering events cannot be prevented by this measure it is .
During some time in the epidemic, the general population may effectively reduce contacts which can be a result of wearing facial masks, increasing "social distance", adopting improved measures of "respiratory hygiene" or simply of a general change in behaviour. This will be implemented in the program by reducing the contacts of susceptible individuals at that time t by factor r gen (t
while mass gathering events are forbidden while m mass gathering events are allowed. while the population reduces their contacts while the population behaves as usual.
The contact rates of cases in the late incubation period and that of asymptomatic cases remain unchanged:
for infected individuals in the late incubation period, for asymptomatic cases.
To allow for a contagiousness which changes over the course of disease, we multiply each contact rate with a weighting factor whereby k is the stage of contagiousness. This leads to the following contact rates:
for asymptomatic cases in For x = 1, contagiousness is equally high in all stages; for x = 0, only the first stage is contagious; for 0 <x < 1, the contagiousness decreases in a geometric procession. We make the simplifying assumption that contagiousness does not change during the late incubation period for cases in stage k = n -l,..,n of the incubation period.
At time t = 0 and in the absence of interventions, the next generation matrix has the following elements where is the fraction of untreated extremely severe cases who die from the disease (see below for details). The dominant eigenvalue of this matrix is called the basic reproduction number R 0 . If κ (which determines the value of the contact rates ) is given, the eigenvectors of this matrix can numerically be calculated. The user-specified value of R 0 is now used to determine numerically the scaling factor κ. Let be the eigenvector which has the largest eigenvalue R 0 .
) ) −
Using the user-specified numbers of people N a in the age classes and the fractions F a of people under high risk within each age class (Table 2) , we obtain the initial population sizes according to age and risk class: Using these initial values, the set of differential equations is solved numerically with a Runge-Kutta method with step-size control.
if and in treatment window otherwise 0 ⎩ ⎩
|
What are the limitations of a deterministic model?
| 5,255 |
cannot reliably represent effects originating from stochasticity, from effects in small populations, or from heterogeneities
| 10,476 |
1,603 |
Genome Sequences of Porcine Epidemic Diarrhea Virus: In Vivo and In Vitro Phenotypes
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4056290/
SHA: f6d6d7efc1686a7d219ecfc55f9a48ce72d4fb00
Authors: Lawrence, Paulraj K.; Bumgardner, Eric; Bey, Russell F.; Stine, Douglas; Bumgarner, Roger E.
Date: 2014-06-12
DOI: 10.1128/genomea.00503-14
License: cc-by
Abstract: Since the outbreak of porcine epidemic diarrhea virus (PEDV) in May 2013, U.S. swine producers have lost almost five million baby pigs. In an attempt to understand the evolution of PEDV in the United States and possibly develop a control strategy, we compared the genome sequences of a PEDV strain isolated from an infected piglet against its in vitro adapted version. The original PEDV strain was grown in Vero cells and passed 10 times serially in a MARC145 cell line. The sequence analysis of the native PEDV strain and in vitro passaged virus shows that the cell culture adaptation specifically modifies PEDV spike protein whereas the open reading frame 1a/b (ORF1a/b)-encoded polyprotein, the nucleoprotein, NS3B (ORF3), and membrane and envelope proteins remain unchanged.
Text: highly contagious swine disease. While older pigs have a chance of survival, 80 to 100 percent of PEDV-infected piglets die within 24 h of being infected. PEDV spreads primarily through fecal-oral contact (1, 2) . Once the virus is internalized, it destroys the lining of piglets' intestines, making them incapable of digesting and deriving nutrition from milk and feed (1) . The virus causes diarrhea, vomiting, and death from dehydration and starvation (2) .
PEDV is a member of the Coronavirinae subfamily and belongs to the Alphacoronavirus genus. Its genomic size ranges from approximately 26 to 32 kb, which is relatively large for an RNA virus. Although vaccines for PEDV exist in China, Japan, and South Korea, there is no approved vaccine in the United States or Europe (3) . Furthermore, PEDV is still evolving within the U.S. swine population.
This report briefly describes the comparison of genome sequences of a PEDV strain isolated from small intestine samples of an infected piglet and its in vitro adapted version. The original PEDV strain was dubbed NPL-PEDV/2013, grown in Vero cells, and passed 10 times in a MARC145 cell line. The serial in vitro passage strain was named NPL-PEDV/2013/P10. The total viral RNA was extracted by TRIzol LS reagent and sequenced by Sanger dideoxy sequencing using a primer walking technique. The raw sequences were imported into the Geneious assembler (Biomatters, CA), assembled, annotated, and compared against each other using USA/Colorado/2013 (GenBank accession no. KF272920) as a reference sequence.
The whole-genome sequences of NPL-PEDV/2013 and NPL-PEDV/2013/P10 contain 28,038 and 28,025 nucleotides (nt), respectively, including the 5= and 3= untranslated regions (UTR). The NPL-PEDV/2013 genome shares 99% identity with all the U.S. isolates sequenced to date and many Chinese isolates as well. The top three BLAST hits were against U.S. isolates, USA/Colora-do/2013 (GenBank accession no. KF272920), IA1 (GenBank accession no. KF468753.1), and an isolate from Iowa, 13-019349 (GenBank accession no. KF267450.1). The NPL-PEDV/2013 isolate also shares 99% identity with the Chinese outbreak isolate AH2012 (GenBank accession no. KC210145).
When the NPL-PEDV/2013/P10 strain was compared against NPL-PEDV/2013 , the open reading frame 1a/b (ORF1a/b) polyprotein, the nucleoprotein, NS3B, and membrane and envelope proteins were found to be 100% identical at the amino acid level. In contrast, the spike gene contains six nonsynonymous single nucleotide polymorphisms, resulting in amino acid (aa) substitutions in the following positions: 375 (F¡L), 486 (T¡P), 856 (D¡E), 1081 (A¡V), 1099 (A¡S), and 1253 (Y¡D). The S1 domain of spike protein contains 2 aa substitutions, whereas the S2 domain contains 4 aa substitutions. PEDV has been shown to use porcine aminopeptidase N (pAPN) as the major receptor for cell entry (4, 5) . However, Vero and MARC145 cells lack pAPN, clearly indicating that other receptors or receptor-independent pathways may be used for entry (6) . The spike protein in its trimeric conformation interacts with the cell receptor and contains numerous neutralizing antibody binding epitopes (7) . Analysis of the spike by PeptideCutter (http://web.expasy.org/ peptide_cutter/) shows that the native spike protein of NPL-PEDV/2013 has 63 trypsin and 2 chymotrypsin cleavage sites at 100% efficiency whereas NPL-PEDV/2013/P10 has lost one trypsin cleavage site but the number of chymotrypsin sites remain unchanged. This indicates that cell culture adaptation specifically modifies the PEDV spike protein; however, the immunological implications are unknown.
Nucleotide sequence accession numbers. The whole-genome sequences of the NPL-PEDV/2013 and NPL-PEDV/2013/P10 strains have been deposited at DDBJ/EMBL/GenBank under accession no. KJ778615 and KJ778616.
|
How does PEDV spread?
| 5,266 |
fecal-oral contact
| 1,335 |
1,603 |
Genome Sequences of Porcine Epidemic Diarrhea Virus: In Vivo and In Vitro Phenotypes
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4056290/
SHA: f6d6d7efc1686a7d219ecfc55f9a48ce72d4fb00
Authors: Lawrence, Paulraj K.; Bumgardner, Eric; Bey, Russell F.; Stine, Douglas; Bumgarner, Roger E.
Date: 2014-06-12
DOI: 10.1128/genomea.00503-14
License: cc-by
Abstract: Since the outbreak of porcine epidemic diarrhea virus (PEDV) in May 2013, U.S. swine producers have lost almost five million baby pigs. In an attempt to understand the evolution of PEDV in the United States and possibly develop a control strategy, we compared the genome sequences of a PEDV strain isolated from an infected piglet against its in vitro adapted version. The original PEDV strain was grown in Vero cells and passed 10 times serially in a MARC145 cell line. The sequence analysis of the native PEDV strain and in vitro passaged virus shows that the cell culture adaptation specifically modifies PEDV spike protein whereas the open reading frame 1a/b (ORF1a/b)-encoded polyprotein, the nucleoprotein, NS3B (ORF3), and membrane and envelope proteins remain unchanged.
Text: highly contagious swine disease. While older pigs have a chance of survival, 80 to 100 percent of PEDV-infected piglets die within 24 h of being infected. PEDV spreads primarily through fecal-oral contact (1, 2) . Once the virus is internalized, it destroys the lining of piglets' intestines, making them incapable of digesting and deriving nutrition from milk and feed (1) . The virus causes diarrhea, vomiting, and death from dehydration and starvation (2) .
PEDV is a member of the Coronavirinae subfamily and belongs to the Alphacoronavirus genus. Its genomic size ranges from approximately 26 to 32 kb, which is relatively large for an RNA virus. Although vaccines for PEDV exist in China, Japan, and South Korea, there is no approved vaccine in the United States or Europe (3) . Furthermore, PEDV is still evolving within the U.S. swine population.
This report briefly describes the comparison of genome sequences of a PEDV strain isolated from small intestine samples of an infected piglet and its in vitro adapted version. The original PEDV strain was dubbed NPL-PEDV/2013, grown in Vero cells, and passed 10 times in a MARC145 cell line. The serial in vitro passage strain was named NPL-PEDV/2013/P10. The total viral RNA was extracted by TRIzol LS reagent and sequenced by Sanger dideoxy sequencing using a primer walking technique. The raw sequences were imported into the Geneious assembler (Biomatters, CA), assembled, annotated, and compared against each other using USA/Colorado/2013 (GenBank accession no. KF272920) as a reference sequence.
The whole-genome sequences of NPL-PEDV/2013 and NPL-PEDV/2013/P10 contain 28,038 and 28,025 nucleotides (nt), respectively, including the 5= and 3= untranslated regions (UTR). The NPL-PEDV/2013 genome shares 99% identity with all the U.S. isolates sequenced to date and many Chinese isolates as well. The top three BLAST hits were against U.S. isolates, USA/Colora-do/2013 (GenBank accession no. KF272920), IA1 (GenBank accession no. KF468753.1), and an isolate from Iowa, 13-019349 (GenBank accession no. KF267450.1). The NPL-PEDV/2013 isolate also shares 99% identity with the Chinese outbreak isolate AH2012 (GenBank accession no. KC210145).
When the NPL-PEDV/2013/P10 strain was compared against NPL-PEDV/2013 , the open reading frame 1a/b (ORF1a/b) polyprotein, the nucleoprotein, NS3B, and membrane and envelope proteins were found to be 100% identical at the amino acid level. In contrast, the spike gene contains six nonsynonymous single nucleotide polymorphisms, resulting in amino acid (aa) substitutions in the following positions: 375 (F¡L), 486 (T¡P), 856 (D¡E), 1081 (A¡V), 1099 (A¡S), and 1253 (Y¡D). The S1 domain of spike protein contains 2 aa substitutions, whereas the S2 domain contains 4 aa substitutions. PEDV has been shown to use porcine aminopeptidase N (pAPN) as the major receptor for cell entry (4, 5) . However, Vero and MARC145 cells lack pAPN, clearly indicating that other receptors or receptor-independent pathways may be used for entry (6) . The spike protein in its trimeric conformation interacts with the cell receptor and contains numerous neutralizing antibody binding epitopes (7) . Analysis of the spike by PeptideCutter (http://web.expasy.org/ peptide_cutter/) shows that the native spike protein of NPL-PEDV/2013 has 63 trypsin and 2 chymotrypsin cleavage sites at 100% efficiency whereas NPL-PEDV/2013/P10 has lost one trypsin cleavage site but the number of chymotrypsin sites remain unchanged. This indicates that cell culture adaptation specifically modifies the PEDV spike protein; however, the immunological implications are unknown.
Nucleotide sequence accession numbers. The whole-genome sequences of the NPL-PEDV/2013 and NPL-PEDV/2013/P10 strains have been deposited at DDBJ/EMBL/GenBank under accession no. KJ778615 and KJ778616.
|
How does PEDV cause illness?
| 5,267 |
destroys the lining of piglets' intestines
| 1,398 |
1,603 |
Genome Sequences of Porcine Epidemic Diarrhea Virus: In Vivo and In Vitro Phenotypes
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4056290/
SHA: f6d6d7efc1686a7d219ecfc55f9a48ce72d4fb00
Authors: Lawrence, Paulraj K.; Bumgardner, Eric; Bey, Russell F.; Stine, Douglas; Bumgarner, Roger E.
Date: 2014-06-12
DOI: 10.1128/genomea.00503-14
License: cc-by
Abstract: Since the outbreak of porcine epidemic diarrhea virus (PEDV) in May 2013, U.S. swine producers have lost almost five million baby pigs. In an attempt to understand the evolution of PEDV in the United States and possibly develop a control strategy, we compared the genome sequences of a PEDV strain isolated from an infected piglet against its in vitro adapted version. The original PEDV strain was grown in Vero cells and passed 10 times serially in a MARC145 cell line. The sequence analysis of the native PEDV strain and in vitro passaged virus shows that the cell culture adaptation specifically modifies PEDV spike protein whereas the open reading frame 1a/b (ORF1a/b)-encoded polyprotein, the nucleoprotein, NS3B (ORF3), and membrane and envelope proteins remain unchanged.
Text: highly contagious swine disease. While older pigs have a chance of survival, 80 to 100 percent of PEDV-infected piglets die within 24 h of being infected. PEDV spreads primarily through fecal-oral contact (1, 2) . Once the virus is internalized, it destroys the lining of piglets' intestines, making them incapable of digesting and deriving nutrition from milk and feed (1) . The virus causes diarrhea, vomiting, and death from dehydration and starvation (2) .
PEDV is a member of the Coronavirinae subfamily and belongs to the Alphacoronavirus genus. Its genomic size ranges from approximately 26 to 32 kb, which is relatively large for an RNA virus. Although vaccines for PEDV exist in China, Japan, and South Korea, there is no approved vaccine in the United States or Europe (3) . Furthermore, PEDV is still evolving within the U.S. swine population.
This report briefly describes the comparison of genome sequences of a PEDV strain isolated from small intestine samples of an infected piglet and its in vitro adapted version. The original PEDV strain was dubbed NPL-PEDV/2013, grown in Vero cells, and passed 10 times in a MARC145 cell line. The serial in vitro passage strain was named NPL-PEDV/2013/P10. The total viral RNA was extracted by TRIzol LS reagent and sequenced by Sanger dideoxy sequencing using a primer walking technique. The raw sequences were imported into the Geneious assembler (Biomatters, CA), assembled, annotated, and compared against each other using USA/Colorado/2013 (GenBank accession no. KF272920) as a reference sequence.
The whole-genome sequences of NPL-PEDV/2013 and NPL-PEDV/2013/P10 contain 28,038 and 28,025 nucleotides (nt), respectively, including the 5= and 3= untranslated regions (UTR). The NPL-PEDV/2013 genome shares 99% identity with all the U.S. isolates sequenced to date and many Chinese isolates as well. The top three BLAST hits were against U.S. isolates, USA/Colora-do/2013 (GenBank accession no. KF272920), IA1 (GenBank accession no. KF468753.1), and an isolate from Iowa, 13-019349 (GenBank accession no. KF267450.1). The NPL-PEDV/2013 isolate also shares 99% identity with the Chinese outbreak isolate AH2012 (GenBank accession no. KC210145).
When the NPL-PEDV/2013/P10 strain was compared against NPL-PEDV/2013 , the open reading frame 1a/b (ORF1a/b) polyprotein, the nucleoprotein, NS3B, and membrane and envelope proteins were found to be 100% identical at the amino acid level. In contrast, the spike gene contains six nonsynonymous single nucleotide polymorphisms, resulting in amino acid (aa) substitutions in the following positions: 375 (F¡L), 486 (T¡P), 856 (D¡E), 1081 (A¡V), 1099 (A¡S), and 1253 (Y¡D). The S1 domain of spike protein contains 2 aa substitutions, whereas the S2 domain contains 4 aa substitutions. PEDV has been shown to use porcine aminopeptidase N (pAPN) as the major receptor for cell entry (4, 5) . However, Vero and MARC145 cells lack pAPN, clearly indicating that other receptors or receptor-independent pathways may be used for entry (6) . The spike protein in its trimeric conformation interacts with the cell receptor and contains numerous neutralizing antibody binding epitopes (7) . Analysis of the spike by PeptideCutter (http://web.expasy.org/ peptide_cutter/) shows that the native spike protein of NPL-PEDV/2013 has 63 trypsin and 2 chymotrypsin cleavage sites at 100% efficiency whereas NPL-PEDV/2013/P10 has lost one trypsin cleavage site but the number of chymotrypsin sites remain unchanged. This indicates that cell culture adaptation specifically modifies the PEDV spike protein; however, the immunological implications are unknown.
Nucleotide sequence accession numbers. The whole-genome sequences of the NPL-PEDV/2013 and NPL-PEDV/2013/P10 strains have been deposited at DDBJ/EMBL/GenBank under accession no. KJ778615 and KJ778616.
|
What is the size of the PEDV genome?
| 5,268 |
26 to 32 kb
| 1,745 |
1,606 |
Serological Assays Based on Recombinant Viral Proteins for the Diagnosis of Arenavirus Hemorrhagic Fevers
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3497043/
SHA: f1d308db379b3c293bcfc8fe251c043fe8842358
Authors: Fukushi, Shuetsu; Tani, Hideki; Yoshikawa, Tomoki; Saijo, Masayuki; Morikawa, Shigeru
Date: 2012-10-12
DOI: 10.3390/v4102097
License: cc-by
Abstract: The family Arenaviridae, genus Arenavirus, consists of two phylogenetically independent groups: Old World (OW) and New World (NW) complexes. The Lassa and Lujo viruses in the OW complex and the Guanarito, Junin, Machupo, Sabia, and Chapare viruses in the NW complex cause viral hemorrhagic fever (VHF) in humans, leading to serious public health concerns. These viruses are also considered potential bioterrorism agents. Therefore, it is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of arenavirus outbreaks. However, these arenaviruses are classified as BSL-4 pathogens, thus making it difficult to develop diagnostic techniques for these virus infections in institutes without BSL-4 facilities. To overcome these difficulties, antibody detection systems in the form of an enzyme-linked immunosorbent assay (ELISA) and an indirect immunofluorescence assay were developed using recombinant nucleoproteins (rNPs) derived from these viruses. Furthermore, several antigen-detection assays were developed. For example, novel monoclonal antibodies (mAbs) to the rNPs of Lassa and Junin viruses were generated. Sandwich antigen-capture (Ag-capture) ELISAs using these mAbs as capture antibodies were developed and confirmed to be sensitive and specific for detecting the respective arenavirus NPs. These rNP-based assays were proposed to be useful not only for an etiological diagnosis of VHFs, but also for seroepidemiological studies on VHFs. We recently developed arenavirus neutralization assays using vesicular stomatitis virus (VSV)-based pseudotypes bearing arenavirus recombinant glycoproteins. The goal of this article is to review the recent advances in developing laboratory diagnostic assays based on recombinant viral proteins for the diagnosis of VHFs and epidemiological studies on the VHFs caused by arenaviruses.
Text: The virus family Arenaviridae consists of only one genus, but most viruses within this genus can be divided into two different groups: the Old World arenaviruses and the New World arenaviruses (also known as the Tacaribe complex) [1, 2] . The differences between the two groups have been established through the use of serological assays. Most of the arenaviruses cause persistent infection in rodents without any symptoms, and humans acquire a variety of diseases when zoonotically infected. Lymphocytic choriomeningitis virus (LCMV) is the only arenavirus to exhibit a worldwide distribution, and causes illnesses such as meningitis [3, 4] . Congenital LCMV infections have also been reported [4, 5] . Most importantly, viral hemorrhagic fever (VHF) can be caused by several arenaviruses. Lassa fever, caused by the Lassa virus (LASV), an Old World arenavirus, is one of the most devastating VHFs in humans [6] . Hemorrhaging and organ failure occur in a subset of patients infected with this virus, and it is associated with high mortality. Many cases of Lassa fever occur in Western Africa in countries such as Guinea, Sierra Leone, and Nigeria [7] [8] [9] [10] [11] [12] [13] . Tacaribe complex lineage B of the New World arenaviruses consists of the Junin virus (JUNV), Guanarito virus (GUNV), Sabia virus (SABV) and Machupo virus (MACV), the etiological agents of Argentine, Venezuelan, Brazilian, and Bolivian hemorrhagic fevers, respectively [14, 15] . Although genetically distinct from one another, they appear to produce similar symptoms, accompanied by hemorrhaging in humans [14, 15] . These pathogenic New World arenavirus species are closely associated with a specific rodent species [6] .
Humans are usually infected with pathogenic arenaviruses through direct contact with tissue or blood, or after inhaling aerosolized particles from urine, feces, and saliva of infected rodents. After an incubation period of 1-3 weeks, infected individuals abruptly develop fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria, and mucosal bleeding. Neurological problems have also been described, including hearing loss, tremors, and encephalitis. Because the symptoms of pathogenic arenavirus-related illness are varied and nonspecific, the clinical diagnosis is often difficult [14, 16] . Human-to-human transmission may occur via mucosal or cutaneous contact, or through nosocomial contamination [14, 16] . These viruses are also considered to be potential bioterrorism agents [2] .
A number of arenavirus species have been recently discovered as a result of both rodent surveys and disease outbreaks [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . A novel pathogenic New World arenavirus, Chapare virus (CHPV), has been isolated from a fatal case of VHF in Bolivia [20] . In addition, five cases of VHF have been reported in South Africa, and a novel arenavirus, named Lujo virus, was isolated from a patient [17] . The Lujo virus is most distantly related to the other Old World arenaviruses [17] . To date, there is no information concerning the vertebrate host for the Chapare and Lujo viruses.
There is some evidence of endemicity of the Lassa virus in neighboring countries [27, 28] . However, as the magnitude of international trade and travel is continuously increasing, and the perturbation of the environment (due either to human activity or natural ecological changes) may result in behavioral changes of reservoir rodents, highly pathogenic arenaviruses could be introduced to virus-free countries from endemic areas. In fact, more than twenty cases of Lassa fever have been reported outside of the endemic region in areas such as the USA, Canada, Europe, and Japan [29] [30] [31] [32] [33] . It is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of outbreaks of VHFs caused by arenaviruses. However, these arenaviruses are classified as biosafety level (BSL)-4 pathogens, making it difficult to develop diagnostic techniques for these virus infections in laboratories without BSL-4 facilities. To overcome these difficulties, we have established recombinant viral nucleoproteins (rNPs)-based serological assays, such as IgG-enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescence assay (IFA), and antigen (Ag)-capture ELISA for the diagnosis of VHFs caused by highly pathogenic arenaviruses. Furthermore, virus neutralization assays using pseudotype virus-bearing arenavirus GPs have been developed. In this review, we describe the usefulness of such recombinant protein-based diagnostic assays for diagnosing VHFs caused by arenaviruses.
In outbreaks of VHFs, infections are confirmed by various laboratory diagnostic methods. Virus detection is performed by virus isolation, reverse transcription-polymerase chain reaction (RT-PCR), and antigen-capture ELISA. It has been shown that monoclonal antibody panels against pathogenic arenaviruses are useful for detecting viral antigens on the virus-infected cells as well as for investigating of antigenic relationships of arenaviruses [34] [35] [36] . Detection of the virus genome is suitable for a rapid and sensitive diagnosis of VHF patients in the early stage of illness, and extensive reviews of such RT-PCR assays have been described [37, 38] . More recently, progress in the RT-PCR method covering genetic variations of the hemorrhagic fever viruses (HFVs) [39, 40] and a multiplexed oligonucleotide microarray for the differential diagnosis of VHFs have also been reported [41] . On the other hand, antibodies against these viruses can be detected by the indirect immunofluorescence assay (IFA), or IgG-and IgM-ELISA. An IFA detects the antibody in the serum, which is able to bind to the fixed monolayer of the virus-infected cells. Although the interpretation of immunofluorescence results requires experience, the assay has advantages over other methods, since each virus generates a characteristic fluorescence pattern that adds specificity to the assay compared to a simple ELISA readout. A serological diagnosis by the detection of specific IgM and IgG antibodies to the HFVs must be sensitive, specific and reliable, because a misdiagnosis can lead to panic in the general population. An IgM-specific ELISA is suitable for detecting recent infection, but the relevance of IgM testing for acute VHF depends on the virus and the duration of illness; specific IgM is not often present in the very early stage of illness, and patients who die of VHF often fail to seroconvert at all. An IgG-specific ELISA is efficacious, not only in the diagnosis of a large number of VHF cases, especially during convalescence, but also for epidemiological studies in the endemic regions. The detailed methods used for the IFA and IgG-and IgM-ELISAs for the diagnosis of VHF using authentic virus-antigens have been described in detail [42] [43] [44] [45] .
Arenaviruses have a bisegmented, negative-sense, single stranded RNA genome with a unique ambisense coding strategy that produces just four known proteins: a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L) [46] . Of these proteins, the NP is the most abundant in virus-infected cells. Recombinant protein technology could meet the demand for a simple and reliable VHF test system, and recombinant NP (rNP) has been shown to be useful for serological surveys of IgM-and IgG antibodies against arenaviruses [47] [48] [49] [50] .
Recombinant baculoviruses that express the full-length rNP of arenaviruses have been generated [48, 50, 51] . The method used for the purification of arenavirus rNP from insect Tn5 cells infected with recombinant baculoviruses is effective and simple compared to those for Ebola, Marburg, and Crimean-Congo hemorrhagic fever virus rNPs [51] [52] [53] [54] [55] . Most of the arenavirus rNPs expressed in insect cells using the recombinant baculoviruses are crystallized [56] and are solubilized in PBS containing 8M urea. Since the majority of Tn5 cellular proteins are solubilized in PBS containing 2M urea, the arenavirus rNPs in the insoluble fraction in PBS containing 2M urea can be solubilized by sonication in PBS containing 8M urea. After a simple centrifugation of the lysates in PBS containing 8M urea, the supernatant fractions can be used as purified rNP antigens without further purification steps [51] . The control antigen is produced from Tn5 cells infected with baculovirus lacking the polyhedrin gene (ΔP) in the same manner as the arenavirus rNPs ( Figure 1 ).
Purified rNPs. The expression and purification efficiency of arenavirus rNP were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after staining the gels with Coomassie blue. Purified NP antigens with approximate molecular weights of 62 kDa from Luna, LCM, Lassa, Lujo, Junin, Machupo, Guanarito, Sabia, and Chapare viruses and the purified negative control antigen (ΔP) are shown.
As described above, recombinant baculoviruses allow the delivery of rNP antigens without using infectious live arenaviruses. An ELISA plate coated with the predetermined optimal quantity of purified rNPs (approximately 100 ng/well) is used for the IgG-antibody detection assay. An advantage of using recombinant rNP for the IgG-ELISA is that it enables a direct comparison of antibody cross-reactivity among arenavirus rNPs, since antigen preparations of all arenavirus rNPs tested are performed using the same method [51] . Rabbit anti-sera raised against LCMV-rNP and LASV-rNP show cross-reactivity to LASV-rNP and LCMV-rNP, respectively, indicating that rabbit antibodies against rNPs of Old World arenaviruses cross-react with rNPs of other Old World arenaviruses (Table 1 ) [51] . Similarly, rabbit anti-sera generated against JUNV-NP show cross-reactivity to the LASV-rNP and LCMV-rNP, although the reaction is weak. However, rabbit anti-sera against LASV-NP and LCMV-NP show a negative reaction to the JUNV-rNP (Table 1 ) [51] , indicating that rabbit antibodies against JUNV (a pathogenic New World arenavirus) NP might cross-react with the Old World arenavirus NP, whereas antibodies against Old World arenavirus NPs may not be able to react with pathogenic New World arenavirus NPs.
The rNP-based IgG-ELISA has also been used for the characterization of a mouse monoclonal antibody (MAb). Nakauchi et al. [50] have investigated the cross-reactivity of MAbs against JUNV rNP to pathogenic New World arenavirus rNPs, as well as LASV rNP. MAb C11-12 reacts at the same level with the rNPs of all of the pathogenic New World arenaviruses, including JUNV, GTOV, MACV, SABV, and CHPV, indicating that this MAb recognizes an epitope conserved among pathogenic New World arenaviruses. Another MAb, C6-9, reacts specifically with the rNP of JUNV, but does not react with those of the other pathogenic New World arenaviruses [50] . This indicates that MAb C6-9 recognizes a JUNV-specific epitope. None of these MAbs reacts with the rNP of the human pathogenic Old World arenavirus LASV. Thus, the MAb C11-12 is considered to be a broadly reactive MAb against New World arenaviruses, whereas MAb C6-9 is JUNV-specific. These findings have been confirmed by detailed epitope analyses using peptide mapping [50] . Similarly, the cross-reactivity of MAbs against LASV rNP has been analyzed [51] . MAb 4A5 cross-reacts with the Mopeia virus (MOPV) but not with the LCMV rNP. MAb 6C11 cross-reacts with LCMV rNP, while MAb 2-11 does not cross-react with LCMV rNP [51] . Table 1 . Anti-serum reactivity for rNPs of different arenaviruses in IgG ELISAs.
Reactivity for rNP from LASV LCMV JUNV anti-LASV NP
It is important to evaluate whether rNP-based ELISA is useful for the diagnosis of human VHF cases. The specificity of the LASV-rNP-based IgG ELISA has been confirmed by using sera obtained from Lassa fever patients [51] . The Lassa fever patients' sera show a highly positive reaction in the LASV-rNP-based IgG-ELISA, but sera from patients with Argentine hemorrhagic fever (AHF), which is caused by JUNV, do not. The serum from an AHF patient showed a highly positive reaction in the JUNV-rNP-based IgG-ELISA [49] . In addition, it was shown that, using sera obtained from AHF cases, the results of the JUNV rNP-based IgG ELISA correlate well with an authentic JUNV antigen-based IgG ELISA [49] . An IgM-capture ELISA using purified LASV-rNP as an antigen has been developed in the same way as in previous reports [54, 57] and detects an LASV-IgM antibody [58] . In addition, immunoblot assays based on N-terminally truncated LASV rNP have been developed for detecting IgG and IgM antibodies against LASV. These methods may provide a rapid and simple Lassa fever test for use under field conditions [47] .
An IFA using virus-infected cells is a common antibody test for VHF viruses [59] [60] [61] [62] [63] . To avoid the use of highly pathogenic viruses for the antigen preparation, mammalian cells expressing recombinant rNP have been developed [51, 57, [64] [65] [66] [67] [68] . Lassa virus NP antigen for IFA can be prepared simply as described [51] . Briefly, the procedure involves (1) transfecting HeLa cells with a mammalian cell expression vector inserted with the cloned NP cDNA; (2) expanding the stable NP-expressing cells by antibiotic selection; (3) mixing the rNP-expressing cells with un-transfected HeLa cells (at a ratio of 1:1); (4) spotting the cell mixtures onto glass slides, then drying and fixing them in acetone.
In the IFA specific for LASV-NP, antibody positive sera show characteristic granular staining patterns in the cytoplasm (Figure 2 ) [69] , thus making it easy to distinguish positive from negative samples. The specificity of the assay has also been confirmed by using sera obtained from Lassa fever patients [51] . In addition, an IFA using JUNV rNP-expressing HeLa cells has been developed to detect antibodies against JUNV, and the assay has been evaluated by using AHF patients' sera [70] . The LASV-rNP-based antibody detection systems such as ELISA and IFA are suggested to be useful not only for the diagnosis of Lassa fever, but also for seroepidemiological studies of LASV infection. In our preliminary study, approximately 15% of the sera collected from 334 Ghanaians and less than 3% of 280 Zambians showed positive reactions in the LASV-rNP-based IgG ELISA [58] . These results are in agreement with the fact that Lassa fever is endemic to the West African region, including Ghana, but less in the East African region.
For the diagnosis of many viral infections, PCR assays have been shown to have an excellent analytical sensitivity, but the established techniques are limited by their requirement for expensive equipment and technical expertise. Moreover, the high degree of genetic variability of the RNA viruses, including arenavirus and bunyavirus, poses difficulties in selecting primers for RT-PCR assays that can detect all strains of the virus. Since the sensitivity of the Ag-capture ELISA is comparable to that of RT-PCR for several virus-mediated infectious diseases, including Lassa fever and filovirus hemorrhagic fever [51, [71] [72] [73] , the Ag-capture ELISA is a sophisticated approach that can be used for the diagnosis of viral infections. Ag-capture ELISAs detecting viral NP in viremic sera have been widely applied to detect various viruses, since they are the most abundant viral antigens and have highly conserved amino acid sequences [50, 51, 54, 71, 72, 74, 75] . Polyclonal anti-sera or a mixture of MAbs present in the ascetic fluids from animals immunized for HFVs have been used for capture-antibodies in the Ag-capture ELISA [36, [76] [77] [78] [79] . MAbs recognizing conserved epitopes of the rNP are also used as capture antibodies since they have a high specificity for the antigens, and an identification of the epitopes of these MAbs is of crucial importance for the assessment of the specificity and cross-reactivity of the assay system [50, 51, 53, 54, 71, 75] . In order to develop a sensitive diagnostic test for Lassa fever and AHF, rNPs of LASV and JUNV (see above) have been prepared, and newly established MAbs against them have been characterized and used for Ag-capture ELISAs [50, 51] . The Ag-capture ELISA using MAb 4A5 has been confirmed to be useful in the detection of authentic LASV antigen in sera serially collected from hamsters infected with LASV [51] . The sensitivity of the MAb 4A5-based Ag-capture ELISA was similar to that of conventional RT-PCR, suggesting that the Ag-capture ELISA can be efficiently used in the diagnosis of Lassa fever [51] . Therefore, the MAb 4A5-based Ag-capture ELISA is considered to be useful in the diagnosis of Lassa fever. Also, by using MAbs raised against the rNP of JUNV, Ag-capture ELISAs specific for JUNV and broadly reactive to human pathogenic New World arenaviruses have been developed [50] . The Ag-capture ELISA using MAb E4-2 and C11-12 detected the Ags of all of the pathogenic New World arenaviruses tested, including JUNV. On the other hand, the Ag-capture ELISA using MAb C6-9 detects only the JUNV Ag. Considering that the symptoms of JUNV infection in humans are indistinguishable from those due to other pathogenic New World arenaviruses, the Ag capture ELISA using MAb C6-9 may be a useful diagnostic tool, especially for AHF [50] .
The virus neutralization assay is accepted as the "gold standard" serodiagnostic assay to quantify the antibody response to infection and vaccination of a wide variety of viruses associated with human diseases [80] [81] [82] [83] [84] [85] [86] . The presence of neutralizing antibodies is a reliable indicator of protective immunity against VHF [87] [88] [89] . The most direct method for detection of neutralizing antibodies against HFVs is by plaque reduction neutralization tests using infectious viruses. However, because of the high pathogenicity of HFVs to humans and the strict regulation of select agents, only a limited number of laboratories are able to perform such neutralization tests. For many HFVs, replication-incompetent pseudotyped virus particles bearing viral envelope protein (GP) have been shown to mimic the respective HFV infections, thus, neutralization assays using the pseudotypes may be advantageous in some laboratory settings for the detection of antibodies to HFVs without the need for heightened biocontainment requirements.
The VSV-based vector has already been used to generate replication-competent recombinant VSVs to study of the role of GPs of various viruses [90] [91] [92] . Recent advances in producing pseudotype virus particles have enabled the investigation of the virus cell entry, viral tropism, and effect of entry inhibitors, as well as measurement of the neutralization titers, by using human immunodeficiency virus-, feline immunodeficiency virus-, murine leukemia virus-, or VSV-based vectors [86, [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] . Pseudotypes based on VSV have advantages compared with other pseudotypes based on retroviruses for the following reasons. First, the pseudotype virus titer obtained with the VSV system is generally higher than that of the pseudotyped retrovirus system [104] . Second, the infection of target cells with a VSV pseudotype can be readily detected as green fluorescent protein (GFP)-positive cells at 7-16 h post-infection because of the high level of GFP expression in the VSV system [104, 105] . In contrast, the time required for infection in the pseudotyped retrovirus system is 48 h [106, 107] , which is similar to the time required for infectious viruses to replicate to a level that results in plaque-forming or cytopathic effects in infected cells. A high-throughput assay for determining neutralizing antibody titers using VSV pseudotypes expressing secreted alkaline phosphatase [108, 109] or luciferase ( Figure 3 ) has also been developed. We have recently developed a VSV-based pseudotype bearing Lassa virus GP (VSV-LAS-GP) for the detection of neutralizing antibodies in the sera obtained from a Lassa fever patient. An example of the LASV neutralization assay using the VSV pseudotype is shown (Figure 4 ). In the presence of serum from Lassa fever patients, the number of GFP-positive cells (infectivity of VSV-LAS-GP) is significantly reduced compared with the number in the absence of the patient's serum ( Figure 4A ). The control VSV pseudotype bearing VSV GP (VSV-VSV-G) is not neutralized by any sera. When the cut-off serum dilution is set at 50% inhibition of infectivity compared with the infectivity in the absence of the test serum, the neutralization titer of this patient's serum for VSV-LAS-GP is calculated to be 75 ( Figure 4B ). Likewise, a VSV-based pseudotype bearing the Junin virus GP has been developed for the detection of neutralizing antibodies from AHF patients' sera. The accuracy of the results of VSV-based neutralization assays has been confirmed by comparison with the results of the neutralization assay using live Junin virus [70] . The Lujo virus is a new member of the hemorrhagic fever-associated arenavirus family from Zambia and southern Africa, and the virus is classified as a BSL-4 pathogen [17] . The genome sequence analysis of the Lujo virus suggests that the virus is genetically distinct from previously characterized arenaviruses. In order to study the infectivity of this newly identified arenavirus, we have recently developed a luciferase-expressing VSV pseudotype bearing Lujo virus GPC (VSV-Lujo-GP). As shown in Figure 3 , infection with VSV-Lujo-GPC is specifically neutralized by rabbit anti-Lujo GPC serum. Thus, the VSV-Lujo-GP may be a useful tool not only for determining the neutralizing antibody titer within the serum, but also for exploring yet-to-be-defined cellular receptor(s) for Lujo virus infection or for screening inhibitors of the Lujo virus GP-mediated cell entry.
Hemorrhagic fever outbreaks caused by pathogenic arenaviruses result in high fatality rates. A rapid and accurate diagnosis is a critical first step in any outbreak. Serologic diagnostic methods for VHFs most often employ an ELISA, IFA, and/or virus neutralization assay. Diagnostic methods using recombinant viral proteins have been developed and their utilities for diagnosing of VHF have been reviewed. IgG-and IgM-ELISAs and IFAs using rNPs as antigens are useful for the detection of antibodies induced in the patients' sera. These methods are also useful for seroepidemiological surveys for HFVs. Ag-capture ELISAs using MAbs to the arenavirus rNPs are specific for the virus species or can be broadly reactive for New World arenaviruses, depending on the MAb used. Furthermore, the VSV-based pseudotype system provides a safe and rapid tool for measuring virus neutralizing antibody titers, as well as a model to analyze the entry of the respective arenavirus in susceptible cells without using live arenaviruses. Recent discoveries of novel arenavirus species [17, 26, 110] and their potential to evolve predominantly via host switching, rather than with their hosts [110, 111] , suggest that an unknown pathogenic arenavirus may emerge in the future, and that the diagnostic methods for VHF caused by arenaviruses should thus be further developed and improved.
|
Which viruses are part of the Old World complex of Arenaviridae?
| 5,271 |
Lassa and Lujo viruses
| 514 |
1,606 |
Serological Assays Based on Recombinant Viral Proteins for the Diagnosis of Arenavirus Hemorrhagic Fevers
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3497043/
SHA: f1d308db379b3c293bcfc8fe251c043fe8842358
Authors: Fukushi, Shuetsu; Tani, Hideki; Yoshikawa, Tomoki; Saijo, Masayuki; Morikawa, Shigeru
Date: 2012-10-12
DOI: 10.3390/v4102097
License: cc-by
Abstract: The family Arenaviridae, genus Arenavirus, consists of two phylogenetically independent groups: Old World (OW) and New World (NW) complexes. The Lassa and Lujo viruses in the OW complex and the Guanarito, Junin, Machupo, Sabia, and Chapare viruses in the NW complex cause viral hemorrhagic fever (VHF) in humans, leading to serious public health concerns. These viruses are also considered potential bioterrorism agents. Therefore, it is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of arenavirus outbreaks. However, these arenaviruses are classified as BSL-4 pathogens, thus making it difficult to develop diagnostic techniques for these virus infections in institutes without BSL-4 facilities. To overcome these difficulties, antibody detection systems in the form of an enzyme-linked immunosorbent assay (ELISA) and an indirect immunofluorescence assay were developed using recombinant nucleoproteins (rNPs) derived from these viruses. Furthermore, several antigen-detection assays were developed. For example, novel monoclonal antibodies (mAbs) to the rNPs of Lassa and Junin viruses were generated. Sandwich antigen-capture (Ag-capture) ELISAs using these mAbs as capture antibodies were developed and confirmed to be sensitive and specific for detecting the respective arenavirus NPs. These rNP-based assays were proposed to be useful not only for an etiological diagnosis of VHFs, but also for seroepidemiological studies on VHFs. We recently developed arenavirus neutralization assays using vesicular stomatitis virus (VSV)-based pseudotypes bearing arenavirus recombinant glycoproteins. The goal of this article is to review the recent advances in developing laboratory diagnostic assays based on recombinant viral proteins for the diagnosis of VHFs and epidemiological studies on the VHFs caused by arenaviruses.
Text: The virus family Arenaviridae consists of only one genus, but most viruses within this genus can be divided into two different groups: the Old World arenaviruses and the New World arenaviruses (also known as the Tacaribe complex) [1, 2] . The differences between the two groups have been established through the use of serological assays. Most of the arenaviruses cause persistent infection in rodents without any symptoms, and humans acquire a variety of diseases when zoonotically infected. Lymphocytic choriomeningitis virus (LCMV) is the only arenavirus to exhibit a worldwide distribution, and causes illnesses such as meningitis [3, 4] . Congenital LCMV infections have also been reported [4, 5] . Most importantly, viral hemorrhagic fever (VHF) can be caused by several arenaviruses. Lassa fever, caused by the Lassa virus (LASV), an Old World arenavirus, is one of the most devastating VHFs in humans [6] . Hemorrhaging and organ failure occur in a subset of patients infected with this virus, and it is associated with high mortality. Many cases of Lassa fever occur in Western Africa in countries such as Guinea, Sierra Leone, and Nigeria [7] [8] [9] [10] [11] [12] [13] . Tacaribe complex lineage B of the New World arenaviruses consists of the Junin virus (JUNV), Guanarito virus (GUNV), Sabia virus (SABV) and Machupo virus (MACV), the etiological agents of Argentine, Venezuelan, Brazilian, and Bolivian hemorrhagic fevers, respectively [14, 15] . Although genetically distinct from one another, they appear to produce similar symptoms, accompanied by hemorrhaging in humans [14, 15] . These pathogenic New World arenavirus species are closely associated with a specific rodent species [6] .
Humans are usually infected with pathogenic arenaviruses through direct contact with tissue or blood, or after inhaling aerosolized particles from urine, feces, and saliva of infected rodents. After an incubation period of 1-3 weeks, infected individuals abruptly develop fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria, and mucosal bleeding. Neurological problems have also been described, including hearing loss, tremors, and encephalitis. Because the symptoms of pathogenic arenavirus-related illness are varied and nonspecific, the clinical diagnosis is often difficult [14, 16] . Human-to-human transmission may occur via mucosal or cutaneous contact, or through nosocomial contamination [14, 16] . These viruses are also considered to be potential bioterrorism agents [2] .
A number of arenavirus species have been recently discovered as a result of both rodent surveys and disease outbreaks [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . A novel pathogenic New World arenavirus, Chapare virus (CHPV), has been isolated from a fatal case of VHF in Bolivia [20] . In addition, five cases of VHF have been reported in South Africa, and a novel arenavirus, named Lujo virus, was isolated from a patient [17] . The Lujo virus is most distantly related to the other Old World arenaviruses [17] . To date, there is no information concerning the vertebrate host for the Chapare and Lujo viruses.
There is some evidence of endemicity of the Lassa virus in neighboring countries [27, 28] . However, as the magnitude of international trade and travel is continuously increasing, and the perturbation of the environment (due either to human activity or natural ecological changes) may result in behavioral changes of reservoir rodents, highly pathogenic arenaviruses could be introduced to virus-free countries from endemic areas. In fact, more than twenty cases of Lassa fever have been reported outside of the endemic region in areas such as the USA, Canada, Europe, and Japan [29] [30] [31] [32] [33] . It is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of outbreaks of VHFs caused by arenaviruses. However, these arenaviruses are classified as biosafety level (BSL)-4 pathogens, making it difficult to develop diagnostic techniques for these virus infections in laboratories without BSL-4 facilities. To overcome these difficulties, we have established recombinant viral nucleoproteins (rNPs)-based serological assays, such as IgG-enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescence assay (IFA), and antigen (Ag)-capture ELISA for the diagnosis of VHFs caused by highly pathogenic arenaviruses. Furthermore, virus neutralization assays using pseudotype virus-bearing arenavirus GPs have been developed. In this review, we describe the usefulness of such recombinant protein-based diagnostic assays for diagnosing VHFs caused by arenaviruses.
In outbreaks of VHFs, infections are confirmed by various laboratory diagnostic methods. Virus detection is performed by virus isolation, reverse transcription-polymerase chain reaction (RT-PCR), and antigen-capture ELISA. It has been shown that monoclonal antibody panels against pathogenic arenaviruses are useful for detecting viral antigens on the virus-infected cells as well as for investigating of antigenic relationships of arenaviruses [34] [35] [36] . Detection of the virus genome is suitable for a rapid and sensitive diagnosis of VHF patients in the early stage of illness, and extensive reviews of such RT-PCR assays have been described [37, 38] . More recently, progress in the RT-PCR method covering genetic variations of the hemorrhagic fever viruses (HFVs) [39, 40] and a multiplexed oligonucleotide microarray for the differential diagnosis of VHFs have also been reported [41] . On the other hand, antibodies against these viruses can be detected by the indirect immunofluorescence assay (IFA), or IgG-and IgM-ELISA. An IFA detects the antibody in the serum, which is able to bind to the fixed monolayer of the virus-infected cells. Although the interpretation of immunofluorescence results requires experience, the assay has advantages over other methods, since each virus generates a characteristic fluorescence pattern that adds specificity to the assay compared to a simple ELISA readout. A serological diagnosis by the detection of specific IgM and IgG antibodies to the HFVs must be sensitive, specific and reliable, because a misdiagnosis can lead to panic in the general population. An IgM-specific ELISA is suitable for detecting recent infection, but the relevance of IgM testing for acute VHF depends on the virus and the duration of illness; specific IgM is not often present in the very early stage of illness, and patients who die of VHF often fail to seroconvert at all. An IgG-specific ELISA is efficacious, not only in the diagnosis of a large number of VHF cases, especially during convalescence, but also for epidemiological studies in the endemic regions. The detailed methods used for the IFA and IgG-and IgM-ELISAs for the diagnosis of VHF using authentic virus-antigens have been described in detail [42] [43] [44] [45] .
Arenaviruses have a bisegmented, negative-sense, single stranded RNA genome with a unique ambisense coding strategy that produces just four known proteins: a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L) [46] . Of these proteins, the NP is the most abundant in virus-infected cells. Recombinant protein technology could meet the demand for a simple and reliable VHF test system, and recombinant NP (rNP) has been shown to be useful for serological surveys of IgM-and IgG antibodies against arenaviruses [47] [48] [49] [50] .
Recombinant baculoviruses that express the full-length rNP of arenaviruses have been generated [48, 50, 51] . The method used for the purification of arenavirus rNP from insect Tn5 cells infected with recombinant baculoviruses is effective and simple compared to those for Ebola, Marburg, and Crimean-Congo hemorrhagic fever virus rNPs [51] [52] [53] [54] [55] . Most of the arenavirus rNPs expressed in insect cells using the recombinant baculoviruses are crystallized [56] and are solubilized in PBS containing 8M urea. Since the majority of Tn5 cellular proteins are solubilized in PBS containing 2M urea, the arenavirus rNPs in the insoluble fraction in PBS containing 2M urea can be solubilized by sonication in PBS containing 8M urea. After a simple centrifugation of the lysates in PBS containing 8M urea, the supernatant fractions can be used as purified rNP antigens without further purification steps [51] . The control antigen is produced from Tn5 cells infected with baculovirus lacking the polyhedrin gene (ΔP) in the same manner as the arenavirus rNPs ( Figure 1 ).
Purified rNPs. The expression and purification efficiency of arenavirus rNP were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after staining the gels with Coomassie blue. Purified NP antigens with approximate molecular weights of 62 kDa from Luna, LCM, Lassa, Lujo, Junin, Machupo, Guanarito, Sabia, and Chapare viruses and the purified negative control antigen (ΔP) are shown.
As described above, recombinant baculoviruses allow the delivery of rNP antigens without using infectious live arenaviruses. An ELISA plate coated with the predetermined optimal quantity of purified rNPs (approximately 100 ng/well) is used for the IgG-antibody detection assay. An advantage of using recombinant rNP for the IgG-ELISA is that it enables a direct comparison of antibody cross-reactivity among arenavirus rNPs, since antigen preparations of all arenavirus rNPs tested are performed using the same method [51] . Rabbit anti-sera raised against LCMV-rNP and LASV-rNP show cross-reactivity to LASV-rNP and LCMV-rNP, respectively, indicating that rabbit antibodies against rNPs of Old World arenaviruses cross-react with rNPs of other Old World arenaviruses (Table 1 ) [51] . Similarly, rabbit anti-sera generated against JUNV-NP show cross-reactivity to the LASV-rNP and LCMV-rNP, although the reaction is weak. However, rabbit anti-sera against LASV-NP and LCMV-NP show a negative reaction to the JUNV-rNP (Table 1 ) [51] , indicating that rabbit antibodies against JUNV (a pathogenic New World arenavirus) NP might cross-react with the Old World arenavirus NP, whereas antibodies against Old World arenavirus NPs may not be able to react with pathogenic New World arenavirus NPs.
The rNP-based IgG-ELISA has also been used for the characterization of a mouse monoclonal antibody (MAb). Nakauchi et al. [50] have investigated the cross-reactivity of MAbs against JUNV rNP to pathogenic New World arenavirus rNPs, as well as LASV rNP. MAb C11-12 reacts at the same level with the rNPs of all of the pathogenic New World arenaviruses, including JUNV, GTOV, MACV, SABV, and CHPV, indicating that this MAb recognizes an epitope conserved among pathogenic New World arenaviruses. Another MAb, C6-9, reacts specifically with the rNP of JUNV, but does not react with those of the other pathogenic New World arenaviruses [50] . This indicates that MAb C6-9 recognizes a JUNV-specific epitope. None of these MAbs reacts with the rNP of the human pathogenic Old World arenavirus LASV. Thus, the MAb C11-12 is considered to be a broadly reactive MAb against New World arenaviruses, whereas MAb C6-9 is JUNV-specific. These findings have been confirmed by detailed epitope analyses using peptide mapping [50] . Similarly, the cross-reactivity of MAbs against LASV rNP has been analyzed [51] . MAb 4A5 cross-reacts with the Mopeia virus (MOPV) but not with the LCMV rNP. MAb 6C11 cross-reacts with LCMV rNP, while MAb 2-11 does not cross-react with LCMV rNP [51] . Table 1 . Anti-serum reactivity for rNPs of different arenaviruses in IgG ELISAs.
Reactivity for rNP from LASV LCMV JUNV anti-LASV NP
It is important to evaluate whether rNP-based ELISA is useful for the diagnosis of human VHF cases. The specificity of the LASV-rNP-based IgG ELISA has been confirmed by using sera obtained from Lassa fever patients [51] . The Lassa fever patients' sera show a highly positive reaction in the LASV-rNP-based IgG-ELISA, but sera from patients with Argentine hemorrhagic fever (AHF), which is caused by JUNV, do not. The serum from an AHF patient showed a highly positive reaction in the JUNV-rNP-based IgG-ELISA [49] . In addition, it was shown that, using sera obtained from AHF cases, the results of the JUNV rNP-based IgG ELISA correlate well with an authentic JUNV antigen-based IgG ELISA [49] . An IgM-capture ELISA using purified LASV-rNP as an antigen has been developed in the same way as in previous reports [54, 57] and detects an LASV-IgM antibody [58] . In addition, immunoblot assays based on N-terminally truncated LASV rNP have been developed for detecting IgG and IgM antibodies against LASV. These methods may provide a rapid and simple Lassa fever test for use under field conditions [47] .
An IFA using virus-infected cells is a common antibody test for VHF viruses [59] [60] [61] [62] [63] . To avoid the use of highly pathogenic viruses for the antigen preparation, mammalian cells expressing recombinant rNP have been developed [51, 57, [64] [65] [66] [67] [68] . Lassa virus NP antigen for IFA can be prepared simply as described [51] . Briefly, the procedure involves (1) transfecting HeLa cells with a mammalian cell expression vector inserted with the cloned NP cDNA; (2) expanding the stable NP-expressing cells by antibiotic selection; (3) mixing the rNP-expressing cells with un-transfected HeLa cells (at a ratio of 1:1); (4) spotting the cell mixtures onto glass slides, then drying and fixing them in acetone.
In the IFA specific for LASV-NP, antibody positive sera show characteristic granular staining patterns in the cytoplasm (Figure 2 ) [69] , thus making it easy to distinguish positive from negative samples. The specificity of the assay has also been confirmed by using sera obtained from Lassa fever patients [51] . In addition, an IFA using JUNV rNP-expressing HeLa cells has been developed to detect antibodies against JUNV, and the assay has been evaluated by using AHF patients' sera [70] . The LASV-rNP-based antibody detection systems such as ELISA and IFA are suggested to be useful not only for the diagnosis of Lassa fever, but also for seroepidemiological studies of LASV infection. In our preliminary study, approximately 15% of the sera collected from 334 Ghanaians and less than 3% of 280 Zambians showed positive reactions in the LASV-rNP-based IgG ELISA [58] . These results are in agreement with the fact that Lassa fever is endemic to the West African region, including Ghana, but less in the East African region.
For the diagnosis of many viral infections, PCR assays have been shown to have an excellent analytical sensitivity, but the established techniques are limited by their requirement for expensive equipment and technical expertise. Moreover, the high degree of genetic variability of the RNA viruses, including arenavirus and bunyavirus, poses difficulties in selecting primers for RT-PCR assays that can detect all strains of the virus. Since the sensitivity of the Ag-capture ELISA is comparable to that of RT-PCR for several virus-mediated infectious diseases, including Lassa fever and filovirus hemorrhagic fever [51, [71] [72] [73] , the Ag-capture ELISA is a sophisticated approach that can be used for the diagnosis of viral infections. Ag-capture ELISAs detecting viral NP in viremic sera have been widely applied to detect various viruses, since they are the most abundant viral antigens and have highly conserved amino acid sequences [50, 51, 54, 71, 72, 74, 75] . Polyclonal anti-sera or a mixture of MAbs present in the ascetic fluids from animals immunized for HFVs have been used for capture-antibodies in the Ag-capture ELISA [36, [76] [77] [78] [79] . MAbs recognizing conserved epitopes of the rNP are also used as capture antibodies since they have a high specificity for the antigens, and an identification of the epitopes of these MAbs is of crucial importance for the assessment of the specificity and cross-reactivity of the assay system [50, 51, 53, 54, 71, 75] . In order to develop a sensitive diagnostic test for Lassa fever and AHF, rNPs of LASV and JUNV (see above) have been prepared, and newly established MAbs against them have been characterized and used for Ag-capture ELISAs [50, 51] . The Ag-capture ELISA using MAb 4A5 has been confirmed to be useful in the detection of authentic LASV antigen in sera serially collected from hamsters infected with LASV [51] . The sensitivity of the MAb 4A5-based Ag-capture ELISA was similar to that of conventional RT-PCR, suggesting that the Ag-capture ELISA can be efficiently used in the diagnosis of Lassa fever [51] . Therefore, the MAb 4A5-based Ag-capture ELISA is considered to be useful in the diagnosis of Lassa fever. Also, by using MAbs raised against the rNP of JUNV, Ag-capture ELISAs specific for JUNV and broadly reactive to human pathogenic New World arenaviruses have been developed [50] . The Ag-capture ELISA using MAb E4-2 and C11-12 detected the Ags of all of the pathogenic New World arenaviruses tested, including JUNV. On the other hand, the Ag-capture ELISA using MAb C6-9 detects only the JUNV Ag. Considering that the symptoms of JUNV infection in humans are indistinguishable from those due to other pathogenic New World arenaviruses, the Ag capture ELISA using MAb C6-9 may be a useful diagnostic tool, especially for AHF [50] .
The virus neutralization assay is accepted as the "gold standard" serodiagnostic assay to quantify the antibody response to infection and vaccination of a wide variety of viruses associated with human diseases [80] [81] [82] [83] [84] [85] [86] . The presence of neutralizing antibodies is a reliable indicator of protective immunity against VHF [87] [88] [89] . The most direct method for detection of neutralizing antibodies against HFVs is by plaque reduction neutralization tests using infectious viruses. However, because of the high pathogenicity of HFVs to humans and the strict regulation of select agents, only a limited number of laboratories are able to perform such neutralization tests. For many HFVs, replication-incompetent pseudotyped virus particles bearing viral envelope protein (GP) have been shown to mimic the respective HFV infections, thus, neutralization assays using the pseudotypes may be advantageous in some laboratory settings for the detection of antibodies to HFVs without the need for heightened biocontainment requirements.
The VSV-based vector has already been used to generate replication-competent recombinant VSVs to study of the role of GPs of various viruses [90] [91] [92] . Recent advances in producing pseudotype virus particles have enabled the investigation of the virus cell entry, viral tropism, and effect of entry inhibitors, as well as measurement of the neutralization titers, by using human immunodeficiency virus-, feline immunodeficiency virus-, murine leukemia virus-, or VSV-based vectors [86, [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] . Pseudotypes based on VSV have advantages compared with other pseudotypes based on retroviruses for the following reasons. First, the pseudotype virus titer obtained with the VSV system is generally higher than that of the pseudotyped retrovirus system [104] . Second, the infection of target cells with a VSV pseudotype can be readily detected as green fluorescent protein (GFP)-positive cells at 7-16 h post-infection because of the high level of GFP expression in the VSV system [104, 105] . In contrast, the time required for infection in the pseudotyped retrovirus system is 48 h [106, 107] , which is similar to the time required for infectious viruses to replicate to a level that results in plaque-forming or cytopathic effects in infected cells. A high-throughput assay for determining neutralizing antibody titers using VSV pseudotypes expressing secreted alkaline phosphatase [108, 109] or luciferase ( Figure 3 ) has also been developed. We have recently developed a VSV-based pseudotype bearing Lassa virus GP (VSV-LAS-GP) for the detection of neutralizing antibodies in the sera obtained from a Lassa fever patient. An example of the LASV neutralization assay using the VSV pseudotype is shown (Figure 4 ). In the presence of serum from Lassa fever patients, the number of GFP-positive cells (infectivity of VSV-LAS-GP) is significantly reduced compared with the number in the absence of the patient's serum ( Figure 4A ). The control VSV pseudotype bearing VSV GP (VSV-VSV-G) is not neutralized by any sera. When the cut-off serum dilution is set at 50% inhibition of infectivity compared with the infectivity in the absence of the test serum, the neutralization titer of this patient's serum for VSV-LAS-GP is calculated to be 75 ( Figure 4B ). Likewise, a VSV-based pseudotype bearing the Junin virus GP has been developed for the detection of neutralizing antibodies from AHF patients' sera. The accuracy of the results of VSV-based neutralization assays has been confirmed by comparison with the results of the neutralization assay using live Junin virus [70] . The Lujo virus is a new member of the hemorrhagic fever-associated arenavirus family from Zambia and southern Africa, and the virus is classified as a BSL-4 pathogen [17] . The genome sequence analysis of the Lujo virus suggests that the virus is genetically distinct from previously characterized arenaviruses. In order to study the infectivity of this newly identified arenavirus, we have recently developed a luciferase-expressing VSV pseudotype bearing Lujo virus GPC (VSV-Lujo-GP). As shown in Figure 3 , infection with VSV-Lujo-GPC is specifically neutralized by rabbit anti-Lujo GPC serum. Thus, the VSV-Lujo-GP may be a useful tool not only for determining the neutralizing antibody titer within the serum, but also for exploring yet-to-be-defined cellular receptor(s) for Lujo virus infection or for screening inhibitors of the Lujo virus GP-mediated cell entry.
Hemorrhagic fever outbreaks caused by pathogenic arenaviruses result in high fatality rates. A rapid and accurate diagnosis is a critical first step in any outbreak. Serologic diagnostic methods for VHFs most often employ an ELISA, IFA, and/or virus neutralization assay. Diagnostic methods using recombinant viral proteins have been developed and their utilities for diagnosing of VHF have been reviewed. IgG-and IgM-ELISAs and IFAs using rNPs as antigens are useful for the detection of antibodies induced in the patients' sera. These methods are also useful for seroepidemiological surveys for HFVs. Ag-capture ELISAs using MAbs to the arenavirus rNPs are specific for the virus species or can be broadly reactive for New World arenaviruses, depending on the MAb used. Furthermore, the VSV-based pseudotype system provides a safe and rapid tool for measuring virus neutralizing antibody titers, as well as a model to analyze the entry of the respective arenavirus in susceptible cells without using live arenaviruses. Recent discoveries of novel arenavirus species [17, 26, 110] and their potential to evolve predominantly via host switching, rather than with their hosts [110, 111] , suggest that an unknown pathogenic arenavirus may emerge in the future, and that the diagnostic methods for VHF caused by arenaviruses should thus be further developed and improved.
|
How can Old World and New World Arenaviruses be differentiated?
| 5,272 |
through the use of serological assays
| 2,565 |
1,606 |
Serological Assays Based on Recombinant Viral Proteins for the Diagnosis of Arenavirus Hemorrhagic Fevers
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3497043/
SHA: f1d308db379b3c293bcfc8fe251c043fe8842358
Authors: Fukushi, Shuetsu; Tani, Hideki; Yoshikawa, Tomoki; Saijo, Masayuki; Morikawa, Shigeru
Date: 2012-10-12
DOI: 10.3390/v4102097
License: cc-by
Abstract: The family Arenaviridae, genus Arenavirus, consists of two phylogenetically independent groups: Old World (OW) and New World (NW) complexes. The Lassa and Lujo viruses in the OW complex and the Guanarito, Junin, Machupo, Sabia, and Chapare viruses in the NW complex cause viral hemorrhagic fever (VHF) in humans, leading to serious public health concerns. These viruses are also considered potential bioterrorism agents. Therefore, it is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of arenavirus outbreaks. However, these arenaviruses are classified as BSL-4 pathogens, thus making it difficult to develop diagnostic techniques for these virus infections in institutes without BSL-4 facilities. To overcome these difficulties, antibody detection systems in the form of an enzyme-linked immunosorbent assay (ELISA) and an indirect immunofluorescence assay were developed using recombinant nucleoproteins (rNPs) derived from these viruses. Furthermore, several antigen-detection assays were developed. For example, novel monoclonal antibodies (mAbs) to the rNPs of Lassa and Junin viruses were generated. Sandwich antigen-capture (Ag-capture) ELISAs using these mAbs as capture antibodies were developed and confirmed to be sensitive and specific for detecting the respective arenavirus NPs. These rNP-based assays were proposed to be useful not only for an etiological diagnosis of VHFs, but also for seroepidemiological studies on VHFs. We recently developed arenavirus neutralization assays using vesicular stomatitis virus (VSV)-based pseudotypes bearing arenavirus recombinant glycoproteins. The goal of this article is to review the recent advances in developing laboratory diagnostic assays based on recombinant viral proteins for the diagnosis of VHFs and epidemiological studies on the VHFs caused by arenaviruses.
Text: The virus family Arenaviridae consists of only one genus, but most viruses within this genus can be divided into two different groups: the Old World arenaviruses and the New World arenaviruses (also known as the Tacaribe complex) [1, 2] . The differences between the two groups have been established through the use of serological assays. Most of the arenaviruses cause persistent infection in rodents without any symptoms, and humans acquire a variety of diseases when zoonotically infected. Lymphocytic choriomeningitis virus (LCMV) is the only arenavirus to exhibit a worldwide distribution, and causes illnesses such as meningitis [3, 4] . Congenital LCMV infections have also been reported [4, 5] . Most importantly, viral hemorrhagic fever (VHF) can be caused by several arenaviruses. Lassa fever, caused by the Lassa virus (LASV), an Old World arenavirus, is one of the most devastating VHFs in humans [6] . Hemorrhaging and organ failure occur in a subset of patients infected with this virus, and it is associated with high mortality. Many cases of Lassa fever occur in Western Africa in countries such as Guinea, Sierra Leone, and Nigeria [7] [8] [9] [10] [11] [12] [13] . Tacaribe complex lineage B of the New World arenaviruses consists of the Junin virus (JUNV), Guanarito virus (GUNV), Sabia virus (SABV) and Machupo virus (MACV), the etiological agents of Argentine, Venezuelan, Brazilian, and Bolivian hemorrhagic fevers, respectively [14, 15] . Although genetically distinct from one another, they appear to produce similar symptoms, accompanied by hemorrhaging in humans [14, 15] . These pathogenic New World arenavirus species are closely associated with a specific rodent species [6] .
Humans are usually infected with pathogenic arenaviruses through direct contact with tissue or blood, or after inhaling aerosolized particles from urine, feces, and saliva of infected rodents. After an incubation period of 1-3 weeks, infected individuals abruptly develop fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria, and mucosal bleeding. Neurological problems have also been described, including hearing loss, tremors, and encephalitis. Because the symptoms of pathogenic arenavirus-related illness are varied and nonspecific, the clinical diagnosis is often difficult [14, 16] . Human-to-human transmission may occur via mucosal or cutaneous contact, or through nosocomial contamination [14, 16] . These viruses are also considered to be potential bioterrorism agents [2] .
A number of arenavirus species have been recently discovered as a result of both rodent surveys and disease outbreaks [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . A novel pathogenic New World arenavirus, Chapare virus (CHPV), has been isolated from a fatal case of VHF in Bolivia [20] . In addition, five cases of VHF have been reported in South Africa, and a novel arenavirus, named Lujo virus, was isolated from a patient [17] . The Lujo virus is most distantly related to the other Old World arenaviruses [17] . To date, there is no information concerning the vertebrate host for the Chapare and Lujo viruses.
There is some evidence of endemicity of the Lassa virus in neighboring countries [27, 28] . However, as the magnitude of international trade and travel is continuously increasing, and the perturbation of the environment (due either to human activity or natural ecological changes) may result in behavioral changes of reservoir rodents, highly pathogenic arenaviruses could be introduced to virus-free countries from endemic areas. In fact, more than twenty cases of Lassa fever have been reported outside of the endemic region in areas such as the USA, Canada, Europe, and Japan [29] [30] [31] [32] [33] . It is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of outbreaks of VHFs caused by arenaviruses. However, these arenaviruses are classified as biosafety level (BSL)-4 pathogens, making it difficult to develop diagnostic techniques for these virus infections in laboratories without BSL-4 facilities. To overcome these difficulties, we have established recombinant viral nucleoproteins (rNPs)-based serological assays, such as IgG-enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescence assay (IFA), and antigen (Ag)-capture ELISA for the diagnosis of VHFs caused by highly pathogenic arenaviruses. Furthermore, virus neutralization assays using pseudotype virus-bearing arenavirus GPs have been developed. In this review, we describe the usefulness of such recombinant protein-based diagnostic assays for diagnosing VHFs caused by arenaviruses.
In outbreaks of VHFs, infections are confirmed by various laboratory diagnostic methods. Virus detection is performed by virus isolation, reverse transcription-polymerase chain reaction (RT-PCR), and antigen-capture ELISA. It has been shown that monoclonal antibody panels against pathogenic arenaviruses are useful for detecting viral antigens on the virus-infected cells as well as for investigating of antigenic relationships of arenaviruses [34] [35] [36] . Detection of the virus genome is suitable for a rapid and sensitive diagnosis of VHF patients in the early stage of illness, and extensive reviews of such RT-PCR assays have been described [37, 38] . More recently, progress in the RT-PCR method covering genetic variations of the hemorrhagic fever viruses (HFVs) [39, 40] and a multiplexed oligonucleotide microarray for the differential diagnosis of VHFs have also been reported [41] . On the other hand, antibodies against these viruses can be detected by the indirect immunofluorescence assay (IFA), or IgG-and IgM-ELISA. An IFA detects the antibody in the serum, which is able to bind to the fixed monolayer of the virus-infected cells. Although the interpretation of immunofluorescence results requires experience, the assay has advantages over other methods, since each virus generates a characteristic fluorescence pattern that adds specificity to the assay compared to a simple ELISA readout. A serological diagnosis by the detection of specific IgM and IgG antibodies to the HFVs must be sensitive, specific and reliable, because a misdiagnosis can lead to panic in the general population. An IgM-specific ELISA is suitable for detecting recent infection, but the relevance of IgM testing for acute VHF depends on the virus and the duration of illness; specific IgM is not often present in the very early stage of illness, and patients who die of VHF often fail to seroconvert at all. An IgG-specific ELISA is efficacious, not only in the diagnosis of a large number of VHF cases, especially during convalescence, but also for epidemiological studies in the endemic regions. The detailed methods used for the IFA and IgG-and IgM-ELISAs for the diagnosis of VHF using authentic virus-antigens have been described in detail [42] [43] [44] [45] .
Arenaviruses have a bisegmented, negative-sense, single stranded RNA genome with a unique ambisense coding strategy that produces just four known proteins: a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L) [46] . Of these proteins, the NP is the most abundant in virus-infected cells. Recombinant protein technology could meet the demand for a simple and reliable VHF test system, and recombinant NP (rNP) has been shown to be useful for serological surveys of IgM-and IgG antibodies against arenaviruses [47] [48] [49] [50] .
Recombinant baculoviruses that express the full-length rNP of arenaviruses have been generated [48, 50, 51] . The method used for the purification of arenavirus rNP from insect Tn5 cells infected with recombinant baculoviruses is effective and simple compared to those for Ebola, Marburg, and Crimean-Congo hemorrhagic fever virus rNPs [51] [52] [53] [54] [55] . Most of the arenavirus rNPs expressed in insect cells using the recombinant baculoviruses are crystallized [56] and are solubilized in PBS containing 8M urea. Since the majority of Tn5 cellular proteins are solubilized in PBS containing 2M urea, the arenavirus rNPs in the insoluble fraction in PBS containing 2M urea can be solubilized by sonication in PBS containing 8M urea. After a simple centrifugation of the lysates in PBS containing 8M urea, the supernatant fractions can be used as purified rNP antigens without further purification steps [51] . The control antigen is produced from Tn5 cells infected with baculovirus lacking the polyhedrin gene (ΔP) in the same manner as the arenavirus rNPs ( Figure 1 ).
Purified rNPs. The expression and purification efficiency of arenavirus rNP were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after staining the gels with Coomassie blue. Purified NP antigens with approximate molecular weights of 62 kDa from Luna, LCM, Lassa, Lujo, Junin, Machupo, Guanarito, Sabia, and Chapare viruses and the purified negative control antigen (ΔP) are shown.
As described above, recombinant baculoviruses allow the delivery of rNP antigens without using infectious live arenaviruses. An ELISA plate coated with the predetermined optimal quantity of purified rNPs (approximately 100 ng/well) is used for the IgG-antibody detection assay. An advantage of using recombinant rNP for the IgG-ELISA is that it enables a direct comparison of antibody cross-reactivity among arenavirus rNPs, since antigen preparations of all arenavirus rNPs tested are performed using the same method [51] . Rabbit anti-sera raised against LCMV-rNP and LASV-rNP show cross-reactivity to LASV-rNP and LCMV-rNP, respectively, indicating that rabbit antibodies against rNPs of Old World arenaviruses cross-react with rNPs of other Old World arenaviruses (Table 1 ) [51] . Similarly, rabbit anti-sera generated against JUNV-NP show cross-reactivity to the LASV-rNP and LCMV-rNP, although the reaction is weak. However, rabbit anti-sera against LASV-NP and LCMV-NP show a negative reaction to the JUNV-rNP (Table 1 ) [51] , indicating that rabbit antibodies against JUNV (a pathogenic New World arenavirus) NP might cross-react with the Old World arenavirus NP, whereas antibodies against Old World arenavirus NPs may not be able to react with pathogenic New World arenavirus NPs.
The rNP-based IgG-ELISA has also been used for the characterization of a mouse monoclonal antibody (MAb). Nakauchi et al. [50] have investigated the cross-reactivity of MAbs against JUNV rNP to pathogenic New World arenavirus rNPs, as well as LASV rNP. MAb C11-12 reacts at the same level with the rNPs of all of the pathogenic New World arenaviruses, including JUNV, GTOV, MACV, SABV, and CHPV, indicating that this MAb recognizes an epitope conserved among pathogenic New World arenaviruses. Another MAb, C6-9, reacts specifically with the rNP of JUNV, but does not react with those of the other pathogenic New World arenaviruses [50] . This indicates that MAb C6-9 recognizes a JUNV-specific epitope. None of these MAbs reacts with the rNP of the human pathogenic Old World arenavirus LASV. Thus, the MAb C11-12 is considered to be a broadly reactive MAb against New World arenaviruses, whereas MAb C6-9 is JUNV-specific. These findings have been confirmed by detailed epitope analyses using peptide mapping [50] . Similarly, the cross-reactivity of MAbs against LASV rNP has been analyzed [51] . MAb 4A5 cross-reacts with the Mopeia virus (MOPV) but not with the LCMV rNP. MAb 6C11 cross-reacts with LCMV rNP, while MAb 2-11 does not cross-react with LCMV rNP [51] . Table 1 . Anti-serum reactivity for rNPs of different arenaviruses in IgG ELISAs.
Reactivity for rNP from LASV LCMV JUNV anti-LASV NP
It is important to evaluate whether rNP-based ELISA is useful for the diagnosis of human VHF cases. The specificity of the LASV-rNP-based IgG ELISA has been confirmed by using sera obtained from Lassa fever patients [51] . The Lassa fever patients' sera show a highly positive reaction in the LASV-rNP-based IgG-ELISA, but sera from patients with Argentine hemorrhagic fever (AHF), which is caused by JUNV, do not. The serum from an AHF patient showed a highly positive reaction in the JUNV-rNP-based IgG-ELISA [49] . In addition, it was shown that, using sera obtained from AHF cases, the results of the JUNV rNP-based IgG ELISA correlate well with an authentic JUNV antigen-based IgG ELISA [49] . An IgM-capture ELISA using purified LASV-rNP as an antigen has been developed in the same way as in previous reports [54, 57] and detects an LASV-IgM antibody [58] . In addition, immunoblot assays based on N-terminally truncated LASV rNP have been developed for detecting IgG and IgM antibodies against LASV. These methods may provide a rapid and simple Lassa fever test for use under field conditions [47] .
An IFA using virus-infected cells is a common antibody test for VHF viruses [59] [60] [61] [62] [63] . To avoid the use of highly pathogenic viruses for the antigen preparation, mammalian cells expressing recombinant rNP have been developed [51, 57, [64] [65] [66] [67] [68] . Lassa virus NP antigen for IFA can be prepared simply as described [51] . Briefly, the procedure involves (1) transfecting HeLa cells with a mammalian cell expression vector inserted with the cloned NP cDNA; (2) expanding the stable NP-expressing cells by antibiotic selection; (3) mixing the rNP-expressing cells with un-transfected HeLa cells (at a ratio of 1:1); (4) spotting the cell mixtures onto glass slides, then drying and fixing them in acetone.
In the IFA specific for LASV-NP, antibody positive sera show characteristic granular staining patterns in the cytoplasm (Figure 2 ) [69] , thus making it easy to distinguish positive from negative samples. The specificity of the assay has also been confirmed by using sera obtained from Lassa fever patients [51] . In addition, an IFA using JUNV rNP-expressing HeLa cells has been developed to detect antibodies against JUNV, and the assay has been evaluated by using AHF patients' sera [70] . The LASV-rNP-based antibody detection systems such as ELISA and IFA are suggested to be useful not only for the diagnosis of Lassa fever, but also for seroepidemiological studies of LASV infection. In our preliminary study, approximately 15% of the sera collected from 334 Ghanaians and less than 3% of 280 Zambians showed positive reactions in the LASV-rNP-based IgG ELISA [58] . These results are in agreement with the fact that Lassa fever is endemic to the West African region, including Ghana, but less in the East African region.
For the diagnosis of many viral infections, PCR assays have been shown to have an excellent analytical sensitivity, but the established techniques are limited by their requirement for expensive equipment and technical expertise. Moreover, the high degree of genetic variability of the RNA viruses, including arenavirus and bunyavirus, poses difficulties in selecting primers for RT-PCR assays that can detect all strains of the virus. Since the sensitivity of the Ag-capture ELISA is comparable to that of RT-PCR for several virus-mediated infectious diseases, including Lassa fever and filovirus hemorrhagic fever [51, [71] [72] [73] , the Ag-capture ELISA is a sophisticated approach that can be used for the diagnosis of viral infections. Ag-capture ELISAs detecting viral NP in viremic sera have been widely applied to detect various viruses, since they are the most abundant viral antigens and have highly conserved amino acid sequences [50, 51, 54, 71, 72, 74, 75] . Polyclonal anti-sera or a mixture of MAbs present in the ascetic fluids from animals immunized for HFVs have been used for capture-antibodies in the Ag-capture ELISA [36, [76] [77] [78] [79] . MAbs recognizing conserved epitopes of the rNP are also used as capture antibodies since they have a high specificity for the antigens, and an identification of the epitopes of these MAbs is of crucial importance for the assessment of the specificity and cross-reactivity of the assay system [50, 51, 53, 54, 71, 75] . In order to develop a sensitive diagnostic test for Lassa fever and AHF, rNPs of LASV and JUNV (see above) have been prepared, and newly established MAbs against them have been characterized and used for Ag-capture ELISAs [50, 51] . The Ag-capture ELISA using MAb 4A5 has been confirmed to be useful in the detection of authentic LASV antigen in sera serially collected from hamsters infected with LASV [51] . The sensitivity of the MAb 4A5-based Ag-capture ELISA was similar to that of conventional RT-PCR, suggesting that the Ag-capture ELISA can be efficiently used in the diagnosis of Lassa fever [51] . Therefore, the MAb 4A5-based Ag-capture ELISA is considered to be useful in the diagnosis of Lassa fever. Also, by using MAbs raised against the rNP of JUNV, Ag-capture ELISAs specific for JUNV and broadly reactive to human pathogenic New World arenaviruses have been developed [50] . The Ag-capture ELISA using MAb E4-2 and C11-12 detected the Ags of all of the pathogenic New World arenaviruses tested, including JUNV. On the other hand, the Ag-capture ELISA using MAb C6-9 detects only the JUNV Ag. Considering that the symptoms of JUNV infection in humans are indistinguishable from those due to other pathogenic New World arenaviruses, the Ag capture ELISA using MAb C6-9 may be a useful diagnostic tool, especially for AHF [50] .
The virus neutralization assay is accepted as the "gold standard" serodiagnostic assay to quantify the antibody response to infection and vaccination of a wide variety of viruses associated with human diseases [80] [81] [82] [83] [84] [85] [86] . The presence of neutralizing antibodies is a reliable indicator of protective immunity against VHF [87] [88] [89] . The most direct method for detection of neutralizing antibodies against HFVs is by plaque reduction neutralization tests using infectious viruses. However, because of the high pathogenicity of HFVs to humans and the strict regulation of select agents, only a limited number of laboratories are able to perform such neutralization tests. For many HFVs, replication-incompetent pseudotyped virus particles bearing viral envelope protein (GP) have been shown to mimic the respective HFV infections, thus, neutralization assays using the pseudotypes may be advantageous in some laboratory settings for the detection of antibodies to HFVs without the need for heightened biocontainment requirements.
The VSV-based vector has already been used to generate replication-competent recombinant VSVs to study of the role of GPs of various viruses [90] [91] [92] . Recent advances in producing pseudotype virus particles have enabled the investigation of the virus cell entry, viral tropism, and effect of entry inhibitors, as well as measurement of the neutralization titers, by using human immunodeficiency virus-, feline immunodeficiency virus-, murine leukemia virus-, or VSV-based vectors [86, [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] . Pseudotypes based on VSV have advantages compared with other pseudotypes based on retroviruses for the following reasons. First, the pseudotype virus titer obtained with the VSV system is generally higher than that of the pseudotyped retrovirus system [104] . Second, the infection of target cells with a VSV pseudotype can be readily detected as green fluorescent protein (GFP)-positive cells at 7-16 h post-infection because of the high level of GFP expression in the VSV system [104, 105] . In contrast, the time required for infection in the pseudotyped retrovirus system is 48 h [106, 107] , which is similar to the time required for infectious viruses to replicate to a level that results in plaque-forming or cytopathic effects in infected cells. A high-throughput assay for determining neutralizing antibody titers using VSV pseudotypes expressing secreted alkaline phosphatase [108, 109] or luciferase ( Figure 3 ) has also been developed. We have recently developed a VSV-based pseudotype bearing Lassa virus GP (VSV-LAS-GP) for the detection of neutralizing antibodies in the sera obtained from a Lassa fever patient. An example of the LASV neutralization assay using the VSV pseudotype is shown (Figure 4 ). In the presence of serum from Lassa fever patients, the number of GFP-positive cells (infectivity of VSV-LAS-GP) is significantly reduced compared with the number in the absence of the patient's serum ( Figure 4A ). The control VSV pseudotype bearing VSV GP (VSV-VSV-G) is not neutralized by any sera. When the cut-off serum dilution is set at 50% inhibition of infectivity compared with the infectivity in the absence of the test serum, the neutralization titer of this patient's serum for VSV-LAS-GP is calculated to be 75 ( Figure 4B ). Likewise, a VSV-based pseudotype bearing the Junin virus GP has been developed for the detection of neutralizing antibodies from AHF patients' sera. The accuracy of the results of VSV-based neutralization assays has been confirmed by comparison with the results of the neutralization assay using live Junin virus [70] . The Lujo virus is a new member of the hemorrhagic fever-associated arenavirus family from Zambia and southern Africa, and the virus is classified as a BSL-4 pathogen [17] . The genome sequence analysis of the Lujo virus suggests that the virus is genetically distinct from previously characterized arenaviruses. In order to study the infectivity of this newly identified arenavirus, we have recently developed a luciferase-expressing VSV pseudotype bearing Lujo virus GPC (VSV-Lujo-GP). As shown in Figure 3 , infection with VSV-Lujo-GPC is specifically neutralized by rabbit anti-Lujo GPC serum. Thus, the VSV-Lujo-GP may be a useful tool not only for determining the neutralizing antibody titer within the serum, but also for exploring yet-to-be-defined cellular receptor(s) for Lujo virus infection or for screening inhibitors of the Lujo virus GP-mediated cell entry.
Hemorrhagic fever outbreaks caused by pathogenic arenaviruses result in high fatality rates. A rapid and accurate diagnosis is a critical first step in any outbreak. Serologic diagnostic methods for VHFs most often employ an ELISA, IFA, and/or virus neutralization assay. Diagnostic methods using recombinant viral proteins have been developed and their utilities for diagnosing of VHF have been reviewed. IgG-and IgM-ELISAs and IFAs using rNPs as antigens are useful for the detection of antibodies induced in the patients' sera. These methods are also useful for seroepidemiological surveys for HFVs. Ag-capture ELISAs using MAbs to the arenavirus rNPs are specific for the virus species or can be broadly reactive for New World arenaviruses, depending on the MAb used. Furthermore, the VSV-based pseudotype system provides a safe and rapid tool for measuring virus neutralizing antibody titers, as well as a model to analyze the entry of the respective arenavirus in susceptible cells without using live arenaviruses. Recent discoveries of novel arenavirus species [17, 26, 110] and their potential to evolve predominantly via host switching, rather than with their hosts [110, 111] , suggest that an unknown pathogenic arenavirus may emerge in the future, and that the diagnostic methods for VHF caused by arenaviruses should thus be further developed and improved.
|
What is the incubation period for arenavirus?
| 5,273 |
1-3 weeks
| 4,195 |
1,606 |
Serological Assays Based on Recombinant Viral Proteins for the Diagnosis of Arenavirus Hemorrhagic Fevers
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3497043/
SHA: f1d308db379b3c293bcfc8fe251c043fe8842358
Authors: Fukushi, Shuetsu; Tani, Hideki; Yoshikawa, Tomoki; Saijo, Masayuki; Morikawa, Shigeru
Date: 2012-10-12
DOI: 10.3390/v4102097
License: cc-by
Abstract: The family Arenaviridae, genus Arenavirus, consists of two phylogenetically independent groups: Old World (OW) and New World (NW) complexes. The Lassa and Lujo viruses in the OW complex and the Guanarito, Junin, Machupo, Sabia, and Chapare viruses in the NW complex cause viral hemorrhagic fever (VHF) in humans, leading to serious public health concerns. These viruses are also considered potential bioterrorism agents. Therefore, it is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of arenavirus outbreaks. However, these arenaviruses are classified as BSL-4 pathogens, thus making it difficult to develop diagnostic techniques for these virus infections in institutes without BSL-4 facilities. To overcome these difficulties, antibody detection systems in the form of an enzyme-linked immunosorbent assay (ELISA) and an indirect immunofluorescence assay were developed using recombinant nucleoproteins (rNPs) derived from these viruses. Furthermore, several antigen-detection assays were developed. For example, novel monoclonal antibodies (mAbs) to the rNPs of Lassa and Junin viruses were generated. Sandwich antigen-capture (Ag-capture) ELISAs using these mAbs as capture antibodies were developed and confirmed to be sensitive and specific for detecting the respective arenavirus NPs. These rNP-based assays were proposed to be useful not only for an etiological diagnosis of VHFs, but also for seroepidemiological studies on VHFs. We recently developed arenavirus neutralization assays using vesicular stomatitis virus (VSV)-based pseudotypes bearing arenavirus recombinant glycoproteins. The goal of this article is to review the recent advances in developing laboratory diagnostic assays based on recombinant viral proteins for the diagnosis of VHFs and epidemiological studies on the VHFs caused by arenaviruses.
Text: The virus family Arenaviridae consists of only one genus, but most viruses within this genus can be divided into two different groups: the Old World arenaviruses and the New World arenaviruses (also known as the Tacaribe complex) [1, 2] . The differences between the two groups have been established through the use of serological assays. Most of the arenaviruses cause persistent infection in rodents without any symptoms, and humans acquire a variety of diseases when zoonotically infected. Lymphocytic choriomeningitis virus (LCMV) is the only arenavirus to exhibit a worldwide distribution, and causes illnesses such as meningitis [3, 4] . Congenital LCMV infections have also been reported [4, 5] . Most importantly, viral hemorrhagic fever (VHF) can be caused by several arenaviruses. Lassa fever, caused by the Lassa virus (LASV), an Old World arenavirus, is one of the most devastating VHFs in humans [6] . Hemorrhaging and organ failure occur in a subset of patients infected with this virus, and it is associated with high mortality. Many cases of Lassa fever occur in Western Africa in countries such as Guinea, Sierra Leone, and Nigeria [7] [8] [9] [10] [11] [12] [13] . Tacaribe complex lineage B of the New World arenaviruses consists of the Junin virus (JUNV), Guanarito virus (GUNV), Sabia virus (SABV) and Machupo virus (MACV), the etiological agents of Argentine, Venezuelan, Brazilian, and Bolivian hemorrhagic fevers, respectively [14, 15] . Although genetically distinct from one another, they appear to produce similar symptoms, accompanied by hemorrhaging in humans [14, 15] . These pathogenic New World arenavirus species are closely associated with a specific rodent species [6] .
Humans are usually infected with pathogenic arenaviruses through direct contact with tissue or blood, or after inhaling aerosolized particles from urine, feces, and saliva of infected rodents. After an incubation period of 1-3 weeks, infected individuals abruptly develop fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria, and mucosal bleeding. Neurological problems have also been described, including hearing loss, tremors, and encephalitis. Because the symptoms of pathogenic arenavirus-related illness are varied and nonspecific, the clinical diagnosis is often difficult [14, 16] . Human-to-human transmission may occur via mucosal or cutaneous contact, or through nosocomial contamination [14, 16] . These viruses are also considered to be potential bioterrorism agents [2] .
A number of arenavirus species have been recently discovered as a result of both rodent surveys and disease outbreaks [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . A novel pathogenic New World arenavirus, Chapare virus (CHPV), has been isolated from a fatal case of VHF in Bolivia [20] . In addition, five cases of VHF have been reported in South Africa, and a novel arenavirus, named Lujo virus, was isolated from a patient [17] . The Lujo virus is most distantly related to the other Old World arenaviruses [17] . To date, there is no information concerning the vertebrate host for the Chapare and Lujo viruses.
There is some evidence of endemicity of the Lassa virus in neighboring countries [27, 28] . However, as the magnitude of international trade and travel is continuously increasing, and the perturbation of the environment (due either to human activity or natural ecological changes) may result in behavioral changes of reservoir rodents, highly pathogenic arenaviruses could be introduced to virus-free countries from endemic areas. In fact, more than twenty cases of Lassa fever have been reported outside of the endemic region in areas such as the USA, Canada, Europe, and Japan [29] [30] [31] [32] [33] . It is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of outbreaks of VHFs caused by arenaviruses. However, these arenaviruses are classified as biosafety level (BSL)-4 pathogens, making it difficult to develop diagnostic techniques for these virus infections in laboratories without BSL-4 facilities. To overcome these difficulties, we have established recombinant viral nucleoproteins (rNPs)-based serological assays, such as IgG-enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescence assay (IFA), and antigen (Ag)-capture ELISA for the diagnosis of VHFs caused by highly pathogenic arenaviruses. Furthermore, virus neutralization assays using pseudotype virus-bearing arenavirus GPs have been developed. In this review, we describe the usefulness of such recombinant protein-based diagnostic assays for diagnosing VHFs caused by arenaviruses.
In outbreaks of VHFs, infections are confirmed by various laboratory diagnostic methods. Virus detection is performed by virus isolation, reverse transcription-polymerase chain reaction (RT-PCR), and antigen-capture ELISA. It has been shown that monoclonal antibody panels against pathogenic arenaviruses are useful for detecting viral antigens on the virus-infected cells as well as for investigating of antigenic relationships of arenaviruses [34] [35] [36] . Detection of the virus genome is suitable for a rapid and sensitive diagnosis of VHF patients in the early stage of illness, and extensive reviews of such RT-PCR assays have been described [37, 38] . More recently, progress in the RT-PCR method covering genetic variations of the hemorrhagic fever viruses (HFVs) [39, 40] and a multiplexed oligonucleotide microarray for the differential diagnosis of VHFs have also been reported [41] . On the other hand, antibodies against these viruses can be detected by the indirect immunofluorescence assay (IFA), or IgG-and IgM-ELISA. An IFA detects the antibody in the serum, which is able to bind to the fixed monolayer of the virus-infected cells. Although the interpretation of immunofluorescence results requires experience, the assay has advantages over other methods, since each virus generates a characteristic fluorescence pattern that adds specificity to the assay compared to a simple ELISA readout. A serological diagnosis by the detection of specific IgM and IgG antibodies to the HFVs must be sensitive, specific and reliable, because a misdiagnosis can lead to panic in the general population. An IgM-specific ELISA is suitable for detecting recent infection, but the relevance of IgM testing for acute VHF depends on the virus and the duration of illness; specific IgM is not often present in the very early stage of illness, and patients who die of VHF often fail to seroconvert at all. An IgG-specific ELISA is efficacious, not only in the diagnosis of a large number of VHF cases, especially during convalescence, but also for epidemiological studies in the endemic regions. The detailed methods used for the IFA and IgG-and IgM-ELISAs for the diagnosis of VHF using authentic virus-antigens have been described in detail [42] [43] [44] [45] .
Arenaviruses have a bisegmented, negative-sense, single stranded RNA genome with a unique ambisense coding strategy that produces just four known proteins: a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L) [46] . Of these proteins, the NP is the most abundant in virus-infected cells. Recombinant protein technology could meet the demand for a simple and reliable VHF test system, and recombinant NP (rNP) has been shown to be useful for serological surveys of IgM-and IgG antibodies against arenaviruses [47] [48] [49] [50] .
Recombinant baculoviruses that express the full-length rNP of arenaviruses have been generated [48, 50, 51] . The method used for the purification of arenavirus rNP from insect Tn5 cells infected with recombinant baculoviruses is effective and simple compared to those for Ebola, Marburg, and Crimean-Congo hemorrhagic fever virus rNPs [51] [52] [53] [54] [55] . Most of the arenavirus rNPs expressed in insect cells using the recombinant baculoviruses are crystallized [56] and are solubilized in PBS containing 8M urea. Since the majority of Tn5 cellular proteins are solubilized in PBS containing 2M urea, the arenavirus rNPs in the insoluble fraction in PBS containing 2M urea can be solubilized by sonication in PBS containing 8M urea. After a simple centrifugation of the lysates in PBS containing 8M urea, the supernatant fractions can be used as purified rNP antigens without further purification steps [51] . The control antigen is produced from Tn5 cells infected with baculovirus lacking the polyhedrin gene (ΔP) in the same manner as the arenavirus rNPs ( Figure 1 ).
Purified rNPs. The expression and purification efficiency of arenavirus rNP were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after staining the gels with Coomassie blue. Purified NP antigens with approximate molecular weights of 62 kDa from Luna, LCM, Lassa, Lujo, Junin, Machupo, Guanarito, Sabia, and Chapare viruses and the purified negative control antigen (ΔP) are shown.
As described above, recombinant baculoviruses allow the delivery of rNP antigens without using infectious live arenaviruses. An ELISA plate coated with the predetermined optimal quantity of purified rNPs (approximately 100 ng/well) is used for the IgG-antibody detection assay. An advantage of using recombinant rNP for the IgG-ELISA is that it enables a direct comparison of antibody cross-reactivity among arenavirus rNPs, since antigen preparations of all arenavirus rNPs tested are performed using the same method [51] . Rabbit anti-sera raised against LCMV-rNP and LASV-rNP show cross-reactivity to LASV-rNP and LCMV-rNP, respectively, indicating that rabbit antibodies against rNPs of Old World arenaviruses cross-react with rNPs of other Old World arenaviruses (Table 1 ) [51] . Similarly, rabbit anti-sera generated against JUNV-NP show cross-reactivity to the LASV-rNP and LCMV-rNP, although the reaction is weak. However, rabbit anti-sera against LASV-NP and LCMV-NP show a negative reaction to the JUNV-rNP (Table 1 ) [51] , indicating that rabbit antibodies against JUNV (a pathogenic New World arenavirus) NP might cross-react with the Old World arenavirus NP, whereas antibodies against Old World arenavirus NPs may not be able to react with pathogenic New World arenavirus NPs.
The rNP-based IgG-ELISA has also been used for the characterization of a mouse monoclonal antibody (MAb). Nakauchi et al. [50] have investigated the cross-reactivity of MAbs against JUNV rNP to pathogenic New World arenavirus rNPs, as well as LASV rNP. MAb C11-12 reacts at the same level with the rNPs of all of the pathogenic New World arenaviruses, including JUNV, GTOV, MACV, SABV, and CHPV, indicating that this MAb recognizes an epitope conserved among pathogenic New World arenaviruses. Another MAb, C6-9, reacts specifically with the rNP of JUNV, but does not react with those of the other pathogenic New World arenaviruses [50] . This indicates that MAb C6-9 recognizes a JUNV-specific epitope. None of these MAbs reacts with the rNP of the human pathogenic Old World arenavirus LASV. Thus, the MAb C11-12 is considered to be a broadly reactive MAb against New World arenaviruses, whereas MAb C6-9 is JUNV-specific. These findings have been confirmed by detailed epitope analyses using peptide mapping [50] . Similarly, the cross-reactivity of MAbs against LASV rNP has been analyzed [51] . MAb 4A5 cross-reacts with the Mopeia virus (MOPV) but not with the LCMV rNP. MAb 6C11 cross-reacts with LCMV rNP, while MAb 2-11 does not cross-react with LCMV rNP [51] . Table 1 . Anti-serum reactivity for rNPs of different arenaviruses in IgG ELISAs.
Reactivity for rNP from LASV LCMV JUNV anti-LASV NP
It is important to evaluate whether rNP-based ELISA is useful for the diagnosis of human VHF cases. The specificity of the LASV-rNP-based IgG ELISA has been confirmed by using sera obtained from Lassa fever patients [51] . The Lassa fever patients' sera show a highly positive reaction in the LASV-rNP-based IgG-ELISA, but sera from patients with Argentine hemorrhagic fever (AHF), which is caused by JUNV, do not. The serum from an AHF patient showed a highly positive reaction in the JUNV-rNP-based IgG-ELISA [49] . In addition, it was shown that, using sera obtained from AHF cases, the results of the JUNV rNP-based IgG ELISA correlate well with an authentic JUNV antigen-based IgG ELISA [49] . An IgM-capture ELISA using purified LASV-rNP as an antigen has been developed in the same way as in previous reports [54, 57] and detects an LASV-IgM antibody [58] . In addition, immunoblot assays based on N-terminally truncated LASV rNP have been developed for detecting IgG and IgM antibodies against LASV. These methods may provide a rapid and simple Lassa fever test for use under field conditions [47] .
An IFA using virus-infected cells is a common antibody test for VHF viruses [59] [60] [61] [62] [63] . To avoid the use of highly pathogenic viruses for the antigen preparation, mammalian cells expressing recombinant rNP have been developed [51, 57, [64] [65] [66] [67] [68] . Lassa virus NP antigen for IFA can be prepared simply as described [51] . Briefly, the procedure involves (1) transfecting HeLa cells with a mammalian cell expression vector inserted with the cloned NP cDNA; (2) expanding the stable NP-expressing cells by antibiotic selection; (3) mixing the rNP-expressing cells with un-transfected HeLa cells (at a ratio of 1:1); (4) spotting the cell mixtures onto glass slides, then drying and fixing them in acetone.
In the IFA specific for LASV-NP, antibody positive sera show characteristic granular staining patterns in the cytoplasm (Figure 2 ) [69] , thus making it easy to distinguish positive from negative samples. The specificity of the assay has also been confirmed by using sera obtained from Lassa fever patients [51] . In addition, an IFA using JUNV rNP-expressing HeLa cells has been developed to detect antibodies against JUNV, and the assay has been evaluated by using AHF patients' sera [70] . The LASV-rNP-based antibody detection systems such as ELISA and IFA are suggested to be useful not only for the diagnosis of Lassa fever, but also for seroepidemiological studies of LASV infection. In our preliminary study, approximately 15% of the sera collected from 334 Ghanaians and less than 3% of 280 Zambians showed positive reactions in the LASV-rNP-based IgG ELISA [58] . These results are in agreement with the fact that Lassa fever is endemic to the West African region, including Ghana, but less in the East African region.
For the diagnosis of many viral infections, PCR assays have been shown to have an excellent analytical sensitivity, but the established techniques are limited by their requirement for expensive equipment and technical expertise. Moreover, the high degree of genetic variability of the RNA viruses, including arenavirus and bunyavirus, poses difficulties in selecting primers for RT-PCR assays that can detect all strains of the virus. Since the sensitivity of the Ag-capture ELISA is comparable to that of RT-PCR for several virus-mediated infectious diseases, including Lassa fever and filovirus hemorrhagic fever [51, [71] [72] [73] , the Ag-capture ELISA is a sophisticated approach that can be used for the diagnosis of viral infections. Ag-capture ELISAs detecting viral NP in viremic sera have been widely applied to detect various viruses, since they are the most abundant viral antigens and have highly conserved amino acid sequences [50, 51, 54, 71, 72, 74, 75] . Polyclonal anti-sera or a mixture of MAbs present in the ascetic fluids from animals immunized for HFVs have been used for capture-antibodies in the Ag-capture ELISA [36, [76] [77] [78] [79] . MAbs recognizing conserved epitopes of the rNP are also used as capture antibodies since they have a high specificity for the antigens, and an identification of the epitopes of these MAbs is of crucial importance for the assessment of the specificity and cross-reactivity of the assay system [50, 51, 53, 54, 71, 75] . In order to develop a sensitive diagnostic test for Lassa fever and AHF, rNPs of LASV and JUNV (see above) have been prepared, and newly established MAbs against them have been characterized and used for Ag-capture ELISAs [50, 51] . The Ag-capture ELISA using MAb 4A5 has been confirmed to be useful in the detection of authentic LASV antigen in sera serially collected from hamsters infected with LASV [51] . The sensitivity of the MAb 4A5-based Ag-capture ELISA was similar to that of conventional RT-PCR, suggesting that the Ag-capture ELISA can be efficiently used in the diagnosis of Lassa fever [51] . Therefore, the MAb 4A5-based Ag-capture ELISA is considered to be useful in the diagnosis of Lassa fever. Also, by using MAbs raised against the rNP of JUNV, Ag-capture ELISAs specific for JUNV and broadly reactive to human pathogenic New World arenaviruses have been developed [50] . The Ag-capture ELISA using MAb E4-2 and C11-12 detected the Ags of all of the pathogenic New World arenaviruses tested, including JUNV. On the other hand, the Ag-capture ELISA using MAb C6-9 detects only the JUNV Ag. Considering that the symptoms of JUNV infection in humans are indistinguishable from those due to other pathogenic New World arenaviruses, the Ag capture ELISA using MAb C6-9 may be a useful diagnostic tool, especially for AHF [50] .
The virus neutralization assay is accepted as the "gold standard" serodiagnostic assay to quantify the antibody response to infection and vaccination of a wide variety of viruses associated with human diseases [80] [81] [82] [83] [84] [85] [86] . The presence of neutralizing antibodies is a reliable indicator of protective immunity against VHF [87] [88] [89] . The most direct method for detection of neutralizing antibodies against HFVs is by plaque reduction neutralization tests using infectious viruses. However, because of the high pathogenicity of HFVs to humans and the strict regulation of select agents, only a limited number of laboratories are able to perform such neutralization tests. For many HFVs, replication-incompetent pseudotyped virus particles bearing viral envelope protein (GP) have been shown to mimic the respective HFV infections, thus, neutralization assays using the pseudotypes may be advantageous in some laboratory settings for the detection of antibodies to HFVs without the need for heightened biocontainment requirements.
The VSV-based vector has already been used to generate replication-competent recombinant VSVs to study of the role of GPs of various viruses [90] [91] [92] . Recent advances in producing pseudotype virus particles have enabled the investigation of the virus cell entry, viral tropism, and effect of entry inhibitors, as well as measurement of the neutralization titers, by using human immunodeficiency virus-, feline immunodeficiency virus-, murine leukemia virus-, or VSV-based vectors [86, [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] . Pseudotypes based on VSV have advantages compared with other pseudotypes based on retroviruses for the following reasons. First, the pseudotype virus titer obtained with the VSV system is generally higher than that of the pseudotyped retrovirus system [104] . Second, the infection of target cells with a VSV pseudotype can be readily detected as green fluorescent protein (GFP)-positive cells at 7-16 h post-infection because of the high level of GFP expression in the VSV system [104, 105] . In contrast, the time required for infection in the pseudotyped retrovirus system is 48 h [106, 107] , which is similar to the time required for infectious viruses to replicate to a level that results in plaque-forming or cytopathic effects in infected cells. A high-throughput assay for determining neutralizing antibody titers using VSV pseudotypes expressing secreted alkaline phosphatase [108, 109] or luciferase ( Figure 3 ) has also been developed. We have recently developed a VSV-based pseudotype bearing Lassa virus GP (VSV-LAS-GP) for the detection of neutralizing antibodies in the sera obtained from a Lassa fever patient. An example of the LASV neutralization assay using the VSV pseudotype is shown (Figure 4 ). In the presence of serum from Lassa fever patients, the number of GFP-positive cells (infectivity of VSV-LAS-GP) is significantly reduced compared with the number in the absence of the patient's serum ( Figure 4A ). The control VSV pseudotype bearing VSV GP (VSV-VSV-G) is not neutralized by any sera. When the cut-off serum dilution is set at 50% inhibition of infectivity compared with the infectivity in the absence of the test serum, the neutralization titer of this patient's serum for VSV-LAS-GP is calculated to be 75 ( Figure 4B ). Likewise, a VSV-based pseudotype bearing the Junin virus GP has been developed for the detection of neutralizing antibodies from AHF patients' sera. The accuracy of the results of VSV-based neutralization assays has been confirmed by comparison with the results of the neutralization assay using live Junin virus [70] . The Lujo virus is a new member of the hemorrhagic fever-associated arenavirus family from Zambia and southern Africa, and the virus is classified as a BSL-4 pathogen [17] . The genome sequence analysis of the Lujo virus suggests that the virus is genetically distinct from previously characterized arenaviruses. In order to study the infectivity of this newly identified arenavirus, we have recently developed a luciferase-expressing VSV pseudotype bearing Lujo virus GPC (VSV-Lujo-GP). As shown in Figure 3 , infection with VSV-Lujo-GPC is specifically neutralized by rabbit anti-Lujo GPC serum. Thus, the VSV-Lujo-GP may be a useful tool not only for determining the neutralizing antibody titer within the serum, but also for exploring yet-to-be-defined cellular receptor(s) for Lujo virus infection or for screening inhibitors of the Lujo virus GP-mediated cell entry.
Hemorrhagic fever outbreaks caused by pathogenic arenaviruses result in high fatality rates. A rapid and accurate diagnosis is a critical first step in any outbreak. Serologic diagnostic methods for VHFs most often employ an ELISA, IFA, and/or virus neutralization assay. Diagnostic methods using recombinant viral proteins have been developed and their utilities for diagnosing of VHF have been reviewed. IgG-and IgM-ELISAs and IFAs using rNPs as antigens are useful for the detection of antibodies induced in the patients' sera. These methods are also useful for seroepidemiological surveys for HFVs. Ag-capture ELISAs using MAbs to the arenavirus rNPs are specific for the virus species or can be broadly reactive for New World arenaviruses, depending on the MAb used. Furthermore, the VSV-based pseudotype system provides a safe and rapid tool for measuring virus neutralizing antibody titers, as well as a model to analyze the entry of the respective arenavirus in susceptible cells without using live arenaviruses. Recent discoveries of novel arenavirus species [17, 26, 110] and their potential to evolve predominantly via host switching, rather than with their hosts [110, 111] , suggest that an unknown pathogenic arenavirus may emerge in the future, and that the diagnostic methods for VHF caused by arenaviruses should thus be further developed and improved.
|
What is the structure of the Arenavirus?
| 5,274 |
bisegmented, negative-sense, single stranded RNA genome
| 9,282 |
1,606 |
Serological Assays Based on Recombinant Viral Proteins for the Diagnosis of Arenavirus Hemorrhagic Fevers
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3497043/
SHA: f1d308db379b3c293bcfc8fe251c043fe8842358
Authors: Fukushi, Shuetsu; Tani, Hideki; Yoshikawa, Tomoki; Saijo, Masayuki; Morikawa, Shigeru
Date: 2012-10-12
DOI: 10.3390/v4102097
License: cc-by
Abstract: The family Arenaviridae, genus Arenavirus, consists of two phylogenetically independent groups: Old World (OW) and New World (NW) complexes. The Lassa and Lujo viruses in the OW complex and the Guanarito, Junin, Machupo, Sabia, and Chapare viruses in the NW complex cause viral hemorrhagic fever (VHF) in humans, leading to serious public health concerns. These viruses are also considered potential bioterrorism agents. Therefore, it is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of arenavirus outbreaks. However, these arenaviruses are classified as BSL-4 pathogens, thus making it difficult to develop diagnostic techniques for these virus infections in institutes without BSL-4 facilities. To overcome these difficulties, antibody detection systems in the form of an enzyme-linked immunosorbent assay (ELISA) and an indirect immunofluorescence assay were developed using recombinant nucleoproteins (rNPs) derived from these viruses. Furthermore, several antigen-detection assays were developed. For example, novel monoclonal antibodies (mAbs) to the rNPs of Lassa and Junin viruses were generated. Sandwich antigen-capture (Ag-capture) ELISAs using these mAbs as capture antibodies were developed and confirmed to be sensitive and specific for detecting the respective arenavirus NPs. These rNP-based assays were proposed to be useful not only for an etiological diagnosis of VHFs, but also for seroepidemiological studies on VHFs. We recently developed arenavirus neutralization assays using vesicular stomatitis virus (VSV)-based pseudotypes bearing arenavirus recombinant glycoproteins. The goal of this article is to review the recent advances in developing laboratory diagnostic assays based on recombinant viral proteins for the diagnosis of VHFs and epidemiological studies on the VHFs caused by arenaviruses.
Text: The virus family Arenaviridae consists of only one genus, but most viruses within this genus can be divided into two different groups: the Old World arenaviruses and the New World arenaviruses (also known as the Tacaribe complex) [1, 2] . The differences between the two groups have been established through the use of serological assays. Most of the arenaviruses cause persistent infection in rodents without any symptoms, and humans acquire a variety of diseases when zoonotically infected. Lymphocytic choriomeningitis virus (LCMV) is the only arenavirus to exhibit a worldwide distribution, and causes illnesses such as meningitis [3, 4] . Congenital LCMV infections have also been reported [4, 5] . Most importantly, viral hemorrhagic fever (VHF) can be caused by several arenaviruses. Lassa fever, caused by the Lassa virus (LASV), an Old World arenavirus, is one of the most devastating VHFs in humans [6] . Hemorrhaging and organ failure occur in a subset of patients infected with this virus, and it is associated with high mortality. Many cases of Lassa fever occur in Western Africa in countries such as Guinea, Sierra Leone, and Nigeria [7] [8] [9] [10] [11] [12] [13] . Tacaribe complex lineage B of the New World arenaviruses consists of the Junin virus (JUNV), Guanarito virus (GUNV), Sabia virus (SABV) and Machupo virus (MACV), the etiological agents of Argentine, Venezuelan, Brazilian, and Bolivian hemorrhagic fevers, respectively [14, 15] . Although genetically distinct from one another, they appear to produce similar symptoms, accompanied by hemorrhaging in humans [14, 15] . These pathogenic New World arenavirus species are closely associated with a specific rodent species [6] .
Humans are usually infected with pathogenic arenaviruses through direct contact with tissue or blood, or after inhaling aerosolized particles from urine, feces, and saliva of infected rodents. After an incubation period of 1-3 weeks, infected individuals abruptly develop fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria, and mucosal bleeding. Neurological problems have also been described, including hearing loss, tremors, and encephalitis. Because the symptoms of pathogenic arenavirus-related illness are varied and nonspecific, the clinical diagnosis is often difficult [14, 16] . Human-to-human transmission may occur via mucosal or cutaneous contact, or through nosocomial contamination [14, 16] . These viruses are also considered to be potential bioterrorism agents [2] .
A number of arenavirus species have been recently discovered as a result of both rodent surveys and disease outbreaks [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . A novel pathogenic New World arenavirus, Chapare virus (CHPV), has been isolated from a fatal case of VHF in Bolivia [20] . In addition, five cases of VHF have been reported in South Africa, and a novel arenavirus, named Lujo virus, was isolated from a patient [17] . The Lujo virus is most distantly related to the other Old World arenaviruses [17] . To date, there is no information concerning the vertebrate host for the Chapare and Lujo viruses.
There is some evidence of endemicity of the Lassa virus in neighboring countries [27, 28] . However, as the magnitude of international trade and travel is continuously increasing, and the perturbation of the environment (due either to human activity or natural ecological changes) may result in behavioral changes of reservoir rodents, highly pathogenic arenaviruses could be introduced to virus-free countries from endemic areas. In fact, more than twenty cases of Lassa fever have been reported outside of the endemic region in areas such as the USA, Canada, Europe, and Japan [29] [30] [31] [32] [33] . It is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of outbreaks of VHFs caused by arenaviruses. However, these arenaviruses are classified as biosafety level (BSL)-4 pathogens, making it difficult to develop diagnostic techniques for these virus infections in laboratories without BSL-4 facilities. To overcome these difficulties, we have established recombinant viral nucleoproteins (rNPs)-based serological assays, such as IgG-enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescence assay (IFA), and antigen (Ag)-capture ELISA for the diagnosis of VHFs caused by highly pathogenic arenaviruses. Furthermore, virus neutralization assays using pseudotype virus-bearing arenavirus GPs have been developed. In this review, we describe the usefulness of such recombinant protein-based diagnostic assays for diagnosing VHFs caused by arenaviruses.
In outbreaks of VHFs, infections are confirmed by various laboratory diagnostic methods. Virus detection is performed by virus isolation, reverse transcription-polymerase chain reaction (RT-PCR), and antigen-capture ELISA. It has been shown that monoclonal antibody panels against pathogenic arenaviruses are useful for detecting viral antigens on the virus-infected cells as well as for investigating of antigenic relationships of arenaviruses [34] [35] [36] . Detection of the virus genome is suitable for a rapid and sensitive diagnosis of VHF patients in the early stage of illness, and extensive reviews of such RT-PCR assays have been described [37, 38] . More recently, progress in the RT-PCR method covering genetic variations of the hemorrhagic fever viruses (HFVs) [39, 40] and a multiplexed oligonucleotide microarray for the differential diagnosis of VHFs have also been reported [41] . On the other hand, antibodies against these viruses can be detected by the indirect immunofluorescence assay (IFA), or IgG-and IgM-ELISA. An IFA detects the antibody in the serum, which is able to bind to the fixed monolayer of the virus-infected cells. Although the interpretation of immunofluorescence results requires experience, the assay has advantages over other methods, since each virus generates a characteristic fluorescence pattern that adds specificity to the assay compared to a simple ELISA readout. A serological diagnosis by the detection of specific IgM and IgG antibodies to the HFVs must be sensitive, specific and reliable, because a misdiagnosis can lead to panic in the general population. An IgM-specific ELISA is suitable for detecting recent infection, but the relevance of IgM testing for acute VHF depends on the virus and the duration of illness; specific IgM is not often present in the very early stage of illness, and patients who die of VHF often fail to seroconvert at all. An IgG-specific ELISA is efficacious, not only in the diagnosis of a large number of VHF cases, especially during convalescence, but also for epidemiological studies in the endemic regions. The detailed methods used for the IFA and IgG-and IgM-ELISAs for the diagnosis of VHF using authentic virus-antigens have been described in detail [42] [43] [44] [45] .
Arenaviruses have a bisegmented, negative-sense, single stranded RNA genome with a unique ambisense coding strategy that produces just four known proteins: a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L) [46] . Of these proteins, the NP is the most abundant in virus-infected cells. Recombinant protein technology could meet the demand for a simple and reliable VHF test system, and recombinant NP (rNP) has been shown to be useful for serological surveys of IgM-and IgG antibodies against arenaviruses [47] [48] [49] [50] .
Recombinant baculoviruses that express the full-length rNP of arenaviruses have been generated [48, 50, 51] . The method used for the purification of arenavirus rNP from insect Tn5 cells infected with recombinant baculoviruses is effective and simple compared to those for Ebola, Marburg, and Crimean-Congo hemorrhagic fever virus rNPs [51] [52] [53] [54] [55] . Most of the arenavirus rNPs expressed in insect cells using the recombinant baculoviruses are crystallized [56] and are solubilized in PBS containing 8M urea. Since the majority of Tn5 cellular proteins are solubilized in PBS containing 2M urea, the arenavirus rNPs in the insoluble fraction in PBS containing 2M urea can be solubilized by sonication in PBS containing 8M urea. After a simple centrifugation of the lysates in PBS containing 8M urea, the supernatant fractions can be used as purified rNP antigens without further purification steps [51] . The control antigen is produced from Tn5 cells infected with baculovirus lacking the polyhedrin gene (ΔP) in the same manner as the arenavirus rNPs ( Figure 1 ).
Purified rNPs. The expression and purification efficiency of arenavirus rNP were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after staining the gels with Coomassie blue. Purified NP antigens with approximate molecular weights of 62 kDa from Luna, LCM, Lassa, Lujo, Junin, Machupo, Guanarito, Sabia, and Chapare viruses and the purified negative control antigen (ΔP) are shown.
As described above, recombinant baculoviruses allow the delivery of rNP antigens without using infectious live arenaviruses. An ELISA plate coated with the predetermined optimal quantity of purified rNPs (approximately 100 ng/well) is used for the IgG-antibody detection assay. An advantage of using recombinant rNP for the IgG-ELISA is that it enables a direct comparison of antibody cross-reactivity among arenavirus rNPs, since antigen preparations of all arenavirus rNPs tested are performed using the same method [51] . Rabbit anti-sera raised against LCMV-rNP and LASV-rNP show cross-reactivity to LASV-rNP and LCMV-rNP, respectively, indicating that rabbit antibodies against rNPs of Old World arenaviruses cross-react with rNPs of other Old World arenaviruses (Table 1 ) [51] . Similarly, rabbit anti-sera generated against JUNV-NP show cross-reactivity to the LASV-rNP and LCMV-rNP, although the reaction is weak. However, rabbit anti-sera against LASV-NP and LCMV-NP show a negative reaction to the JUNV-rNP (Table 1 ) [51] , indicating that rabbit antibodies against JUNV (a pathogenic New World arenavirus) NP might cross-react with the Old World arenavirus NP, whereas antibodies against Old World arenavirus NPs may not be able to react with pathogenic New World arenavirus NPs.
The rNP-based IgG-ELISA has also been used for the characterization of a mouse monoclonal antibody (MAb). Nakauchi et al. [50] have investigated the cross-reactivity of MAbs against JUNV rNP to pathogenic New World arenavirus rNPs, as well as LASV rNP. MAb C11-12 reacts at the same level with the rNPs of all of the pathogenic New World arenaviruses, including JUNV, GTOV, MACV, SABV, and CHPV, indicating that this MAb recognizes an epitope conserved among pathogenic New World arenaviruses. Another MAb, C6-9, reacts specifically with the rNP of JUNV, but does not react with those of the other pathogenic New World arenaviruses [50] . This indicates that MAb C6-9 recognizes a JUNV-specific epitope. None of these MAbs reacts with the rNP of the human pathogenic Old World arenavirus LASV. Thus, the MAb C11-12 is considered to be a broadly reactive MAb against New World arenaviruses, whereas MAb C6-9 is JUNV-specific. These findings have been confirmed by detailed epitope analyses using peptide mapping [50] . Similarly, the cross-reactivity of MAbs against LASV rNP has been analyzed [51] . MAb 4A5 cross-reacts with the Mopeia virus (MOPV) but not with the LCMV rNP. MAb 6C11 cross-reacts with LCMV rNP, while MAb 2-11 does not cross-react with LCMV rNP [51] . Table 1 . Anti-serum reactivity for rNPs of different arenaviruses in IgG ELISAs.
Reactivity for rNP from LASV LCMV JUNV anti-LASV NP
It is important to evaluate whether rNP-based ELISA is useful for the diagnosis of human VHF cases. The specificity of the LASV-rNP-based IgG ELISA has been confirmed by using sera obtained from Lassa fever patients [51] . The Lassa fever patients' sera show a highly positive reaction in the LASV-rNP-based IgG-ELISA, but sera from patients with Argentine hemorrhagic fever (AHF), which is caused by JUNV, do not. The serum from an AHF patient showed a highly positive reaction in the JUNV-rNP-based IgG-ELISA [49] . In addition, it was shown that, using sera obtained from AHF cases, the results of the JUNV rNP-based IgG ELISA correlate well with an authentic JUNV antigen-based IgG ELISA [49] . An IgM-capture ELISA using purified LASV-rNP as an antigen has been developed in the same way as in previous reports [54, 57] and detects an LASV-IgM antibody [58] . In addition, immunoblot assays based on N-terminally truncated LASV rNP have been developed for detecting IgG and IgM antibodies against LASV. These methods may provide a rapid and simple Lassa fever test for use under field conditions [47] .
An IFA using virus-infected cells is a common antibody test for VHF viruses [59] [60] [61] [62] [63] . To avoid the use of highly pathogenic viruses for the antigen preparation, mammalian cells expressing recombinant rNP have been developed [51, 57, [64] [65] [66] [67] [68] . Lassa virus NP antigen for IFA can be prepared simply as described [51] . Briefly, the procedure involves (1) transfecting HeLa cells with a mammalian cell expression vector inserted with the cloned NP cDNA; (2) expanding the stable NP-expressing cells by antibiotic selection; (3) mixing the rNP-expressing cells with un-transfected HeLa cells (at a ratio of 1:1); (4) spotting the cell mixtures onto glass slides, then drying and fixing them in acetone.
In the IFA specific for LASV-NP, antibody positive sera show characteristic granular staining patterns in the cytoplasm (Figure 2 ) [69] , thus making it easy to distinguish positive from negative samples. The specificity of the assay has also been confirmed by using sera obtained from Lassa fever patients [51] . In addition, an IFA using JUNV rNP-expressing HeLa cells has been developed to detect antibodies against JUNV, and the assay has been evaluated by using AHF patients' sera [70] . The LASV-rNP-based antibody detection systems such as ELISA and IFA are suggested to be useful not only for the diagnosis of Lassa fever, but also for seroepidemiological studies of LASV infection. In our preliminary study, approximately 15% of the sera collected from 334 Ghanaians and less than 3% of 280 Zambians showed positive reactions in the LASV-rNP-based IgG ELISA [58] . These results are in agreement with the fact that Lassa fever is endemic to the West African region, including Ghana, but less in the East African region.
For the diagnosis of many viral infections, PCR assays have been shown to have an excellent analytical sensitivity, but the established techniques are limited by their requirement for expensive equipment and technical expertise. Moreover, the high degree of genetic variability of the RNA viruses, including arenavirus and bunyavirus, poses difficulties in selecting primers for RT-PCR assays that can detect all strains of the virus. Since the sensitivity of the Ag-capture ELISA is comparable to that of RT-PCR for several virus-mediated infectious diseases, including Lassa fever and filovirus hemorrhagic fever [51, [71] [72] [73] , the Ag-capture ELISA is a sophisticated approach that can be used for the diagnosis of viral infections. Ag-capture ELISAs detecting viral NP in viremic sera have been widely applied to detect various viruses, since they are the most abundant viral antigens and have highly conserved amino acid sequences [50, 51, 54, 71, 72, 74, 75] . Polyclonal anti-sera or a mixture of MAbs present in the ascetic fluids from animals immunized for HFVs have been used for capture-antibodies in the Ag-capture ELISA [36, [76] [77] [78] [79] . MAbs recognizing conserved epitopes of the rNP are also used as capture antibodies since they have a high specificity for the antigens, and an identification of the epitopes of these MAbs is of crucial importance for the assessment of the specificity and cross-reactivity of the assay system [50, 51, 53, 54, 71, 75] . In order to develop a sensitive diagnostic test for Lassa fever and AHF, rNPs of LASV and JUNV (see above) have been prepared, and newly established MAbs against them have been characterized and used for Ag-capture ELISAs [50, 51] . The Ag-capture ELISA using MAb 4A5 has been confirmed to be useful in the detection of authentic LASV antigen in sera serially collected from hamsters infected with LASV [51] . The sensitivity of the MAb 4A5-based Ag-capture ELISA was similar to that of conventional RT-PCR, suggesting that the Ag-capture ELISA can be efficiently used in the diagnosis of Lassa fever [51] . Therefore, the MAb 4A5-based Ag-capture ELISA is considered to be useful in the diagnosis of Lassa fever. Also, by using MAbs raised against the rNP of JUNV, Ag-capture ELISAs specific for JUNV and broadly reactive to human pathogenic New World arenaviruses have been developed [50] . The Ag-capture ELISA using MAb E4-2 and C11-12 detected the Ags of all of the pathogenic New World arenaviruses tested, including JUNV. On the other hand, the Ag-capture ELISA using MAb C6-9 detects only the JUNV Ag. Considering that the symptoms of JUNV infection in humans are indistinguishable from those due to other pathogenic New World arenaviruses, the Ag capture ELISA using MAb C6-9 may be a useful diagnostic tool, especially for AHF [50] .
The virus neutralization assay is accepted as the "gold standard" serodiagnostic assay to quantify the antibody response to infection and vaccination of a wide variety of viruses associated with human diseases [80] [81] [82] [83] [84] [85] [86] . The presence of neutralizing antibodies is a reliable indicator of protective immunity against VHF [87] [88] [89] . The most direct method for detection of neutralizing antibodies against HFVs is by plaque reduction neutralization tests using infectious viruses. However, because of the high pathogenicity of HFVs to humans and the strict regulation of select agents, only a limited number of laboratories are able to perform such neutralization tests. For many HFVs, replication-incompetent pseudotyped virus particles bearing viral envelope protein (GP) have been shown to mimic the respective HFV infections, thus, neutralization assays using the pseudotypes may be advantageous in some laboratory settings for the detection of antibodies to HFVs without the need for heightened biocontainment requirements.
The VSV-based vector has already been used to generate replication-competent recombinant VSVs to study of the role of GPs of various viruses [90] [91] [92] . Recent advances in producing pseudotype virus particles have enabled the investigation of the virus cell entry, viral tropism, and effect of entry inhibitors, as well as measurement of the neutralization titers, by using human immunodeficiency virus-, feline immunodeficiency virus-, murine leukemia virus-, or VSV-based vectors [86, [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] . Pseudotypes based on VSV have advantages compared with other pseudotypes based on retroviruses for the following reasons. First, the pseudotype virus titer obtained with the VSV system is generally higher than that of the pseudotyped retrovirus system [104] . Second, the infection of target cells with a VSV pseudotype can be readily detected as green fluorescent protein (GFP)-positive cells at 7-16 h post-infection because of the high level of GFP expression in the VSV system [104, 105] . In contrast, the time required for infection in the pseudotyped retrovirus system is 48 h [106, 107] , which is similar to the time required for infectious viruses to replicate to a level that results in plaque-forming or cytopathic effects in infected cells. A high-throughput assay for determining neutralizing antibody titers using VSV pseudotypes expressing secreted alkaline phosphatase [108, 109] or luciferase ( Figure 3 ) has also been developed. We have recently developed a VSV-based pseudotype bearing Lassa virus GP (VSV-LAS-GP) for the detection of neutralizing antibodies in the sera obtained from a Lassa fever patient. An example of the LASV neutralization assay using the VSV pseudotype is shown (Figure 4 ). In the presence of serum from Lassa fever patients, the number of GFP-positive cells (infectivity of VSV-LAS-GP) is significantly reduced compared with the number in the absence of the patient's serum ( Figure 4A ). The control VSV pseudotype bearing VSV GP (VSV-VSV-G) is not neutralized by any sera. When the cut-off serum dilution is set at 50% inhibition of infectivity compared with the infectivity in the absence of the test serum, the neutralization titer of this patient's serum for VSV-LAS-GP is calculated to be 75 ( Figure 4B ). Likewise, a VSV-based pseudotype bearing the Junin virus GP has been developed for the detection of neutralizing antibodies from AHF patients' sera. The accuracy of the results of VSV-based neutralization assays has been confirmed by comparison with the results of the neutralization assay using live Junin virus [70] . The Lujo virus is a new member of the hemorrhagic fever-associated arenavirus family from Zambia and southern Africa, and the virus is classified as a BSL-4 pathogen [17] . The genome sequence analysis of the Lujo virus suggests that the virus is genetically distinct from previously characterized arenaviruses. In order to study the infectivity of this newly identified arenavirus, we have recently developed a luciferase-expressing VSV pseudotype bearing Lujo virus GPC (VSV-Lujo-GP). As shown in Figure 3 , infection with VSV-Lujo-GPC is specifically neutralized by rabbit anti-Lujo GPC serum. Thus, the VSV-Lujo-GP may be a useful tool not only for determining the neutralizing antibody titer within the serum, but also for exploring yet-to-be-defined cellular receptor(s) for Lujo virus infection or for screening inhibitors of the Lujo virus GP-mediated cell entry.
Hemorrhagic fever outbreaks caused by pathogenic arenaviruses result in high fatality rates. A rapid and accurate diagnosis is a critical first step in any outbreak. Serologic diagnostic methods for VHFs most often employ an ELISA, IFA, and/or virus neutralization assay. Diagnostic methods using recombinant viral proteins have been developed and their utilities for diagnosing of VHF have been reviewed. IgG-and IgM-ELISAs and IFAs using rNPs as antigens are useful for the detection of antibodies induced in the patients' sera. These methods are also useful for seroepidemiological surveys for HFVs. Ag-capture ELISAs using MAbs to the arenavirus rNPs are specific for the virus species or can be broadly reactive for New World arenaviruses, depending on the MAb used. Furthermore, the VSV-based pseudotype system provides a safe and rapid tool for measuring virus neutralizing antibody titers, as well as a model to analyze the entry of the respective arenavirus in susceptible cells without using live arenaviruses. Recent discoveries of novel arenavirus species [17, 26, 110] and their potential to evolve predominantly via host switching, rather than with their hosts [110, 111] , suggest that an unknown pathogenic arenavirus may emerge in the future, and that the diagnostic methods for VHF caused by arenaviruses should thus be further developed and improved.
|
What proteins does the Arenavirus produce?
| 5,275 |
a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L)
| 9,418 |
1,606 |
Serological Assays Based on Recombinant Viral Proteins for the Diagnosis of Arenavirus Hemorrhagic Fevers
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3497043/
SHA: f1d308db379b3c293bcfc8fe251c043fe8842358
Authors: Fukushi, Shuetsu; Tani, Hideki; Yoshikawa, Tomoki; Saijo, Masayuki; Morikawa, Shigeru
Date: 2012-10-12
DOI: 10.3390/v4102097
License: cc-by
Abstract: The family Arenaviridae, genus Arenavirus, consists of two phylogenetically independent groups: Old World (OW) and New World (NW) complexes. The Lassa and Lujo viruses in the OW complex and the Guanarito, Junin, Machupo, Sabia, and Chapare viruses in the NW complex cause viral hemorrhagic fever (VHF) in humans, leading to serious public health concerns. These viruses are also considered potential bioterrorism agents. Therefore, it is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of arenavirus outbreaks. However, these arenaviruses are classified as BSL-4 pathogens, thus making it difficult to develop diagnostic techniques for these virus infections in institutes without BSL-4 facilities. To overcome these difficulties, antibody detection systems in the form of an enzyme-linked immunosorbent assay (ELISA) and an indirect immunofluorescence assay were developed using recombinant nucleoproteins (rNPs) derived from these viruses. Furthermore, several antigen-detection assays were developed. For example, novel monoclonal antibodies (mAbs) to the rNPs of Lassa and Junin viruses were generated. Sandwich antigen-capture (Ag-capture) ELISAs using these mAbs as capture antibodies were developed and confirmed to be sensitive and specific for detecting the respective arenavirus NPs. These rNP-based assays were proposed to be useful not only for an etiological diagnosis of VHFs, but also for seroepidemiological studies on VHFs. We recently developed arenavirus neutralization assays using vesicular stomatitis virus (VSV)-based pseudotypes bearing arenavirus recombinant glycoproteins. The goal of this article is to review the recent advances in developing laboratory diagnostic assays based on recombinant viral proteins for the diagnosis of VHFs and epidemiological studies on the VHFs caused by arenaviruses.
Text: The virus family Arenaviridae consists of only one genus, but most viruses within this genus can be divided into two different groups: the Old World arenaviruses and the New World arenaviruses (also known as the Tacaribe complex) [1, 2] . The differences between the two groups have been established through the use of serological assays. Most of the arenaviruses cause persistent infection in rodents without any symptoms, and humans acquire a variety of diseases when zoonotically infected. Lymphocytic choriomeningitis virus (LCMV) is the only arenavirus to exhibit a worldwide distribution, and causes illnesses such as meningitis [3, 4] . Congenital LCMV infections have also been reported [4, 5] . Most importantly, viral hemorrhagic fever (VHF) can be caused by several arenaviruses. Lassa fever, caused by the Lassa virus (LASV), an Old World arenavirus, is one of the most devastating VHFs in humans [6] . Hemorrhaging and organ failure occur in a subset of patients infected with this virus, and it is associated with high mortality. Many cases of Lassa fever occur in Western Africa in countries such as Guinea, Sierra Leone, and Nigeria [7] [8] [9] [10] [11] [12] [13] . Tacaribe complex lineage B of the New World arenaviruses consists of the Junin virus (JUNV), Guanarito virus (GUNV), Sabia virus (SABV) and Machupo virus (MACV), the etiological agents of Argentine, Venezuelan, Brazilian, and Bolivian hemorrhagic fevers, respectively [14, 15] . Although genetically distinct from one another, they appear to produce similar symptoms, accompanied by hemorrhaging in humans [14, 15] . These pathogenic New World arenavirus species are closely associated with a specific rodent species [6] .
Humans are usually infected with pathogenic arenaviruses through direct contact with tissue or blood, or after inhaling aerosolized particles from urine, feces, and saliva of infected rodents. After an incubation period of 1-3 weeks, infected individuals abruptly develop fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria, and mucosal bleeding. Neurological problems have also been described, including hearing loss, tremors, and encephalitis. Because the symptoms of pathogenic arenavirus-related illness are varied and nonspecific, the clinical diagnosis is often difficult [14, 16] . Human-to-human transmission may occur via mucosal or cutaneous contact, or through nosocomial contamination [14, 16] . These viruses are also considered to be potential bioterrorism agents [2] .
A number of arenavirus species have been recently discovered as a result of both rodent surveys and disease outbreaks [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . A novel pathogenic New World arenavirus, Chapare virus (CHPV), has been isolated from a fatal case of VHF in Bolivia [20] . In addition, five cases of VHF have been reported in South Africa, and a novel arenavirus, named Lujo virus, was isolated from a patient [17] . The Lujo virus is most distantly related to the other Old World arenaviruses [17] . To date, there is no information concerning the vertebrate host for the Chapare and Lujo viruses.
There is some evidence of endemicity of the Lassa virus in neighboring countries [27, 28] . However, as the magnitude of international trade and travel is continuously increasing, and the perturbation of the environment (due either to human activity or natural ecological changes) may result in behavioral changes of reservoir rodents, highly pathogenic arenaviruses could be introduced to virus-free countries from endemic areas. In fact, more than twenty cases of Lassa fever have been reported outside of the endemic region in areas such as the USA, Canada, Europe, and Japan [29] [30] [31] [32] [33] . It is of great importance to detect these pathogens rapidly and specifically in order to minimize the risk and scale of outbreaks of VHFs caused by arenaviruses. However, these arenaviruses are classified as biosafety level (BSL)-4 pathogens, making it difficult to develop diagnostic techniques for these virus infections in laboratories without BSL-4 facilities. To overcome these difficulties, we have established recombinant viral nucleoproteins (rNPs)-based serological assays, such as IgG-enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescence assay (IFA), and antigen (Ag)-capture ELISA for the diagnosis of VHFs caused by highly pathogenic arenaviruses. Furthermore, virus neutralization assays using pseudotype virus-bearing arenavirus GPs have been developed. In this review, we describe the usefulness of such recombinant protein-based diagnostic assays for diagnosing VHFs caused by arenaviruses.
In outbreaks of VHFs, infections are confirmed by various laboratory diagnostic methods. Virus detection is performed by virus isolation, reverse transcription-polymerase chain reaction (RT-PCR), and antigen-capture ELISA. It has been shown that monoclonal antibody panels against pathogenic arenaviruses are useful for detecting viral antigens on the virus-infected cells as well as for investigating of antigenic relationships of arenaviruses [34] [35] [36] . Detection of the virus genome is suitable for a rapid and sensitive diagnosis of VHF patients in the early stage of illness, and extensive reviews of such RT-PCR assays have been described [37, 38] . More recently, progress in the RT-PCR method covering genetic variations of the hemorrhagic fever viruses (HFVs) [39, 40] and a multiplexed oligonucleotide microarray for the differential diagnosis of VHFs have also been reported [41] . On the other hand, antibodies against these viruses can be detected by the indirect immunofluorescence assay (IFA), or IgG-and IgM-ELISA. An IFA detects the antibody in the serum, which is able to bind to the fixed monolayer of the virus-infected cells. Although the interpretation of immunofluorescence results requires experience, the assay has advantages over other methods, since each virus generates a characteristic fluorescence pattern that adds specificity to the assay compared to a simple ELISA readout. A serological diagnosis by the detection of specific IgM and IgG antibodies to the HFVs must be sensitive, specific and reliable, because a misdiagnosis can lead to panic in the general population. An IgM-specific ELISA is suitable for detecting recent infection, but the relevance of IgM testing for acute VHF depends on the virus and the duration of illness; specific IgM is not often present in the very early stage of illness, and patients who die of VHF often fail to seroconvert at all. An IgG-specific ELISA is efficacious, not only in the diagnosis of a large number of VHF cases, especially during convalescence, but also for epidemiological studies in the endemic regions. The detailed methods used for the IFA and IgG-and IgM-ELISAs for the diagnosis of VHF using authentic virus-antigens have been described in detail [42] [43] [44] [45] .
Arenaviruses have a bisegmented, negative-sense, single stranded RNA genome with a unique ambisense coding strategy that produces just four known proteins: a glycoprotein, a nucleoprotein (NP), a matrix protein (Z), and a polymerase (L) [46] . Of these proteins, the NP is the most abundant in virus-infected cells. Recombinant protein technology could meet the demand for a simple and reliable VHF test system, and recombinant NP (rNP) has been shown to be useful for serological surveys of IgM-and IgG antibodies against arenaviruses [47] [48] [49] [50] .
Recombinant baculoviruses that express the full-length rNP of arenaviruses have been generated [48, 50, 51] . The method used for the purification of arenavirus rNP from insect Tn5 cells infected with recombinant baculoviruses is effective and simple compared to those for Ebola, Marburg, and Crimean-Congo hemorrhagic fever virus rNPs [51] [52] [53] [54] [55] . Most of the arenavirus rNPs expressed in insect cells using the recombinant baculoviruses are crystallized [56] and are solubilized in PBS containing 8M urea. Since the majority of Tn5 cellular proteins are solubilized in PBS containing 2M urea, the arenavirus rNPs in the insoluble fraction in PBS containing 2M urea can be solubilized by sonication in PBS containing 8M urea. After a simple centrifugation of the lysates in PBS containing 8M urea, the supernatant fractions can be used as purified rNP antigens without further purification steps [51] . The control antigen is produced from Tn5 cells infected with baculovirus lacking the polyhedrin gene (ΔP) in the same manner as the arenavirus rNPs ( Figure 1 ).
Purified rNPs. The expression and purification efficiency of arenavirus rNP were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after staining the gels with Coomassie blue. Purified NP antigens with approximate molecular weights of 62 kDa from Luna, LCM, Lassa, Lujo, Junin, Machupo, Guanarito, Sabia, and Chapare viruses and the purified negative control antigen (ΔP) are shown.
As described above, recombinant baculoviruses allow the delivery of rNP antigens without using infectious live arenaviruses. An ELISA plate coated with the predetermined optimal quantity of purified rNPs (approximately 100 ng/well) is used for the IgG-antibody detection assay. An advantage of using recombinant rNP for the IgG-ELISA is that it enables a direct comparison of antibody cross-reactivity among arenavirus rNPs, since antigen preparations of all arenavirus rNPs tested are performed using the same method [51] . Rabbit anti-sera raised against LCMV-rNP and LASV-rNP show cross-reactivity to LASV-rNP and LCMV-rNP, respectively, indicating that rabbit antibodies against rNPs of Old World arenaviruses cross-react with rNPs of other Old World arenaviruses (Table 1 ) [51] . Similarly, rabbit anti-sera generated against JUNV-NP show cross-reactivity to the LASV-rNP and LCMV-rNP, although the reaction is weak. However, rabbit anti-sera against LASV-NP and LCMV-NP show a negative reaction to the JUNV-rNP (Table 1 ) [51] , indicating that rabbit antibodies against JUNV (a pathogenic New World arenavirus) NP might cross-react with the Old World arenavirus NP, whereas antibodies against Old World arenavirus NPs may not be able to react with pathogenic New World arenavirus NPs.
The rNP-based IgG-ELISA has also been used for the characterization of a mouse monoclonal antibody (MAb). Nakauchi et al. [50] have investigated the cross-reactivity of MAbs against JUNV rNP to pathogenic New World arenavirus rNPs, as well as LASV rNP. MAb C11-12 reacts at the same level with the rNPs of all of the pathogenic New World arenaviruses, including JUNV, GTOV, MACV, SABV, and CHPV, indicating that this MAb recognizes an epitope conserved among pathogenic New World arenaviruses. Another MAb, C6-9, reacts specifically with the rNP of JUNV, but does not react with those of the other pathogenic New World arenaviruses [50] . This indicates that MAb C6-9 recognizes a JUNV-specific epitope. None of these MAbs reacts with the rNP of the human pathogenic Old World arenavirus LASV. Thus, the MAb C11-12 is considered to be a broadly reactive MAb against New World arenaviruses, whereas MAb C6-9 is JUNV-specific. These findings have been confirmed by detailed epitope analyses using peptide mapping [50] . Similarly, the cross-reactivity of MAbs against LASV rNP has been analyzed [51] . MAb 4A5 cross-reacts with the Mopeia virus (MOPV) but not with the LCMV rNP. MAb 6C11 cross-reacts with LCMV rNP, while MAb 2-11 does not cross-react with LCMV rNP [51] . Table 1 . Anti-serum reactivity for rNPs of different arenaviruses in IgG ELISAs.
Reactivity for rNP from LASV LCMV JUNV anti-LASV NP
It is important to evaluate whether rNP-based ELISA is useful for the diagnosis of human VHF cases. The specificity of the LASV-rNP-based IgG ELISA has been confirmed by using sera obtained from Lassa fever patients [51] . The Lassa fever patients' sera show a highly positive reaction in the LASV-rNP-based IgG-ELISA, but sera from patients with Argentine hemorrhagic fever (AHF), which is caused by JUNV, do not. The serum from an AHF patient showed a highly positive reaction in the JUNV-rNP-based IgG-ELISA [49] . In addition, it was shown that, using sera obtained from AHF cases, the results of the JUNV rNP-based IgG ELISA correlate well with an authentic JUNV antigen-based IgG ELISA [49] . An IgM-capture ELISA using purified LASV-rNP as an antigen has been developed in the same way as in previous reports [54, 57] and detects an LASV-IgM antibody [58] . In addition, immunoblot assays based on N-terminally truncated LASV rNP have been developed for detecting IgG and IgM antibodies against LASV. These methods may provide a rapid and simple Lassa fever test for use under field conditions [47] .
An IFA using virus-infected cells is a common antibody test for VHF viruses [59] [60] [61] [62] [63] . To avoid the use of highly pathogenic viruses for the antigen preparation, mammalian cells expressing recombinant rNP have been developed [51, 57, [64] [65] [66] [67] [68] . Lassa virus NP antigen for IFA can be prepared simply as described [51] . Briefly, the procedure involves (1) transfecting HeLa cells with a mammalian cell expression vector inserted with the cloned NP cDNA; (2) expanding the stable NP-expressing cells by antibiotic selection; (3) mixing the rNP-expressing cells with un-transfected HeLa cells (at a ratio of 1:1); (4) spotting the cell mixtures onto glass slides, then drying and fixing them in acetone.
In the IFA specific for LASV-NP, antibody positive sera show characteristic granular staining patterns in the cytoplasm (Figure 2 ) [69] , thus making it easy to distinguish positive from negative samples. The specificity of the assay has also been confirmed by using sera obtained from Lassa fever patients [51] . In addition, an IFA using JUNV rNP-expressing HeLa cells has been developed to detect antibodies against JUNV, and the assay has been evaluated by using AHF patients' sera [70] . The LASV-rNP-based antibody detection systems such as ELISA and IFA are suggested to be useful not only for the diagnosis of Lassa fever, but also for seroepidemiological studies of LASV infection. In our preliminary study, approximately 15% of the sera collected from 334 Ghanaians and less than 3% of 280 Zambians showed positive reactions in the LASV-rNP-based IgG ELISA [58] . These results are in agreement with the fact that Lassa fever is endemic to the West African region, including Ghana, but less in the East African region.
For the diagnosis of many viral infections, PCR assays have been shown to have an excellent analytical sensitivity, but the established techniques are limited by their requirement for expensive equipment and technical expertise. Moreover, the high degree of genetic variability of the RNA viruses, including arenavirus and bunyavirus, poses difficulties in selecting primers for RT-PCR assays that can detect all strains of the virus. Since the sensitivity of the Ag-capture ELISA is comparable to that of RT-PCR for several virus-mediated infectious diseases, including Lassa fever and filovirus hemorrhagic fever [51, [71] [72] [73] , the Ag-capture ELISA is a sophisticated approach that can be used for the diagnosis of viral infections. Ag-capture ELISAs detecting viral NP in viremic sera have been widely applied to detect various viruses, since they are the most abundant viral antigens and have highly conserved amino acid sequences [50, 51, 54, 71, 72, 74, 75] . Polyclonal anti-sera or a mixture of MAbs present in the ascetic fluids from animals immunized for HFVs have been used for capture-antibodies in the Ag-capture ELISA [36, [76] [77] [78] [79] . MAbs recognizing conserved epitopes of the rNP are also used as capture antibodies since they have a high specificity for the antigens, and an identification of the epitopes of these MAbs is of crucial importance for the assessment of the specificity and cross-reactivity of the assay system [50, 51, 53, 54, 71, 75] . In order to develop a sensitive diagnostic test for Lassa fever and AHF, rNPs of LASV and JUNV (see above) have been prepared, and newly established MAbs against them have been characterized and used for Ag-capture ELISAs [50, 51] . The Ag-capture ELISA using MAb 4A5 has been confirmed to be useful in the detection of authentic LASV antigen in sera serially collected from hamsters infected with LASV [51] . The sensitivity of the MAb 4A5-based Ag-capture ELISA was similar to that of conventional RT-PCR, suggesting that the Ag-capture ELISA can be efficiently used in the diagnosis of Lassa fever [51] . Therefore, the MAb 4A5-based Ag-capture ELISA is considered to be useful in the diagnosis of Lassa fever. Also, by using MAbs raised against the rNP of JUNV, Ag-capture ELISAs specific for JUNV and broadly reactive to human pathogenic New World arenaviruses have been developed [50] . The Ag-capture ELISA using MAb E4-2 and C11-12 detected the Ags of all of the pathogenic New World arenaviruses tested, including JUNV. On the other hand, the Ag-capture ELISA using MAb C6-9 detects only the JUNV Ag. Considering that the symptoms of JUNV infection in humans are indistinguishable from those due to other pathogenic New World arenaviruses, the Ag capture ELISA using MAb C6-9 may be a useful diagnostic tool, especially for AHF [50] .
The virus neutralization assay is accepted as the "gold standard" serodiagnostic assay to quantify the antibody response to infection and vaccination of a wide variety of viruses associated with human diseases [80] [81] [82] [83] [84] [85] [86] . The presence of neutralizing antibodies is a reliable indicator of protective immunity against VHF [87] [88] [89] . The most direct method for detection of neutralizing antibodies against HFVs is by plaque reduction neutralization tests using infectious viruses. However, because of the high pathogenicity of HFVs to humans and the strict regulation of select agents, only a limited number of laboratories are able to perform such neutralization tests. For many HFVs, replication-incompetent pseudotyped virus particles bearing viral envelope protein (GP) have been shown to mimic the respective HFV infections, thus, neutralization assays using the pseudotypes may be advantageous in some laboratory settings for the detection of antibodies to HFVs without the need for heightened biocontainment requirements.
The VSV-based vector has already been used to generate replication-competent recombinant VSVs to study of the role of GPs of various viruses [90] [91] [92] . Recent advances in producing pseudotype virus particles have enabled the investigation of the virus cell entry, viral tropism, and effect of entry inhibitors, as well as measurement of the neutralization titers, by using human immunodeficiency virus-, feline immunodeficiency virus-, murine leukemia virus-, or VSV-based vectors [86, [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] . Pseudotypes based on VSV have advantages compared with other pseudotypes based on retroviruses for the following reasons. First, the pseudotype virus titer obtained with the VSV system is generally higher than that of the pseudotyped retrovirus system [104] . Second, the infection of target cells with a VSV pseudotype can be readily detected as green fluorescent protein (GFP)-positive cells at 7-16 h post-infection because of the high level of GFP expression in the VSV system [104, 105] . In contrast, the time required for infection in the pseudotyped retrovirus system is 48 h [106, 107] , which is similar to the time required for infectious viruses to replicate to a level that results in plaque-forming or cytopathic effects in infected cells. A high-throughput assay for determining neutralizing antibody titers using VSV pseudotypes expressing secreted alkaline phosphatase [108, 109] or luciferase ( Figure 3 ) has also been developed. We have recently developed a VSV-based pseudotype bearing Lassa virus GP (VSV-LAS-GP) for the detection of neutralizing antibodies in the sera obtained from a Lassa fever patient. An example of the LASV neutralization assay using the VSV pseudotype is shown (Figure 4 ). In the presence of serum from Lassa fever patients, the number of GFP-positive cells (infectivity of VSV-LAS-GP) is significantly reduced compared with the number in the absence of the patient's serum ( Figure 4A ). The control VSV pseudotype bearing VSV GP (VSV-VSV-G) is not neutralized by any sera. When the cut-off serum dilution is set at 50% inhibition of infectivity compared with the infectivity in the absence of the test serum, the neutralization titer of this patient's serum for VSV-LAS-GP is calculated to be 75 ( Figure 4B ). Likewise, a VSV-based pseudotype bearing the Junin virus GP has been developed for the detection of neutralizing antibodies from AHF patients' sera. The accuracy of the results of VSV-based neutralization assays has been confirmed by comparison with the results of the neutralization assay using live Junin virus [70] . The Lujo virus is a new member of the hemorrhagic fever-associated arenavirus family from Zambia and southern Africa, and the virus is classified as a BSL-4 pathogen [17] . The genome sequence analysis of the Lujo virus suggests that the virus is genetically distinct from previously characterized arenaviruses. In order to study the infectivity of this newly identified arenavirus, we have recently developed a luciferase-expressing VSV pseudotype bearing Lujo virus GPC (VSV-Lujo-GP). As shown in Figure 3 , infection with VSV-Lujo-GPC is specifically neutralized by rabbit anti-Lujo GPC serum. Thus, the VSV-Lujo-GP may be a useful tool not only for determining the neutralizing antibody titer within the serum, but also for exploring yet-to-be-defined cellular receptor(s) for Lujo virus infection or for screening inhibitors of the Lujo virus GP-mediated cell entry.
Hemorrhagic fever outbreaks caused by pathogenic arenaviruses result in high fatality rates. A rapid and accurate diagnosis is a critical first step in any outbreak. Serologic diagnostic methods for VHFs most often employ an ELISA, IFA, and/or virus neutralization assay. Diagnostic methods using recombinant viral proteins have been developed and their utilities for diagnosing of VHF have been reviewed. IgG-and IgM-ELISAs and IFAs using rNPs as antigens are useful for the detection of antibodies induced in the patients' sera. These methods are also useful for seroepidemiological surveys for HFVs. Ag-capture ELISAs using MAbs to the arenavirus rNPs are specific for the virus species or can be broadly reactive for New World arenaviruses, depending on the MAb used. Furthermore, the VSV-based pseudotype system provides a safe and rapid tool for measuring virus neutralizing antibody titers, as well as a model to analyze the entry of the respective arenavirus in susceptible cells without using live arenaviruses. Recent discoveries of novel arenavirus species [17, 26, 110] and their potential to evolve predominantly via host switching, rather than with their hosts [110, 111] , suggest that an unknown pathogenic arenavirus may emerge in the future, and that the diagnostic methods for VHF caused by arenaviruses should thus be further developed and improved.
|
What diagnostic test has been show to have excellent sensitivity in detecting viral infections?
| 5,276 |
PCR assays
| 16,937 |
1,596 |
Glycyrrhizin Exerts Antioxidative Effects in H5N1 Influenza A Virus-Infected Cells and Inhibits Virus Replication and Pro-Inflammatory Gene Expression
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3096629/
SHA: f3b7f4469ac01f1ce916d24172570c43c537627e
Authors: Michaelis, Martin; Geiler, Janina; Naczk, Patrizia; Sithisarn, Patchima; Leutz, Anke; Doerr, Hans Wilhelm; Cinatl, Jindrich
Date: 2011-05-17
DOI: 10.1371/journal.pone.0019705
License: cc-by
Abstract: Glycyrrhizin is known to exert antiviral and anti-inflammatory effects. Here, the effects of an approved parenteral glycyrrhizin preparation (Stronger Neo-Minophafen C) were investigated on highly pathogenic influenza A H5N1 virus replication, H5N1-induced apoptosis, and H5N1-induced pro-inflammatory responses in lung epithelial (A549) cells. Therapeutic glycyrrhizin concentrations substantially inhibited H5N1-induced expression of the pro-inflammatory molecules CXCL10, interleukin 6, CCL2, and CCL5 (effective glycyrrhizin concentrations 25 to 50 µg/ml) but interfered with H5N1 replication and H5N1-induced apoptosis to a lesser extent (effective glycyrrhizin concentrations 100 µg/ml or higher). Glycyrrhizin also diminished monocyte migration towards supernatants of H5N1-infected A549 cells. The mechanism by which glycyrrhizin interferes with H5N1 replication and H5N1-induced pro-inflammatory gene expression includes inhibition of H5N1-induced formation of reactive oxygen species and (in turn) reduced activation of NFκB, JNK, and p38, redox-sensitive signalling events known to be relevant for influenza A virus replication. Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1 disease.
Text: Highly pathogenic H5N1 influenza A viruses are considered to be potential influenza pandemic progenitors [1] [2] [3] [4] [5] [6] . At least for the first wave of an H5N1 pandemic, no sufficient amounts of adequate vaccines will be available [1] [2] [3] [4] [6] [7] [8] . Therefore, antiviral therapy for influenza A viruses including highly pathogenic H5N1 virus strains remains of great importance for the first line defense against the virus [1] [2] [3] [4] 6, 9] .
The neuraminidase inhibitors oseltamivir and zanamivir as well as the adamantanes amantadin and rimantadin that interfere with the influenza M2 protein are licensed for the treament of influenza [1] [2] [3] [4] 6] . However, the use of both drug classes is limited by the emergence of resistant virus strains. In seasonal influenza strains, the majority of H3N2 viruses and a great proportion of H1N1 viruses in humans are now considered to be amantadine-and rimantadine-resistant [10] [11] [12] [13] . Moreover, a drastic increase in oseltamivir-resistant H1N1 viruses has been reported during the 2007/2008 influenza season in the northern hemisphere [14] [15] [16] [17] . Preliminary data from the United States predict a further rise for the 2008/2009 season, possibly resulting in more than 90% of the circulating H1N1 strains to be oseltamivir resistant [14] .
H5N1 virus strains appear to be generally less sensitive to antiviral treatment than seasonal influenza A virus strains and treatment-resistant H5N1 strains emerge [1] [2] [3] [4] 6, [18] [19] [20] [21] . More-over, parenteral agents for the treatment of seriously ill patients are missing. Glycyrrhizin, a triterpene saponine, is a constituent of licorice root. It has been found to interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses including respiratory viruses such as respiratory syncytial virus, SARS coronavirus, HIV, and influenza viruses [22] [23] [24] [25] [26] [27] [28] . Moreover, antiinflammatory and immunomodulatory properties were attributed to glycyrrhizin [26] . The severity of human H5N1 disease has been associated with hypercytokinaemia (''cytokine storm'') [29, 30] . Delayed antiviral plus immunomodulator treatment reduced H5N1-induced mortality in mice [31] . Therefore, antiinflammatory and immunomodulatory effects exerted by glycyrrhizin may be beneficial for treatment of H5N1. Also, glycyrrhizin is a known antioxidant [26] and antioxidants were already shown to interfere with influenza A virus replication and virus-induced pro-inflammatory responses [32] [33] [34] .
Stronger Neo-Minophagen C (SNMC) is a glycyrrhizin preparation (available as tablets or parenteral formulation) that is approved in Japan for the treatment of chronic hepatic diseases and is marketed in Japan, China, Korea, Taiwan, Indonesia, India, and Mongolia. Here, we investigated the influence of SNMC on H5N1 replication, on H5N1-induced cytokine expression, on H5N1-induced cellular oxidative stress, and on critical H5N1-induced cellular signalling events in human pneumocytes (A549 cell line).
Glycyrrhizin (Stronger Neo Minophagen C) was obtained from Minophagen Pharmaceuticals Co., Ltd. (Tokyo, Japan).
The influenza strain A/Vietnam/1203/04 (H5N1) was received from the WHO Influenza Centre (National Institute for Medical Research, London, UK). The H5N1 influenza strain A/Thailand/ 1(Kan-1)/04 was obtained from Prof. Pilaipan Puthavathana (Mahidol University, Bangkok, Thailand).
Virus stocks were prepared by infecting Vero cells (African green monkey kidney; ATCC, Manassas, VA) and aliquots were stored at 280uC. Virus titres were determined as 50% tissue culture infectious dose (TCID 50 /ml) in confluent Vero cells in 96-well microtiter plates.
A549 cells (human lung carcinoma; ATCC: CCL-185, obtained from LGC Standards GmbH, Wesel, Germany) were grown at 37uC in minimal essential medium (MEM) supplemented with 10% FBS, 100 IU/ml of penicillin and 100 mg/ml streptomycin.
Human monocytes were isolated from buffy coats of healthy donors, obtained from Institute of Transfusion Medicine and Immune Haematology, German Red Cross Blood Donor Center, Johann Wolfgang Goethe-University, Frankfurt am Main. After centrifugation on Ficoll (Biocoll)-Hypaque density gradient (Biochrom AG, Berlin, Germany), mononuclear cells were collected from the interface and washed with PBS. Then, monocytes were isolated using magnetically labeled CD14 MicroBeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) following the manufacturer's instructions. Monocytes were cultivated in IMDM supplemented with 10% pooled human serum, 100 IU/ml of penicillin, and 100 mg/ml streptomycin.
The cellular viability was assessed on confluent cell layers with CellTiter-GloH Luminescent Cell Viability Assay (Promega GmbH, Mannheim, Germany) according to the manufacturers' protocol. Cell viability was expressed as percentage of non-treated control.
To determine intracellular NP localisation, H5N1-infected A549 were fixed 8 hours p.i. for 15 min with ice-cold acetone/ methanol (40:60, Mallinckrodt Baker B.V., Deventer, The Netherlands) and stained with a mouse monoclonal antibody (1 h incubation, 1:1000 in PBS) directed against the influenza A virus nucleoprotein (NP) (Millipore, Molsheim, France). An Alexa Fluor 488 goat anti-mouse IgG (H&L) (Invitrogen, Eugene, Oregon, USA) was used (1 h incubation, 1:1000 in PBS) as secondary antibody. Nuclei were stained using 49,6-diamidino-2phenylindole (DAPI) (Sigma-Aldrich Chemie GmbH, Munich, Germany). Fluorescence was visualised using Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
For flow cytometric analysis, the same antibodies were used.
The cytopathogenic effect (CPE) reduction assay was performed as described before [34] . Confluent A549 cell monolayers grown in 96-well microtitre plates were infected with influenza A strains at the indicated multiplicities of infection (MOIs). After a one hour adsorption period, cells were washed to remove non-detached virus. The virus-induced CPE was recorded at 24 h post infection (p.i.).
Unless otherwise stated, A549 cells were continuously treated with glycyrrhizin starting with a 1 h pre-incubation period. For time-ofaddition experiments, glycyrrhizin was added exclusively during the 1 h pre-incubation period, exclusively during the 1 h adsorption period, or after exclusively after the wash-out of input virus.
Total RNA was isolated from cell cultures using TRI reagent (Sigma-Aldrich, Munich, Germany). Real time PCR for H5 was performed using described methods [35] . The following primers were used: sense 59 acg tat gac tac ccg cag tat tca g 39; antisense 59 aga cca gcy acc atg att gc 39; probe 6-FAM-tca aca gtg gcg agt tcc cta gca-TAMRA.
The fraction of cells with fractional DNA content (''sub-G1'' cell subpopulation) indicates cytotoxicity. Sub-G1 cells are considered to be dead (usually apoptotic) cells. Cells were fixed with 70% ethanol for two hours at 220uC. The cellular DNA was stained using propidium iodide (20 mg/ml) and analysed by flow cytometry (FacsCalibur, BD Biosciences, Heidelberg, Germany).
Caspase activation was measured using the Caspase-Glo 8, 9, or 3/7 Assays (Promega, Mannheim, Germany) following the manufacturer's instructions.
Cell culture supernatants were collected and frozen at 280uC. Cytokines/chemokines were quantified by specific ELISA Duo Sets (R&D Systems GmbH, Wiesbaden, Germany) following the manufacturer's instructions.
NFkB activity was investigated in H5N1 (MOI 0.01)-infected cells by quantification of the NFkB subunits Rel A (p65) and NFkB1 (p50) from nuclear extracts using the TransAM TM transcription factor DNA-binding ELISAs (Active Motif, Rixensart, Belgium). Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction.
Cell culture supernatants were investigated for chemotactic activity by measurement of the activity to induce monocyte migration through membrane inserts in 24-well plates (pore size 8 mm; BD Biosciences, Heidelberg, Germany). Monocytes (1610 6 in 100 ml of IMDM with 10% pooled human serum) were added into the cell culture inserts (upper chamber) and cell culture supernatants (300 ml), were added to the lower chamber of the well. After a 48 h incubation period, cells were fixed with 4% paraformaldehyde and permeabilised with PBS containing 0.3% Tritron X-100. Then, nuclei were stained with 49,6-diamidino-2phenylindole (DAPI). The upper side of the membrane was wiped with a wet swab to remove the cells, while the lower side of the membrane was rinsed with PBS. The number of cells at the lower side of each membrane was quantified by counting of cells from three randomly chosen sections (3.7 mm 2 ) using an Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
Cells were lysed in Triton X-sample buffer and separated by SDS-PAGE. Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction. Proteins were detected using specific antibodies against bactin (Sigma-Aldrich Chemie GmbH, Munich, Germany), JNK, phosphorylated JNK, p38, or phosphorylated p38, (all purchased from New England Biolabs GmbH, Frankfurt am Main, Germany) and were visualised by enhanced chemiluminescence using a commercially available kit (Amersham, Freiburg, Germany).
Reactive oxygen species (ROS) were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes, distributed by Invitrogen, Karlsruhe, Germany).
Two groups were compared by t-test. More groups were compared by ANOVA with subsequent Student-Newman-Keuls test.
The A549 cell line, derived from a human pulmonary adenocarcinoma, is an established model for type II pneumocytes [36] , and commonly used for the investigation of the effect of influenza viruses on this cell type [see e.g. 6,37,38]. If not otherwise stated, glycyrrhizin was continuously present in cell culture media starting with a 1 h preinfection period. Glycyrrhizin 200 mg/ml (the maximum tested concentration) did not affect A549 cell viability (data not shown) but clearly decreased CPE formation in A549 cells infected with the H5N1 influenza strain A/Thailand/1(Kan-1)/04 at MOIs of 0.01, 0.1 or 1 ( Figure 1A ). Similar results were obtained in A549 cells infected with strain A/Vietnam/1203/04 (H5N1) (Suppl. Figure 1A) . Staining of A549 cells for influenza A nucleoprotein 24 h after infection with strain H5N1 A/Thailand/1(Kan-1)/04 indicated that glycyrrhizin 200 mg/ml significantly reduces the number of influenza A nucleoprotein positive cells ( Figure 1B) .
To examine the influence of glycyrrhizin on virus progeny, A549 cells were infected with the H5N1 influenza strain A/ Thailand/1(Kan-1)/04 at MOI 0.01 or MOI 1 and infectious virus titres were determined 24 h post infection ( Figure 1C ). While glycyrrhizin in concentrations up to 50 mg/ml did not affect H5N1 replication, moderate effects were exerted by glycyrrhizin 100 mg/ ml and more pronounced effects by glycyrrhizin 200 mg/ml (MOI 0.01: 13-fold reduction, MOI 1: 10-fold reduction). Next, influence of glycyrrhizin on H5N1 replication was confirmed by the detection of viral (H5) RNA using quantitative PCR. Only glycyrrhizin concentrations $100 mg/ml significantly reduced Figure 1B) or H5N1 A/Vietnam/1203/04-infected (Suppl. Figure 1C ) A549 cells (MOI 0.01) 24 h post infection.
Time-of-addition experiments revealed that maximal effects were achieved when glycyrrhizin was continuously present starting with a 1 h pre-incubation period ( Figure 1D ). Addition of glycyrrhizin post infection showed reduced antiviral effects while pre-incubation alone or glycyrrhizin addition during the adsorption period did not significantly affect H5N1 replication.
For investigation of H5N1-induced cytokine expression, five pro-inflammatory genes were chosen that had been correlated to severity of influenza disease: CXCL10 (also known as interferon-cinducible protein 10, IP-10), interleukin 6 (IL6), interleukin 8, (IL8; also known as CXCL8), CCL2 (also known as monocyte chemoattractant protein 1, MCP-1), and CCL5 (also known as RANTES). A549 cells were infected with H5N1 A/Thailand/ 1(Kan-1)/04 or H5N1 A/Vietnam/1203/04 at MOI 0.01, 0.1, or 1. Glycyrrhizin treatment was performed with 25, 50, 100, or 200 mg/ml. Cytokine expression was detected 24 h post infection by ELISA. Glycyrrhizin did not affect cytokine expression of noninfected cells (data not shown) but inhibited expression of all cytokines investigated in H5N1-infected cells in a dose-dependent manner (Figure 2, Figure 3A ). Effects were more pronounced at lower MOIs. Notably, expression of all cytokines except IL8 was significantly inhibited after treatment with glycyrrhizin 50 mg/ml Figure 3A ) although these glycyrrhizin concentrations had no effect on H5N1 replication in A549 cells (Figure 1, Figure S1 ).
Cytokine expression by influenza A virus-infected respiratory cells causes recruitment of peripheral blood monocytes into the lungs of patients where they differentiate to macrophages which are thought to contribute to influenza A virus pathogenicity [5, 39] . In a chemotaxis assay, the influence of glycyrrhizin was investigated on migration of monocytes towards supernatants of H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.1)-infected A549 cells through 8 mm filters. Monocyte migration towards supernatants of H5N1-infected cells was strongly increased relative to migration towards supernatants of non-infected cells. Treatment of H5N1- infected cells with glycyrrhizin 100 mg/ml clearly suppressed chemoattraction activity of supernatants ( Figure 3B ).
Influenza viruses including H5N1 have been shown to induce caspase-dependent apoptosis in airway cells and this apoptosis has been correlated to the virus pathogenicity [40, 41] . Glycyrrhizin concentrations up to 200 mg/ml did not affect caspase activation in non-infected cells ( Figure 4A-C) . Glycyrrhizin concentrations $100 mg/ml inhibited H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.01)-induced activation of the initiator caspases 8 and 9 as well as of the effector caspases 3/7 in A549 cells as determined 24 h post infection ( Figure 4A-C) . Lower glycyrrhizin concentrations did not affect H5N1-induced apoptosis. The detection of cells in sub-G1 phase resulted in similar findings ( Figure 4D ).
Substances that inhibit H5N1-induced caspase 3 activation including caspase 3 inhibitors cause nuclear retention of RNP complexes [34, 42] . In accordance, glycyrrhizin also interfered with nuclear export RNP at MOI 1 ( Figure S2 ). Similar results were obtained in MOI 0.01 H5N1 A/Thailand/1(Kan-1)/04infected cells ( Figure S3 ).
Influence of glycyrrhizin on H5N1-induced activation of nuclear factor kB (NFkB), p38, and on H5N1-induced cellular reactive oxygen species (ROS) formation Activation of NFkB, p38, and JNK have been associated with influenza A virus replication and virus-induced pro-inflammatory gene expression [34, [43] [44] [45] [46] [47] . While glycyrrhizin did not influence NFkB activity in non-infected A549 cells in the tested concentra-tions (data not shown), glycyrrhizin inhibited NFkB activation in H5N1-infected cells ( Figure 5A ). Moreover, glycyrrhizin inhibited H5N1-induced phosphorylation of the MAPKs p38 and JNK ( Figure 5B ).
In addition to their roles during influenza A virus replication and virus-induced cytokine/chemokine expression, NFkB, p38, and JNK are constituents of redox-sensitive signalling pathways [48] [49] [50] [51] . Antioxidants had been already found to interfere with influenza A virus-induced signalling through NFkB, p38, and JNK, with influenza A virus replication, and with influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] . Since glycyrrhizin is known to exert antioxidative effects [26] we speculated that glycyrrhizin may interfere with H5N1-induced ROS formation. Indeed glycyrrhizin exerted clear antioxidative effects in H5N1 (MOI 0.01)-infected cells ( Figure 5C ) causing significant reduction of ROS formation already at a concentration of 25 mg/ml ( Figure 5D ).
Here, we show that glycyrrhizin inhibits the replication of highly pathogenic H5N1 influenza A virus, H5N1-induced apoptosis, and H5N1-induced expression of pro-inflammatory cytokines in lung-derived A549 cells. After intravenous administration, achievable plasma concentrations of glycyrrhizin have been described to be about 100 mg/ml [52] . Therefore, the glycyrrhizin concentrations found to interfere with H5N1 replication and H5N1-induced pro-inflammatory gene expression in the present report are in the range of therapeutic plasma levels. Notably, although higher glycyrrhizin concentrations were needed to interfere with SARS coronavirus replication [22] than with H5N1 replication, beneficial results were reported in glycyrrhizin (SNMC)-treated SARS patients in comparison to SARS patients who did not receive glycyrrhizin [23] . Notably, investigation of different glycyrrhizin derivatives against SARS coronavirus led to the identification of compounds with enhanced antiviral activity [53] . Therefore, glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs.
Experimental results suggested that glycyrrhizin might be able to affect seasonal influenza A virus disease by antiviral and immunomodulatory effects [26, 27] . Mice were prevented from lethal H2N2 infection by glycyrrhizin although no influence on virus replication was detected. The mechanism was suggested to be induction of interferon-c in T-cells by glycyrrhizin [54] . Moreover, glycyrrhizin was shown to influence seasonal influenza A virus replication through interaction with the cell membrane [25, 28] . However, these effects were observed only in concentrations $200 mg/ml when glycyrrhizin was added during the virus adsorption period. Since glycyrrhizin addition during the adsorption period did not influence H5N1 replication in our experiments it appears not likely that membrane effects contribute to anti-H5N1 effects detected here in lower concentrations.
Our results rather suggest that glycyrrhizin interferes with H5N1-induced oxidative stress. Influenza A virus (including H5N1) infection induces ROS formation. Antioxidants were found to inhibit influenza A virus replication and influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] and glycyrrhizin is known to exert antioxidative effects [26] . Here, glycyrrhizin interfered with H5N1-induced activation of NFkB, p38, and JNK representing redox-sensitive signalling events [48] [49] [50] [51] involved in influenza A virus replication and influenza A virusinduced cellular cytokine/chemokine production [34, [43] [44] [45] [46] 55] . Glycyrrhizin 50 mg/ml significantly reduced H5N1-induced activation of NFkB. In addition, glycyrrhizin concentrations as low as 25 mg/ml effectively interfered with H5N1-induced ROS formation and with phosphorylation of the redox-sensitive MAPKs p38 and JNK. In our model, activation of p38 appears to be critical for H5N1-associated redox signalling since p38 inhibition had been shown before to mimick effects of the antioxidant N-acetyl-cysteine (NAC) [34] . Interestingly and in contrast to glycyrrhizin, NAC failed to inhibit H5N1 replication or H5N1-induced cytokine/chemokine expression in therapeutically relevant concentrations.
Glycyrrhizin diminished H5N1-induced cellular cytokine/ chemokine production in concentrations (#50 mg/ml) that did not interfere with H5N1 replication although redox-sensitive signalling pathways have been described to be involved in both processes. Therefore, H5N1-induced proinflammatory gene expression appears to be more sensitive to inhibition of ROS formation than H5N1 replication. Indeed, influenza viruses had been shown to induce cellular pathways through replicationdependent and -independent events [56] . In a previous report, we could show that similar glycyrrhizin concentrations like those investigated here interfered with H5N1-induced pro-inflammatory gene expression but not with H5N1 replication in human monocyte-derived macrophages [57] . In addition, other immunomodulatory treatment regimens that did not influence H5N1 replication reduced mortality in H5N1-infected mice [31, 58] . Therefore, glycyrrhizin represents a potential additional treatment option that interfers with both H5N1 replication and H5N1induced expression of pro-inflammatory cytokines in lung cells.
Interference with immune responses may also result in the loss of control of virus replication by cytotoxic immune cells including natural killer cells and cytotoxic CD8 + T-lymphocytes. Global immunosuppressants like corticosteroids failed to protect from lethal influenza virus infection [59] . Moreover, antiviral drugs may interfere with cytotoxic cells that control virus replication as demonstrated for ribavirin that was shown to hamper NK cell cytolytic activity [60] . In this context, glycyrrhizin had already been shown not to affect natural killer cell activity in the concentrations used here [57] .
In conclusion, we show in this report that therapeutic concentrations of glycyrrhizin (used as clinically approved parenteral preparation SNMC) interfere with highly pathogenic H5N1 influenza A virus replication and H5N1-induced proinflammatory gene expression at least in part through interference with H5N1-induced ROS formation and in turn reduced activation of p38, JNK, and NFkB in lung cells. Since we used the clinical formulation SNMC effects of other ingredients like glycin or cystein cannot be excluded. Vaccines and antiviral agents will fail to meet global needs at least at the beginning of a severe influenza A virus pandemic [61] . Anti-inflammatory and immunomodulatory agents are considered to be important candidates as constituents of anti-influenza treatment strategies that may save lives in an influenza pandemic situation [61] . Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1-caused disease.
|
What is the effect of oseltamivir and zanamivir?
| 5,236 |
neuraminidase inhibitors
| 2,189 |
1,596 |
Glycyrrhizin Exerts Antioxidative Effects in H5N1 Influenza A Virus-Infected Cells and Inhibits Virus Replication and Pro-Inflammatory Gene Expression
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3096629/
SHA: f3b7f4469ac01f1ce916d24172570c43c537627e
Authors: Michaelis, Martin; Geiler, Janina; Naczk, Patrizia; Sithisarn, Patchima; Leutz, Anke; Doerr, Hans Wilhelm; Cinatl, Jindrich
Date: 2011-05-17
DOI: 10.1371/journal.pone.0019705
License: cc-by
Abstract: Glycyrrhizin is known to exert antiviral and anti-inflammatory effects. Here, the effects of an approved parenteral glycyrrhizin preparation (Stronger Neo-Minophafen C) were investigated on highly pathogenic influenza A H5N1 virus replication, H5N1-induced apoptosis, and H5N1-induced pro-inflammatory responses in lung epithelial (A549) cells. Therapeutic glycyrrhizin concentrations substantially inhibited H5N1-induced expression of the pro-inflammatory molecules CXCL10, interleukin 6, CCL2, and CCL5 (effective glycyrrhizin concentrations 25 to 50 µg/ml) but interfered with H5N1 replication and H5N1-induced apoptosis to a lesser extent (effective glycyrrhizin concentrations 100 µg/ml or higher). Glycyrrhizin also diminished monocyte migration towards supernatants of H5N1-infected A549 cells. The mechanism by which glycyrrhizin interferes with H5N1 replication and H5N1-induced pro-inflammatory gene expression includes inhibition of H5N1-induced formation of reactive oxygen species and (in turn) reduced activation of NFκB, JNK, and p38, redox-sensitive signalling events known to be relevant for influenza A virus replication. Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1 disease.
Text: Highly pathogenic H5N1 influenza A viruses are considered to be potential influenza pandemic progenitors [1] [2] [3] [4] [5] [6] . At least for the first wave of an H5N1 pandemic, no sufficient amounts of adequate vaccines will be available [1] [2] [3] [4] [6] [7] [8] . Therefore, antiviral therapy for influenza A viruses including highly pathogenic H5N1 virus strains remains of great importance for the first line defense against the virus [1] [2] [3] [4] 6, 9] .
The neuraminidase inhibitors oseltamivir and zanamivir as well as the adamantanes amantadin and rimantadin that interfere with the influenza M2 protein are licensed for the treament of influenza [1] [2] [3] [4] 6] . However, the use of both drug classes is limited by the emergence of resistant virus strains. In seasonal influenza strains, the majority of H3N2 viruses and a great proportion of H1N1 viruses in humans are now considered to be amantadine-and rimantadine-resistant [10] [11] [12] [13] . Moreover, a drastic increase in oseltamivir-resistant H1N1 viruses has been reported during the 2007/2008 influenza season in the northern hemisphere [14] [15] [16] [17] . Preliminary data from the United States predict a further rise for the 2008/2009 season, possibly resulting in more than 90% of the circulating H1N1 strains to be oseltamivir resistant [14] .
H5N1 virus strains appear to be generally less sensitive to antiviral treatment than seasonal influenza A virus strains and treatment-resistant H5N1 strains emerge [1] [2] [3] [4] 6, [18] [19] [20] [21] . More-over, parenteral agents for the treatment of seriously ill patients are missing. Glycyrrhizin, a triterpene saponine, is a constituent of licorice root. It has been found to interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses including respiratory viruses such as respiratory syncytial virus, SARS coronavirus, HIV, and influenza viruses [22] [23] [24] [25] [26] [27] [28] . Moreover, antiinflammatory and immunomodulatory properties were attributed to glycyrrhizin [26] . The severity of human H5N1 disease has been associated with hypercytokinaemia (''cytokine storm'') [29, 30] . Delayed antiviral plus immunomodulator treatment reduced H5N1-induced mortality in mice [31] . Therefore, antiinflammatory and immunomodulatory effects exerted by glycyrrhizin may be beneficial for treatment of H5N1. Also, glycyrrhizin is a known antioxidant [26] and antioxidants were already shown to interfere with influenza A virus replication and virus-induced pro-inflammatory responses [32] [33] [34] .
Stronger Neo-Minophagen C (SNMC) is a glycyrrhizin preparation (available as tablets or parenteral formulation) that is approved in Japan for the treatment of chronic hepatic diseases and is marketed in Japan, China, Korea, Taiwan, Indonesia, India, and Mongolia. Here, we investigated the influence of SNMC on H5N1 replication, on H5N1-induced cytokine expression, on H5N1-induced cellular oxidative stress, and on critical H5N1-induced cellular signalling events in human pneumocytes (A549 cell line).
Glycyrrhizin (Stronger Neo Minophagen C) was obtained from Minophagen Pharmaceuticals Co., Ltd. (Tokyo, Japan).
The influenza strain A/Vietnam/1203/04 (H5N1) was received from the WHO Influenza Centre (National Institute for Medical Research, London, UK). The H5N1 influenza strain A/Thailand/ 1(Kan-1)/04 was obtained from Prof. Pilaipan Puthavathana (Mahidol University, Bangkok, Thailand).
Virus stocks were prepared by infecting Vero cells (African green monkey kidney; ATCC, Manassas, VA) and aliquots were stored at 280uC. Virus titres were determined as 50% tissue culture infectious dose (TCID 50 /ml) in confluent Vero cells in 96-well microtiter plates.
A549 cells (human lung carcinoma; ATCC: CCL-185, obtained from LGC Standards GmbH, Wesel, Germany) were grown at 37uC in minimal essential medium (MEM) supplemented with 10% FBS, 100 IU/ml of penicillin and 100 mg/ml streptomycin.
Human monocytes were isolated from buffy coats of healthy donors, obtained from Institute of Transfusion Medicine and Immune Haematology, German Red Cross Blood Donor Center, Johann Wolfgang Goethe-University, Frankfurt am Main. After centrifugation on Ficoll (Biocoll)-Hypaque density gradient (Biochrom AG, Berlin, Germany), mononuclear cells were collected from the interface and washed with PBS. Then, monocytes were isolated using magnetically labeled CD14 MicroBeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) following the manufacturer's instructions. Monocytes were cultivated in IMDM supplemented with 10% pooled human serum, 100 IU/ml of penicillin, and 100 mg/ml streptomycin.
The cellular viability was assessed on confluent cell layers with CellTiter-GloH Luminescent Cell Viability Assay (Promega GmbH, Mannheim, Germany) according to the manufacturers' protocol. Cell viability was expressed as percentage of non-treated control.
To determine intracellular NP localisation, H5N1-infected A549 were fixed 8 hours p.i. for 15 min with ice-cold acetone/ methanol (40:60, Mallinckrodt Baker B.V., Deventer, The Netherlands) and stained with a mouse monoclonal antibody (1 h incubation, 1:1000 in PBS) directed against the influenza A virus nucleoprotein (NP) (Millipore, Molsheim, France). An Alexa Fluor 488 goat anti-mouse IgG (H&L) (Invitrogen, Eugene, Oregon, USA) was used (1 h incubation, 1:1000 in PBS) as secondary antibody. Nuclei were stained using 49,6-diamidino-2phenylindole (DAPI) (Sigma-Aldrich Chemie GmbH, Munich, Germany). Fluorescence was visualised using Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
For flow cytometric analysis, the same antibodies were used.
The cytopathogenic effect (CPE) reduction assay was performed as described before [34] . Confluent A549 cell monolayers grown in 96-well microtitre plates were infected with influenza A strains at the indicated multiplicities of infection (MOIs). After a one hour adsorption period, cells were washed to remove non-detached virus. The virus-induced CPE was recorded at 24 h post infection (p.i.).
Unless otherwise stated, A549 cells were continuously treated with glycyrrhizin starting with a 1 h pre-incubation period. For time-ofaddition experiments, glycyrrhizin was added exclusively during the 1 h pre-incubation period, exclusively during the 1 h adsorption period, or after exclusively after the wash-out of input virus.
Total RNA was isolated from cell cultures using TRI reagent (Sigma-Aldrich, Munich, Germany). Real time PCR for H5 was performed using described methods [35] . The following primers were used: sense 59 acg tat gac tac ccg cag tat tca g 39; antisense 59 aga cca gcy acc atg att gc 39; probe 6-FAM-tca aca gtg gcg agt tcc cta gca-TAMRA.
The fraction of cells with fractional DNA content (''sub-G1'' cell subpopulation) indicates cytotoxicity. Sub-G1 cells are considered to be dead (usually apoptotic) cells. Cells were fixed with 70% ethanol for two hours at 220uC. The cellular DNA was stained using propidium iodide (20 mg/ml) and analysed by flow cytometry (FacsCalibur, BD Biosciences, Heidelberg, Germany).
Caspase activation was measured using the Caspase-Glo 8, 9, or 3/7 Assays (Promega, Mannheim, Germany) following the manufacturer's instructions.
Cell culture supernatants were collected and frozen at 280uC. Cytokines/chemokines were quantified by specific ELISA Duo Sets (R&D Systems GmbH, Wiesbaden, Germany) following the manufacturer's instructions.
NFkB activity was investigated in H5N1 (MOI 0.01)-infected cells by quantification of the NFkB subunits Rel A (p65) and NFkB1 (p50) from nuclear extracts using the TransAM TM transcription factor DNA-binding ELISAs (Active Motif, Rixensart, Belgium). Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction.
Cell culture supernatants were investigated for chemotactic activity by measurement of the activity to induce monocyte migration through membrane inserts in 24-well plates (pore size 8 mm; BD Biosciences, Heidelberg, Germany). Monocytes (1610 6 in 100 ml of IMDM with 10% pooled human serum) were added into the cell culture inserts (upper chamber) and cell culture supernatants (300 ml), were added to the lower chamber of the well. After a 48 h incubation period, cells were fixed with 4% paraformaldehyde and permeabilised with PBS containing 0.3% Tritron X-100. Then, nuclei were stained with 49,6-diamidino-2phenylindole (DAPI). The upper side of the membrane was wiped with a wet swab to remove the cells, while the lower side of the membrane was rinsed with PBS. The number of cells at the lower side of each membrane was quantified by counting of cells from three randomly chosen sections (3.7 mm 2 ) using an Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
Cells were lysed in Triton X-sample buffer and separated by SDS-PAGE. Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction. Proteins were detected using specific antibodies against bactin (Sigma-Aldrich Chemie GmbH, Munich, Germany), JNK, phosphorylated JNK, p38, or phosphorylated p38, (all purchased from New England Biolabs GmbH, Frankfurt am Main, Germany) and were visualised by enhanced chemiluminescence using a commercially available kit (Amersham, Freiburg, Germany).
Reactive oxygen species (ROS) were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes, distributed by Invitrogen, Karlsruhe, Germany).
Two groups were compared by t-test. More groups were compared by ANOVA with subsequent Student-Newman-Keuls test.
The A549 cell line, derived from a human pulmonary adenocarcinoma, is an established model for type II pneumocytes [36] , and commonly used for the investigation of the effect of influenza viruses on this cell type [see e.g. 6,37,38]. If not otherwise stated, glycyrrhizin was continuously present in cell culture media starting with a 1 h preinfection period. Glycyrrhizin 200 mg/ml (the maximum tested concentration) did not affect A549 cell viability (data not shown) but clearly decreased CPE formation in A549 cells infected with the H5N1 influenza strain A/Thailand/1(Kan-1)/04 at MOIs of 0.01, 0.1 or 1 ( Figure 1A ). Similar results were obtained in A549 cells infected with strain A/Vietnam/1203/04 (H5N1) (Suppl. Figure 1A) . Staining of A549 cells for influenza A nucleoprotein 24 h after infection with strain H5N1 A/Thailand/1(Kan-1)/04 indicated that glycyrrhizin 200 mg/ml significantly reduces the number of influenza A nucleoprotein positive cells ( Figure 1B) .
To examine the influence of glycyrrhizin on virus progeny, A549 cells were infected with the H5N1 influenza strain A/ Thailand/1(Kan-1)/04 at MOI 0.01 or MOI 1 and infectious virus titres were determined 24 h post infection ( Figure 1C ). While glycyrrhizin in concentrations up to 50 mg/ml did not affect H5N1 replication, moderate effects were exerted by glycyrrhizin 100 mg/ ml and more pronounced effects by glycyrrhizin 200 mg/ml (MOI 0.01: 13-fold reduction, MOI 1: 10-fold reduction). Next, influence of glycyrrhizin on H5N1 replication was confirmed by the detection of viral (H5) RNA using quantitative PCR. Only glycyrrhizin concentrations $100 mg/ml significantly reduced Figure 1B) or H5N1 A/Vietnam/1203/04-infected (Suppl. Figure 1C ) A549 cells (MOI 0.01) 24 h post infection.
Time-of-addition experiments revealed that maximal effects were achieved when glycyrrhizin was continuously present starting with a 1 h pre-incubation period ( Figure 1D ). Addition of glycyrrhizin post infection showed reduced antiviral effects while pre-incubation alone or glycyrrhizin addition during the adsorption period did not significantly affect H5N1 replication.
For investigation of H5N1-induced cytokine expression, five pro-inflammatory genes were chosen that had been correlated to severity of influenza disease: CXCL10 (also known as interferon-cinducible protein 10, IP-10), interleukin 6 (IL6), interleukin 8, (IL8; also known as CXCL8), CCL2 (also known as monocyte chemoattractant protein 1, MCP-1), and CCL5 (also known as RANTES). A549 cells were infected with H5N1 A/Thailand/ 1(Kan-1)/04 or H5N1 A/Vietnam/1203/04 at MOI 0.01, 0.1, or 1. Glycyrrhizin treatment was performed with 25, 50, 100, or 200 mg/ml. Cytokine expression was detected 24 h post infection by ELISA. Glycyrrhizin did not affect cytokine expression of noninfected cells (data not shown) but inhibited expression of all cytokines investigated in H5N1-infected cells in a dose-dependent manner (Figure 2, Figure 3A ). Effects were more pronounced at lower MOIs. Notably, expression of all cytokines except IL8 was significantly inhibited after treatment with glycyrrhizin 50 mg/ml Figure 3A ) although these glycyrrhizin concentrations had no effect on H5N1 replication in A549 cells (Figure 1, Figure S1 ).
Cytokine expression by influenza A virus-infected respiratory cells causes recruitment of peripheral blood monocytes into the lungs of patients where they differentiate to macrophages which are thought to contribute to influenza A virus pathogenicity [5, 39] . In a chemotaxis assay, the influence of glycyrrhizin was investigated on migration of monocytes towards supernatants of H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.1)-infected A549 cells through 8 mm filters. Monocyte migration towards supernatants of H5N1-infected cells was strongly increased relative to migration towards supernatants of non-infected cells. Treatment of H5N1- infected cells with glycyrrhizin 100 mg/ml clearly suppressed chemoattraction activity of supernatants ( Figure 3B ).
Influenza viruses including H5N1 have been shown to induce caspase-dependent apoptosis in airway cells and this apoptosis has been correlated to the virus pathogenicity [40, 41] . Glycyrrhizin concentrations up to 200 mg/ml did not affect caspase activation in non-infected cells ( Figure 4A-C) . Glycyrrhizin concentrations $100 mg/ml inhibited H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.01)-induced activation of the initiator caspases 8 and 9 as well as of the effector caspases 3/7 in A549 cells as determined 24 h post infection ( Figure 4A-C) . Lower glycyrrhizin concentrations did not affect H5N1-induced apoptosis. The detection of cells in sub-G1 phase resulted in similar findings ( Figure 4D ).
Substances that inhibit H5N1-induced caspase 3 activation including caspase 3 inhibitors cause nuclear retention of RNP complexes [34, 42] . In accordance, glycyrrhizin also interfered with nuclear export RNP at MOI 1 ( Figure S2 ). Similar results were obtained in MOI 0.01 H5N1 A/Thailand/1(Kan-1)/04infected cells ( Figure S3 ).
Influence of glycyrrhizin on H5N1-induced activation of nuclear factor kB (NFkB), p38, and on H5N1-induced cellular reactive oxygen species (ROS) formation Activation of NFkB, p38, and JNK have been associated with influenza A virus replication and virus-induced pro-inflammatory gene expression [34, [43] [44] [45] [46] [47] . While glycyrrhizin did not influence NFkB activity in non-infected A549 cells in the tested concentra-tions (data not shown), glycyrrhizin inhibited NFkB activation in H5N1-infected cells ( Figure 5A ). Moreover, glycyrrhizin inhibited H5N1-induced phosphorylation of the MAPKs p38 and JNK ( Figure 5B ).
In addition to their roles during influenza A virus replication and virus-induced cytokine/chemokine expression, NFkB, p38, and JNK are constituents of redox-sensitive signalling pathways [48] [49] [50] [51] . Antioxidants had been already found to interfere with influenza A virus-induced signalling through NFkB, p38, and JNK, with influenza A virus replication, and with influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] . Since glycyrrhizin is known to exert antioxidative effects [26] we speculated that glycyrrhizin may interfere with H5N1-induced ROS formation. Indeed glycyrrhizin exerted clear antioxidative effects in H5N1 (MOI 0.01)-infected cells ( Figure 5C ) causing significant reduction of ROS formation already at a concentration of 25 mg/ml ( Figure 5D ).
Here, we show that glycyrrhizin inhibits the replication of highly pathogenic H5N1 influenza A virus, H5N1-induced apoptosis, and H5N1-induced expression of pro-inflammatory cytokines in lung-derived A549 cells. After intravenous administration, achievable plasma concentrations of glycyrrhizin have been described to be about 100 mg/ml [52] . Therefore, the glycyrrhizin concentrations found to interfere with H5N1 replication and H5N1-induced pro-inflammatory gene expression in the present report are in the range of therapeutic plasma levels. Notably, although higher glycyrrhizin concentrations were needed to interfere with SARS coronavirus replication [22] than with H5N1 replication, beneficial results were reported in glycyrrhizin (SNMC)-treated SARS patients in comparison to SARS patients who did not receive glycyrrhizin [23] . Notably, investigation of different glycyrrhizin derivatives against SARS coronavirus led to the identification of compounds with enhanced antiviral activity [53] . Therefore, glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs.
Experimental results suggested that glycyrrhizin might be able to affect seasonal influenza A virus disease by antiviral and immunomodulatory effects [26, 27] . Mice were prevented from lethal H2N2 infection by glycyrrhizin although no influence on virus replication was detected. The mechanism was suggested to be induction of interferon-c in T-cells by glycyrrhizin [54] . Moreover, glycyrrhizin was shown to influence seasonal influenza A virus replication through interaction with the cell membrane [25, 28] . However, these effects were observed only in concentrations $200 mg/ml when glycyrrhizin was added during the virus adsorption period. Since glycyrrhizin addition during the adsorption period did not influence H5N1 replication in our experiments it appears not likely that membrane effects contribute to anti-H5N1 effects detected here in lower concentrations.
Our results rather suggest that glycyrrhizin interferes with H5N1-induced oxidative stress. Influenza A virus (including H5N1) infection induces ROS formation. Antioxidants were found to inhibit influenza A virus replication and influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] and glycyrrhizin is known to exert antioxidative effects [26] . Here, glycyrrhizin interfered with H5N1-induced activation of NFkB, p38, and JNK representing redox-sensitive signalling events [48] [49] [50] [51] involved in influenza A virus replication and influenza A virusinduced cellular cytokine/chemokine production [34, [43] [44] [45] [46] 55] . Glycyrrhizin 50 mg/ml significantly reduced H5N1-induced activation of NFkB. In addition, glycyrrhizin concentrations as low as 25 mg/ml effectively interfered with H5N1-induced ROS formation and with phosphorylation of the redox-sensitive MAPKs p38 and JNK. In our model, activation of p38 appears to be critical for H5N1-associated redox signalling since p38 inhibition had been shown before to mimick effects of the antioxidant N-acetyl-cysteine (NAC) [34] . Interestingly and in contrast to glycyrrhizin, NAC failed to inhibit H5N1 replication or H5N1-induced cytokine/chemokine expression in therapeutically relevant concentrations.
Glycyrrhizin diminished H5N1-induced cellular cytokine/ chemokine production in concentrations (#50 mg/ml) that did not interfere with H5N1 replication although redox-sensitive signalling pathways have been described to be involved in both processes. Therefore, H5N1-induced proinflammatory gene expression appears to be more sensitive to inhibition of ROS formation than H5N1 replication. Indeed, influenza viruses had been shown to induce cellular pathways through replicationdependent and -independent events [56] . In a previous report, we could show that similar glycyrrhizin concentrations like those investigated here interfered with H5N1-induced pro-inflammatory gene expression but not with H5N1 replication in human monocyte-derived macrophages [57] . In addition, other immunomodulatory treatment regimens that did not influence H5N1 replication reduced mortality in H5N1-infected mice [31, 58] . Therefore, glycyrrhizin represents a potential additional treatment option that interfers with both H5N1 replication and H5N1induced expression of pro-inflammatory cytokines in lung cells.
Interference with immune responses may also result in the loss of control of virus replication by cytotoxic immune cells including natural killer cells and cytotoxic CD8 + T-lymphocytes. Global immunosuppressants like corticosteroids failed to protect from lethal influenza virus infection [59] . Moreover, antiviral drugs may interfere with cytotoxic cells that control virus replication as demonstrated for ribavirin that was shown to hamper NK cell cytolytic activity [60] . In this context, glycyrrhizin had already been shown not to affect natural killer cell activity in the concentrations used here [57] .
In conclusion, we show in this report that therapeutic concentrations of glycyrrhizin (used as clinically approved parenteral preparation SNMC) interfere with highly pathogenic H5N1 influenza A virus replication and H5N1-induced proinflammatory gene expression at least in part through interference with H5N1-induced ROS formation and in turn reduced activation of p38, JNK, and NFkB in lung cells. Since we used the clinical formulation SNMC effects of other ingredients like glycin or cystein cannot be excluded. Vaccines and antiviral agents will fail to meet global needs at least at the beginning of a severe influenza A virus pandemic [61] . Anti-inflammatory and immunomodulatory agents are considered to be important candidates as constituents of anti-influenza treatment strategies that may save lives in an influenza pandemic situation [61] . Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1-caused disease.
|
What is Glycyrrhizin?
| 5,237 |
a triterpene saponine
| 3,358 |
1,596 |
Glycyrrhizin Exerts Antioxidative Effects in H5N1 Influenza A Virus-Infected Cells and Inhibits Virus Replication and Pro-Inflammatory Gene Expression
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3096629/
SHA: f3b7f4469ac01f1ce916d24172570c43c537627e
Authors: Michaelis, Martin; Geiler, Janina; Naczk, Patrizia; Sithisarn, Patchima; Leutz, Anke; Doerr, Hans Wilhelm; Cinatl, Jindrich
Date: 2011-05-17
DOI: 10.1371/journal.pone.0019705
License: cc-by
Abstract: Glycyrrhizin is known to exert antiviral and anti-inflammatory effects. Here, the effects of an approved parenteral glycyrrhizin preparation (Stronger Neo-Minophafen C) were investigated on highly pathogenic influenza A H5N1 virus replication, H5N1-induced apoptosis, and H5N1-induced pro-inflammatory responses in lung epithelial (A549) cells. Therapeutic glycyrrhizin concentrations substantially inhibited H5N1-induced expression of the pro-inflammatory molecules CXCL10, interleukin 6, CCL2, and CCL5 (effective glycyrrhizin concentrations 25 to 50 µg/ml) but interfered with H5N1 replication and H5N1-induced apoptosis to a lesser extent (effective glycyrrhizin concentrations 100 µg/ml or higher). Glycyrrhizin also diminished monocyte migration towards supernatants of H5N1-infected A549 cells. The mechanism by which glycyrrhizin interferes with H5N1 replication and H5N1-induced pro-inflammatory gene expression includes inhibition of H5N1-induced formation of reactive oxygen species and (in turn) reduced activation of NFκB, JNK, and p38, redox-sensitive signalling events known to be relevant for influenza A virus replication. Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1 disease.
Text: Highly pathogenic H5N1 influenza A viruses are considered to be potential influenza pandemic progenitors [1] [2] [3] [4] [5] [6] . At least for the first wave of an H5N1 pandemic, no sufficient amounts of adequate vaccines will be available [1] [2] [3] [4] [6] [7] [8] . Therefore, antiviral therapy for influenza A viruses including highly pathogenic H5N1 virus strains remains of great importance for the first line defense against the virus [1] [2] [3] [4] 6, 9] .
The neuraminidase inhibitors oseltamivir and zanamivir as well as the adamantanes amantadin and rimantadin that interfere with the influenza M2 protein are licensed for the treament of influenza [1] [2] [3] [4] 6] . However, the use of both drug classes is limited by the emergence of resistant virus strains. In seasonal influenza strains, the majority of H3N2 viruses and a great proportion of H1N1 viruses in humans are now considered to be amantadine-and rimantadine-resistant [10] [11] [12] [13] . Moreover, a drastic increase in oseltamivir-resistant H1N1 viruses has been reported during the 2007/2008 influenza season in the northern hemisphere [14] [15] [16] [17] . Preliminary data from the United States predict a further rise for the 2008/2009 season, possibly resulting in more than 90% of the circulating H1N1 strains to be oseltamivir resistant [14] .
H5N1 virus strains appear to be generally less sensitive to antiviral treatment than seasonal influenza A virus strains and treatment-resistant H5N1 strains emerge [1] [2] [3] [4] 6, [18] [19] [20] [21] . More-over, parenteral agents for the treatment of seriously ill patients are missing. Glycyrrhizin, a triterpene saponine, is a constituent of licorice root. It has been found to interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses including respiratory viruses such as respiratory syncytial virus, SARS coronavirus, HIV, and influenza viruses [22] [23] [24] [25] [26] [27] [28] . Moreover, antiinflammatory and immunomodulatory properties were attributed to glycyrrhizin [26] . The severity of human H5N1 disease has been associated with hypercytokinaemia (''cytokine storm'') [29, 30] . Delayed antiviral plus immunomodulator treatment reduced H5N1-induced mortality in mice [31] . Therefore, antiinflammatory and immunomodulatory effects exerted by glycyrrhizin may be beneficial for treatment of H5N1. Also, glycyrrhizin is a known antioxidant [26] and antioxidants were already shown to interfere with influenza A virus replication and virus-induced pro-inflammatory responses [32] [33] [34] .
Stronger Neo-Minophagen C (SNMC) is a glycyrrhizin preparation (available as tablets or parenteral formulation) that is approved in Japan for the treatment of chronic hepatic diseases and is marketed in Japan, China, Korea, Taiwan, Indonesia, India, and Mongolia. Here, we investigated the influence of SNMC on H5N1 replication, on H5N1-induced cytokine expression, on H5N1-induced cellular oxidative stress, and on critical H5N1-induced cellular signalling events in human pneumocytes (A549 cell line).
Glycyrrhizin (Stronger Neo Minophagen C) was obtained from Minophagen Pharmaceuticals Co., Ltd. (Tokyo, Japan).
The influenza strain A/Vietnam/1203/04 (H5N1) was received from the WHO Influenza Centre (National Institute for Medical Research, London, UK). The H5N1 influenza strain A/Thailand/ 1(Kan-1)/04 was obtained from Prof. Pilaipan Puthavathana (Mahidol University, Bangkok, Thailand).
Virus stocks were prepared by infecting Vero cells (African green monkey kidney; ATCC, Manassas, VA) and aliquots were stored at 280uC. Virus titres were determined as 50% tissue culture infectious dose (TCID 50 /ml) in confluent Vero cells in 96-well microtiter plates.
A549 cells (human lung carcinoma; ATCC: CCL-185, obtained from LGC Standards GmbH, Wesel, Germany) were grown at 37uC in minimal essential medium (MEM) supplemented with 10% FBS, 100 IU/ml of penicillin and 100 mg/ml streptomycin.
Human monocytes were isolated from buffy coats of healthy donors, obtained from Institute of Transfusion Medicine and Immune Haematology, German Red Cross Blood Donor Center, Johann Wolfgang Goethe-University, Frankfurt am Main. After centrifugation on Ficoll (Biocoll)-Hypaque density gradient (Biochrom AG, Berlin, Germany), mononuclear cells were collected from the interface and washed with PBS. Then, monocytes were isolated using magnetically labeled CD14 MicroBeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) following the manufacturer's instructions. Monocytes were cultivated in IMDM supplemented with 10% pooled human serum, 100 IU/ml of penicillin, and 100 mg/ml streptomycin.
The cellular viability was assessed on confluent cell layers with CellTiter-GloH Luminescent Cell Viability Assay (Promega GmbH, Mannheim, Germany) according to the manufacturers' protocol. Cell viability was expressed as percentage of non-treated control.
To determine intracellular NP localisation, H5N1-infected A549 were fixed 8 hours p.i. for 15 min with ice-cold acetone/ methanol (40:60, Mallinckrodt Baker B.V., Deventer, The Netherlands) and stained with a mouse monoclonal antibody (1 h incubation, 1:1000 in PBS) directed against the influenza A virus nucleoprotein (NP) (Millipore, Molsheim, France). An Alexa Fluor 488 goat anti-mouse IgG (H&L) (Invitrogen, Eugene, Oregon, USA) was used (1 h incubation, 1:1000 in PBS) as secondary antibody. Nuclei were stained using 49,6-diamidino-2phenylindole (DAPI) (Sigma-Aldrich Chemie GmbH, Munich, Germany). Fluorescence was visualised using Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
For flow cytometric analysis, the same antibodies were used.
The cytopathogenic effect (CPE) reduction assay was performed as described before [34] . Confluent A549 cell monolayers grown in 96-well microtitre plates were infected with influenza A strains at the indicated multiplicities of infection (MOIs). After a one hour adsorption period, cells were washed to remove non-detached virus. The virus-induced CPE was recorded at 24 h post infection (p.i.).
Unless otherwise stated, A549 cells were continuously treated with glycyrrhizin starting with a 1 h pre-incubation period. For time-ofaddition experiments, glycyrrhizin was added exclusively during the 1 h pre-incubation period, exclusively during the 1 h adsorption period, or after exclusively after the wash-out of input virus.
Total RNA was isolated from cell cultures using TRI reagent (Sigma-Aldrich, Munich, Germany). Real time PCR for H5 was performed using described methods [35] . The following primers were used: sense 59 acg tat gac tac ccg cag tat tca g 39; antisense 59 aga cca gcy acc atg att gc 39; probe 6-FAM-tca aca gtg gcg agt tcc cta gca-TAMRA.
The fraction of cells with fractional DNA content (''sub-G1'' cell subpopulation) indicates cytotoxicity. Sub-G1 cells are considered to be dead (usually apoptotic) cells. Cells were fixed with 70% ethanol for two hours at 220uC. The cellular DNA was stained using propidium iodide (20 mg/ml) and analysed by flow cytometry (FacsCalibur, BD Biosciences, Heidelberg, Germany).
Caspase activation was measured using the Caspase-Glo 8, 9, or 3/7 Assays (Promega, Mannheim, Germany) following the manufacturer's instructions.
Cell culture supernatants were collected and frozen at 280uC. Cytokines/chemokines were quantified by specific ELISA Duo Sets (R&D Systems GmbH, Wiesbaden, Germany) following the manufacturer's instructions.
NFkB activity was investigated in H5N1 (MOI 0.01)-infected cells by quantification of the NFkB subunits Rel A (p65) and NFkB1 (p50) from nuclear extracts using the TransAM TM transcription factor DNA-binding ELISAs (Active Motif, Rixensart, Belgium). Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction.
Cell culture supernatants were investigated for chemotactic activity by measurement of the activity to induce monocyte migration through membrane inserts in 24-well plates (pore size 8 mm; BD Biosciences, Heidelberg, Germany). Monocytes (1610 6 in 100 ml of IMDM with 10% pooled human serum) were added into the cell culture inserts (upper chamber) and cell culture supernatants (300 ml), were added to the lower chamber of the well. After a 48 h incubation period, cells were fixed with 4% paraformaldehyde and permeabilised with PBS containing 0.3% Tritron X-100. Then, nuclei were stained with 49,6-diamidino-2phenylindole (DAPI). The upper side of the membrane was wiped with a wet swab to remove the cells, while the lower side of the membrane was rinsed with PBS. The number of cells at the lower side of each membrane was quantified by counting of cells from three randomly chosen sections (3.7 mm 2 ) using an Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
Cells were lysed in Triton X-sample buffer and separated by SDS-PAGE. Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction. Proteins were detected using specific antibodies against bactin (Sigma-Aldrich Chemie GmbH, Munich, Germany), JNK, phosphorylated JNK, p38, or phosphorylated p38, (all purchased from New England Biolabs GmbH, Frankfurt am Main, Germany) and were visualised by enhanced chemiluminescence using a commercially available kit (Amersham, Freiburg, Germany).
Reactive oxygen species (ROS) were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes, distributed by Invitrogen, Karlsruhe, Germany).
Two groups were compared by t-test. More groups were compared by ANOVA with subsequent Student-Newman-Keuls test.
The A549 cell line, derived from a human pulmonary adenocarcinoma, is an established model for type II pneumocytes [36] , and commonly used for the investigation of the effect of influenza viruses on this cell type [see e.g. 6,37,38]. If not otherwise stated, glycyrrhizin was continuously present in cell culture media starting with a 1 h preinfection period. Glycyrrhizin 200 mg/ml (the maximum tested concentration) did not affect A549 cell viability (data not shown) but clearly decreased CPE formation in A549 cells infected with the H5N1 influenza strain A/Thailand/1(Kan-1)/04 at MOIs of 0.01, 0.1 or 1 ( Figure 1A ). Similar results were obtained in A549 cells infected with strain A/Vietnam/1203/04 (H5N1) (Suppl. Figure 1A) . Staining of A549 cells for influenza A nucleoprotein 24 h after infection with strain H5N1 A/Thailand/1(Kan-1)/04 indicated that glycyrrhizin 200 mg/ml significantly reduces the number of influenza A nucleoprotein positive cells ( Figure 1B) .
To examine the influence of glycyrrhizin on virus progeny, A549 cells were infected with the H5N1 influenza strain A/ Thailand/1(Kan-1)/04 at MOI 0.01 or MOI 1 and infectious virus titres were determined 24 h post infection ( Figure 1C ). While glycyrrhizin in concentrations up to 50 mg/ml did not affect H5N1 replication, moderate effects were exerted by glycyrrhizin 100 mg/ ml and more pronounced effects by glycyrrhizin 200 mg/ml (MOI 0.01: 13-fold reduction, MOI 1: 10-fold reduction). Next, influence of glycyrrhizin on H5N1 replication was confirmed by the detection of viral (H5) RNA using quantitative PCR. Only glycyrrhizin concentrations $100 mg/ml significantly reduced Figure 1B) or H5N1 A/Vietnam/1203/04-infected (Suppl. Figure 1C ) A549 cells (MOI 0.01) 24 h post infection.
Time-of-addition experiments revealed that maximal effects were achieved when glycyrrhizin was continuously present starting with a 1 h pre-incubation period ( Figure 1D ). Addition of glycyrrhizin post infection showed reduced antiviral effects while pre-incubation alone or glycyrrhizin addition during the adsorption period did not significantly affect H5N1 replication.
For investigation of H5N1-induced cytokine expression, five pro-inflammatory genes were chosen that had been correlated to severity of influenza disease: CXCL10 (also known as interferon-cinducible protein 10, IP-10), interleukin 6 (IL6), interleukin 8, (IL8; also known as CXCL8), CCL2 (also known as monocyte chemoattractant protein 1, MCP-1), and CCL5 (also known as RANTES). A549 cells were infected with H5N1 A/Thailand/ 1(Kan-1)/04 or H5N1 A/Vietnam/1203/04 at MOI 0.01, 0.1, or 1. Glycyrrhizin treatment was performed with 25, 50, 100, or 200 mg/ml. Cytokine expression was detected 24 h post infection by ELISA. Glycyrrhizin did not affect cytokine expression of noninfected cells (data not shown) but inhibited expression of all cytokines investigated in H5N1-infected cells in a dose-dependent manner (Figure 2, Figure 3A ). Effects were more pronounced at lower MOIs. Notably, expression of all cytokines except IL8 was significantly inhibited after treatment with glycyrrhizin 50 mg/ml Figure 3A ) although these glycyrrhizin concentrations had no effect on H5N1 replication in A549 cells (Figure 1, Figure S1 ).
Cytokine expression by influenza A virus-infected respiratory cells causes recruitment of peripheral blood monocytes into the lungs of patients where they differentiate to macrophages which are thought to contribute to influenza A virus pathogenicity [5, 39] . In a chemotaxis assay, the influence of glycyrrhizin was investigated on migration of monocytes towards supernatants of H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.1)-infected A549 cells through 8 mm filters. Monocyte migration towards supernatants of H5N1-infected cells was strongly increased relative to migration towards supernatants of non-infected cells. Treatment of H5N1- infected cells with glycyrrhizin 100 mg/ml clearly suppressed chemoattraction activity of supernatants ( Figure 3B ).
Influenza viruses including H5N1 have been shown to induce caspase-dependent apoptosis in airway cells and this apoptosis has been correlated to the virus pathogenicity [40, 41] . Glycyrrhizin concentrations up to 200 mg/ml did not affect caspase activation in non-infected cells ( Figure 4A-C) . Glycyrrhizin concentrations $100 mg/ml inhibited H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.01)-induced activation of the initiator caspases 8 and 9 as well as of the effector caspases 3/7 in A549 cells as determined 24 h post infection ( Figure 4A-C) . Lower glycyrrhizin concentrations did not affect H5N1-induced apoptosis. The detection of cells in sub-G1 phase resulted in similar findings ( Figure 4D ).
Substances that inhibit H5N1-induced caspase 3 activation including caspase 3 inhibitors cause nuclear retention of RNP complexes [34, 42] . In accordance, glycyrrhizin also interfered with nuclear export RNP at MOI 1 ( Figure S2 ). Similar results were obtained in MOI 0.01 H5N1 A/Thailand/1(Kan-1)/04infected cells ( Figure S3 ).
Influence of glycyrrhizin on H5N1-induced activation of nuclear factor kB (NFkB), p38, and on H5N1-induced cellular reactive oxygen species (ROS) formation Activation of NFkB, p38, and JNK have been associated with influenza A virus replication and virus-induced pro-inflammatory gene expression [34, [43] [44] [45] [46] [47] . While glycyrrhizin did not influence NFkB activity in non-infected A549 cells in the tested concentra-tions (data not shown), glycyrrhizin inhibited NFkB activation in H5N1-infected cells ( Figure 5A ). Moreover, glycyrrhizin inhibited H5N1-induced phosphorylation of the MAPKs p38 and JNK ( Figure 5B ).
In addition to their roles during influenza A virus replication and virus-induced cytokine/chemokine expression, NFkB, p38, and JNK are constituents of redox-sensitive signalling pathways [48] [49] [50] [51] . Antioxidants had been already found to interfere with influenza A virus-induced signalling through NFkB, p38, and JNK, with influenza A virus replication, and with influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] . Since glycyrrhizin is known to exert antioxidative effects [26] we speculated that glycyrrhizin may interfere with H5N1-induced ROS formation. Indeed glycyrrhizin exerted clear antioxidative effects in H5N1 (MOI 0.01)-infected cells ( Figure 5C ) causing significant reduction of ROS formation already at a concentration of 25 mg/ml ( Figure 5D ).
Here, we show that glycyrrhizin inhibits the replication of highly pathogenic H5N1 influenza A virus, H5N1-induced apoptosis, and H5N1-induced expression of pro-inflammatory cytokines in lung-derived A549 cells. After intravenous administration, achievable plasma concentrations of glycyrrhizin have been described to be about 100 mg/ml [52] . Therefore, the glycyrrhizin concentrations found to interfere with H5N1 replication and H5N1-induced pro-inflammatory gene expression in the present report are in the range of therapeutic plasma levels. Notably, although higher glycyrrhizin concentrations were needed to interfere with SARS coronavirus replication [22] than with H5N1 replication, beneficial results were reported in glycyrrhizin (SNMC)-treated SARS patients in comparison to SARS patients who did not receive glycyrrhizin [23] . Notably, investigation of different glycyrrhizin derivatives against SARS coronavirus led to the identification of compounds with enhanced antiviral activity [53] . Therefore, glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs.
Experimental results suggested that glycyrrhizin might be able to affect seasonal influenza A virus disease by antiviral and immunomodulatory effects [26, 27] . Mice were prevented from lethal H2N2 infection by glycyrrhizin although no influence on virus replication was detected. The mechanism was suggested to be induction of interferon-c in T-cells by glycyrrhizin [54] . Moreover, glycyrrhizin was shown to influence seasonal influenza A virus replication through interaction with the cell membrane [25, 28] . However, these effects were observed only in concentrations $200 mg/ml when glycyrrhizin was added during the virus adsorption period. Since glycyrrhizin addition during the adsorption period did not influence H5N1 replication in our experiments it appears not likely that membrane effects contribute to anti-H5N1 effects detected here in lower concentrations.
Our results rather suggest that glycyrrhizin interferes with H5N1-induced oxidative stress. Influenza A virus (including H5N1) infection induces ROS formation. Antioxidants were found to inhibit influenza A virus replication and influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] and glycyrrhizin is known to exert antioxidative effects [26] . Here, glycyrrhizin interfered with H5N1-induced activation of NFkB, p38, and JNK representing redox-sensitive signalling events [48] [49] [50] [51] involved in influenza A virus replication and influenza A virusinduced cellular cytokine/chemokine production [34, [43] [44] [45] [46] 55] . Glycyrrhizin 50 mg/ml significantly reduced H5N1-induced activation of NFkB. In addition, glycyrrhizin concentrations as low as 25 mg/ml effectively interfered with H5N1-induced ROS formation and with phosphorylation of the redox-sensitive MAPKs p38 and JNK. In our model, activation of p38 appears to be critical for H5N1-associated redox signalling since p38 inhibition had been shown before to mimick effects of the antioxidant N-acetyl-cysteine (NAC) [34] . Interestingly and in contrast to glycyrrhizin, NAC failed to inhibit H5N1 replication or H5N1-induced cytokine/chemokine expression in therapeutically relevant concentrations.
Glycyrrhizin diminished H5N1-induced cellular cytokine/ chemokine production in concentrations (#50 mg/ml) that did not interfere with H5N1 replication although redox-sensitive signalling pathways have been described to be involved in both processes. Therefore, H5N1-induced proinflammatory gene expression appears to be more sensitive to inhibition of ROS formation than H5N1 replication. Indeed, influenza viruses had been shown to induce cellular pathways through replicationdependent and -independent events [56] . In a previous report, we could show that similar glycyrrhizin concentrations like those investigated here interfered with H5N1-induced pro-inflammatory gene expression but not with H5N1 replication in human monocyte-derived macrophages [57] . In addition, other immunomodulatory treatment regimens that did not influence H5N1 replication reduced mortality in H5N1-infected mice [31, 58] . Therefore, glycyrrhizin represents a potential additional treatment option that interfers with both H5N1 replication and H5N1induced expression of pro-inflammatory cytokines in lung cells.
Interference with immune responses may also result in the loss of control of virus replication by cytotoxic immune cells including natural killer cells and cytotoxic CD8 + T-lymphocytes. Global immunosuppressants like corticosteroids failed to protect from lethal influenza virus infection [59] . Moreover, antiviral drugs may interfere with cytotoxic cells that control virus replication as demonstrated for ribavirin that was shown to hamper NK cell cytolytic activity [60] . In this context, glycyrrhizin had already been shown not to affect natural killer cell activity in the concentrations used here [57] .
In conclusion, we show in this report that therapeutic concentrations of glycyrrhizin (used as clinically approved parenteral preparation SNMC) interfere with highly pathogenic H5N1 influenza A virus replication and H5N1-induced proinflammatory gene expression at least in part through interference with H5N1-induced ROS formation and in turn reduced activation of p38, JNK, and NFkB in lung cells. Since we used the clinical formulation SNMC effects of other ingredients like glycin or cystein cannot be excluded. Vaccines and antiviral agents will fail to meet global needs at least at the beginning of a severe influenza A virus pandemic [61] . Anti-inflammatory and immunomodulatory agents are considered to be important candidates as constituents of anti-influenza treatment strategies that may save lives in an influenza pandemic situation [61] . Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1-caused disease.
|
What is the effect of Glycyrrhizin in viral infections?
| 5,238 |
interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses
| 3,437 |
1,596 |
Glycyrrhizin Exerts Antioxidative Effects in H5N1 Influenza A Virus-Infected Cells and Inhibits Virus Replication and Pro-Inflammatory Gene Expression
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3096629/
SHA: f3b7f4469ac01f1ce916d24172570c43c537627e
Authors: Michaelis, Martin; Geiler, Janina; Naczk, Patrizia; Sithisarn, Patchima; Leutz, Anke; Doerr, Hans Wilhelm; Cinatl, Jindrich
Date: 2011-05-17
DOI: 10.1371/journal.pone.0019705
License: cc-by
Abstract: Glycyrrhizin is known to exert antiviral and anti-inflammatory effects. Here, the effects of an approved parenteral glycyrrhizin preparation (Stronger Neo-Minophafen C) were investigated on highly pathogenic influenza A H5N1 virus replication, H5N1-induced apoptosis, and H5N1-induced pro-inflammatory responses in lung epithelial (A549) cells. Therapeutic glycyrrhizin concentrations substantially inhibited H5N1-induced expression of the pro-inflammatory molecules CXCL10, interleukin 6, CCL2, and CCL5 (effective glycyrrhizin concentrations 25 to 50 µg/ml) but interfered with H5N1 replication and H5N1-induced apoptosis to a lesser extent (effective glycyrrhizin concentrations 100 µg/ml or higher). Glycyrrhizin also diminished monocyte migration towards supernatants of H5N1-infected A549 cells. The mechanism by which glycyrrhizin interferes with H5N1 replication and H5N1-induced pro-inflammatory gene expression includes inhibition of H5N1-induced formation of reactive oxygen species and (in turn) reduced activation of NFκB, JNK, and p38, redox-sensitive signalling events known to be relevant for influenza A virus replication. Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1 disease.
Text: Highly pathogenic H5N1 influenza A viruses are considered to be potential influenza pandemic progenitors [1] [2] [3] [4] [5] [6] . At least for the first wave of an H5N1 pandemic, no sufficient amounts of adequate vaccines will be available [1] [2] [3] [4] [6] [7] [8] . Therefore, antiviral therapy for influenza A viruses including highly pathogenic H5N1 virus strains remains of great importance for the first line defense against the virus [1] [2] [3] [4] 6, 9] .
The neuraminidase inhibitors oseltamivir and zanamivir as well as the adamantanes amantadin and rimantadin that interfere with the influenza M2 protein are licensed for the treament of influenza [1] [2] [3] [4] 6] . However, the use of both drug classes is limited by the emergence of resistant virus strains. In seasonal influenza strains, the majority of H3N2 viruses and a great proportion of H1N1 viruses in humans are now considered to be amantadine-and rimantadine-resistant [10] [11] [12] [13] . Moreover, a drastic increase in oseltamivir-resistant H1N1 viruses has been reported during the 2007/2008 influenza season in the northern hemisphere [14] [15] [16] [17] . Preliminary data from the United States predict a further rise for the 2008/2009 season, possibly resulting in more than 90% of the circulating H1N1 strains to be oseltamivir resistant [14] .
H5N1 virus strains appear to be generally less sensitive to antiviral treatment than seasonal influenza A virus strains and treatment-resistant H5N1 strains emerge [1] [2] [3] [4] 6, [18] [19] [20] [21] . More-over, parenteral agents for the treatment of seriously ill patients are missing. Glycyrrhizin, a triterpene saponine, is a constituent of licorice root. It has been found to interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses including respiratory viruses such as respiratory syncytial virus, SARS coronavirus, HIV, and influenza viruses [22] [23] [24] [25] [26] [27] [28] . Moreover, antiinflammatory and immunomodulatory properties were attributed to glycyrrhizin [26] . The severity of human H5N1 disease has been associated with hypercytokinaemia (''cytokine storm'') [29, 30] . Delayed antiviral plus immunomodulator treatment reduced H5N1-induced mortality in mice [31] . Therefore, antiinflammatory and immunomodulatory effects exerted by glycyrrhizin may be beneficial for treatment of H5N1. Also, glycyrrhizin is a known antioxidant [26] and antioxidants were already shown to interfere with influenza A virus replication and virus-induced pro-inflammatory responses [32] [33] [34] .
Stronger Neo-Minophagen C (SNMC) is a glycyrrhizin preparation (available as tablets or parenteral formulation) that is approved in Japan for the treatment of chronic hepatic diseases and is marketed in Japan, China, Korea, Taiwan, Indonesia, India, and Mongolia. Here, we investigated the influence of SNMC on H5N1 replication, on H5N1-induced cytokine expression, on H5N1-induced cellular oxidative stress, and on critical H5N1-induced cellular signalling events in human pneumocytes (A549 cell line).
Glycyrrhizin (Stronger Neo Minophagen C) was obtained from Minophagen Pharmaceuticals Co., Ltd. (Tokyo, Japan).
The influenza strain A/Vietnam/1203/04 (H5N1) was received from the WHO Influenza Centre (National Institute for Medical Research, London, UK). The H5N1 influenza strain A/Thailand/ 1(Kan-1)/04 was obtained from Prof. Pilaipan Puthavathana (Mahidol University, Bangkok, Thailand).
Virus stocks were prepared by infecting Vero cells (African green monkey kidney; ATCC, Manassas, VA) and aliquots were stored at 280uC. Virus titres were determined as 50% tissue culture infectious dose (TCID 50 /ml) in confluent Vero cells in 96-well microtiter plates.
A549 cells (human lung carcinoma; ATCC: CCL-185, obtained from LGC Standards GmbH, Wesel, Germany) were grown at 37uC in minimal essential medium (MEM) supplemented with 10% FBS, 100 IU/ml of penicillin and 100 mg/ml streptomycin.
Human monocytes were isolated from buffy coats of healthy donors, obtained from Institute of Transfusion Medicine and Immune Haematology, German Red Cross Blood Donor Center, Johann Wolfgang Goethe-University, Frankfurt am Main. After centrifugation on Ficoll (Biocoll)-Hypaque density gradient (Biochrom AG, Berlin, Germany), mononuclear cells were collected from the interface and washed with PBS. Then, monocytes were isolated using magnetically labeled CD14 MicroBeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) following the manufacturer's instructions. Monocytes were cultivated in IMDM supplemented with 10% pooled human serum, 100 IU/ml of penicillin, and 100 mg/ml streptomycin.
The cellular viability was assessed on confluent cell layers with CellTiter-GloH Luminescent Cell Viability Assay (Promega GmbH, Mannheim, Germany) according to the manufacturers' protocol. Cell viability was expressed as percentage of non-treated control.
To determine intracellular NP localisation, H5N1-infected A549 were fixed 8 hours p.i. for 15 min with ice-cold acetone/ methanol (40:60, Mallinckrodt Baker B.V., Deventer, The Netherlands) and stained with a mouse monoclonal antibody (1 h incubation, 1:1000 in PBS) directed against the influenza A virus nucleoprotein (NP) (Millipore, Molsheim, France). An Alexa Fluor 488 goat anti-mouse IgG (H&L) (Invitrogen, Eugene, Oregon, USA) was used (1 h incubation, 1:1000 in PBS) as secondary antibody. Nuclei were stained using 49,6-diamidino-2phenylindole (DAPI) (Sigma-Aldrich Chemie GmbH, Munich, Germany). Fluorescence was visualised using Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
For flow cytometric analysis, the same antibodies were used.
The cytopathogenic effect (CPE) reduction assay was performed as described before [34] . Confluent A549 cell monolayers grown in 96-well microtitre plates were infected with influenza A strains at the indicated multiplicities of infection (MOIs). After a one hour adsorption period, cells were washed to remove non-detached virus. The virus-induced CPE was recorded at 24 h post infection (p.i.).
Unless otherwise stated, A549 cells were continuously treated with glycyrrhizin starting with a 1 h pre-incubation period. For time-ofaddition experiments, glycyrrhizin was added exclusively during the 1 h pre-incubation period, exclusively during the 1 h adsorption period, or after exclusively after the wash-out of input virus.
Total RNA was isolated from cell cultures using TRI reagent (Sigma-Aldrich, Munich, Germany). Real time PCR for H5 was performed using described methods [35] . The following primers were used: sense 59 acg tat gac tac ccg cag tat tca g 39; antisense 59 aga cca gcy acc atg att gc 39; probe 6-FAM-tca aca gtg gcg agt tcc cta gca-TAMRA.
The fraction of cells with fractional DNA content (''sub-G1'' cell subpopulation) indicates cytotoxicity. Sub-G1 cells are considered to be dead (usually apoptotic) cells. Cells were fixed with 70% ethanol for two hours at 220uC. The cellular DNA was stained using propidium iodide (20 mg/ml) and analysed by flow cytometry (FacsCalibur, BD Biosciences, Heidelberg, Germany).
Caspase activation was measured using the Caspase-Glo 8, 9, or 3/7 Assays (Promega, Mannheim, Germany) following the manufacturer's instructions.
Cell culture supernatants were collected and frozen at 280uC. Cytokines/chemokines were quantified by specific ELISA Duo Sets (R&D Systems GmbH, Wiesbaden, Germany) following the manufacturer's instructions.
NFkB activity was investigated in H5N1 (MOI 0.01)-infected cells by quantification of the NFkB subunits Rel A (p65) and NFkB1 (p50) from nuclear extracts using the TransAM TM transcription factor DNA-binding ELISAs (Active Motif, Rixensart, Belgium). Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction.
Cell culture supernatants were investigated for chemotactic activity by measurement of the activity to induce monocyte migration through membrane inserts in 24-well plates (pore size 8 mm; BD Biosciences, Heidelberg, Germany). Monocytes (1610 6 in 100 ml of IMDM with 10% pooled human serum) were added into the cell culture inserts (upper chamber) and cell culture supernatants (300 ml), were added to the lower chamber of the well. After a 48 h incubation period, cells were fixed with 4% paraformaldehyde and permeabilised with PBS containing 0.3% Tritron X-100. Then, nuclei were stained with 49,6-diamidino-2phenylindole (DAPI). The upper side of the membrane was wiped with a wet swab to remove the cells, while the lower side of the membrane was rinsed with PBS. The number of cells at the lower side of each membrane was quantified by counting of cells from three randomly chosen sections (3.7 mm 2 ) using an Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
Cells were lysed in Triton X-sample buffer and separated by SDS-PAGE. Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction. Proteins were detected using specific antibodies against bactin (Sigma-Aldrich Chemie GmbH, Munich, Germany), JNK, phosphorylated JNK, p38, or phosphorylated p38, (all purchased from New England Biolabs GmbH, Frankfurt am Main, Germany) and were visualised by enhanced chemiluminescence using a commercially available kit (Amersham, Freiburg, Germany).
Reactive oxygen species (ROS) were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes, distributed by Invitrogen, Karlsruhe, Germany).
Two groups were compared by t-test. More groups were compared by ANOVA with subsequent Student-Newman-Keuls test.
The A549 cell line, derived from a human pulmonary adenocarcinoma, is an established model for type II pneumocytes [36] , and commonly used for the investigation of the effect of influenza viruses on this cell type [see e.g. 6,37,38]. If not otherwise stated, glycyrrhizin was continuously present in cell culture media starting with a 1 h preinfection period. Glycyrrhizin 200 mg/ml (the maximum tested concentration) did not affect A549 cell viability (data not shown) but clearly decreased CPE formation in A549 cells infected with the H5N1 influenza strain A/Thailand/1(Kan-1)/04 at MOIs of 0.01, 0.1 or 1 ( Figure 1A ). Similar results were obtained in A549 cells infected with strain A/Vietnam/1203/04 (H5N1) (Suppl. Figure 1A) . Staining of A549 cells for influenza A nucleoprotein 24 h after infection with strain H5N1 A/Thailand/1(Kan-1)/04 indicated that glycyrrhizin 200 mg/ml significantly reduces the number of influenza A nucleoprotein positive cells ( Figure 1B) .
To examine the influence of glycyrrhizin on virus progeny, A549 cells were infected with the H5N1 influenza strain A/ Thailand/1(Kan-1)/04 at MOI 0.01 or MOI 1 and infectious virus titres were determined 24 h post infection ( Figure 1C ). While glycyrrhizin in concentrations up to 50 mg/ml did not affect H5N1 replication, moderate effects were exerted by glycyrrhizin 100 mg/ ml and more pronounced effects by glycyrrhizin 200 mg/ml (MOI 0.01: 13-fold reduction, MOI 1: 10-fold reduction). Next, influence of glycyrrhizin on H5N1 replication was confirmed by the detection of viral (H5) RNA using quantitative PCR. Only glycyrrhizin concentrations $100 mg/ml significantly reduced Figure 1B) or H5N1 A/Vietnam/1203/04-infected (Suppl. Figure 1C ) A549 cells (MOI 0.01) 24 h post infection.
Time-of-addition experiments revealed that maximal effects were achieved when glycyrrhizin was continuously present starting with a 1 h pre-incubation period ( Figure 1D ). Addition of glycyrrhizin post infection showed reduced antiviral effects while pre-incubation alone or glycyrrhizin addition during the adsorption period did not significantly affect H5N1 replication.
For investigation of H5N1-induced cytokine expression, five pro-inflammatory genes were chosen that had been correlated to severity of influenza disease: CXCL10 (also known as interferon-cinducible protein 10, IP-10), interleukin 6 (IL6), interleukin 8, (IL8; also known as CXCL8), CCL2 (also known as monocyte chemoattractant protein 1, MCP-1), and CCL5 (also known as RANTES). A549 cells were infected with H5N1 A/Thailand/ 1(Kan-1)/04 or H5N1 A/Vietnam/1203/04 at MOI 0.01, 0.1, or 1. Glycyrrhizin treatment was performed with 25, 50, 100, or 200 mg/ml. Cytokine expression was detected 24 h post infection by ELISA. Glycyrrhizin did not affect cytokine expression of noninfected cells (data not shown) but inhibited expression of all cytokines investigated in H5N1-infected cells in a dose-dependent manner (Figure 2, Figure 3A ). Effects were more pronounced at lower MOIs. Notably, expression of all cytokines except IL8 was significantly inhibited after treatment with glycyrrhizin 50 mg/ml Figure 3A ) although these glycyrrhizin concentrations had no effect on H5N1 replication in A549 cells (Figure 1, Figure S1 ).
Cytokine expression by influenza A virus-infected respiratory cells causes recruitment of peripheral blood monocytes into the lungs of patients where they differentiate to macrophages which are thought to contribute to influenza A virus pathogenicity [5, 39] . In a chemotaxis assay, the influence of glycyrrhizin was investigated on migration of monocytes towards supernatants of H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.1)-infected A549 cells through 8 mm filters. Monocyte migration towards supernatants of H5N1-infected cells was strongly increased relative to migration towards supernatants of non-infected cells. Treatment of H5N1- infected cells with glycyrrhizin 100 mg/ml clearly suppressed chemoattraction activity of supernatants ( Figure 3B ).
Influenza viruses including H5N1 have been shown to induce caspase-dependent apoptosis in airway cells and this apoptosis has been correlated to the virus pathogenicity [40, 41] . Glycyrrhizin concentrations up to 200 mg/ml did not affect caspase activation in non-infected cells ( Figure 4A-C) . Glycyrrhizin concentrations $100 mg/ml inhibited H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.01)-induced activation of the initiator caspases 8 and 9 as well as of the effector caspases 3/7 in A549 cells as determined 24 h post infection ( Figure 4A-C) . Lower glycyrrhizin concentrations did not affect H5N1-induced apoptosis. The detection of cells in sub-G1 phase resulted in similar findings ( Figure 4D ).
Substances that inhibit H5N1-induced caspase 3 activation including caspase 3 inhibitors cause nuclear retention of RNP complexes [34, 42] . In accordance, glycyrrhizin also interfered with nuclear export RNP at MOI 1 ( Figure S2 ). Similar results were obtained in MOI 0.01 H5N1 A/Thailand/1(Kan-1)/04infected cells ( Figure S3 ).
Influence of glycyrrhizin on H5N1-induced activation of nuclear factor kB (NFkB), p38, and on H5N1-induced cellular reactive oxygen species (ROS) formation Activation of NFkB, p38, and JNK have been associated with influenza A virus replication and virus-induced pro-inflammatory gene expression [34, [43] [44] [45] [46] [47] . While glycyrrhizin did not influence NFkB activity in non-infected A549 cells in the tested concentra-tions (data not shown), glycyrrhizin inhibited NFkB activation in H5N1-infected cells ( Figure 5A ). Moreover, glycyrrhizin inhibited H5N1-induced phosphorylation of the MAPKs p38 and JNK ( Figure 5B ).
In addition to their roles during influenza A virus replication and virus-induced cytokine/chemokine expression, NFkB, p38, and JNK are constituents of redox-sensitive signalling pathways [48] [49] [50] [51] . Antioxidants had been already found to interfere with influenza A virus-induced signalling through NFkB, p38, and JNK, with influenza A virus replication, and with influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] . Since glycyrrhizin is known to exert antioxidative effects [26] we speculated that glycyrrhizin may interfere with H5N1-induced ROS formation. Indeed glycyrrhizin exerted clear antioxidative effects in H5N1 (MOI 0.01)-infected cells ( Figure 5C ) causing significant reduction of ROS formation already at a concentration of 25 mg/ml ( Figure 5D ).
Here, we show that glycyrrhizin inhibits the replication of highly pathogenic H5N1 influenza A virus, H5N1-induced apoptosis, and H5N1-induced expression of pro-inflammatory cytokines in lung-derived A549 cells. After intravenous administration, achievable plasma concentrations of glycyrrhizin have been described to be about 100 mg/ml [52] . Therefore, the glycyrrhizin concentrations found to interfere with H5N1 replication and H5N1-induced pro-inflammatory gene expression in the present report are in the range of therapeutic plasma levels. Notably, although higher glycyrrhizin concentrations were needed to interfere with SARS coronavirus replication [22] than with H5N1 replication, beneficial results were reported in glycyrrhizin (SNMC)-treated SARS patients in comparison to SARS patients who did not receive glycyrrhizin [23] . Notably, investigation of different glycyrrhizin derivatives against SARS coronavirus led to the identification of compounds with enhanced antiviral activity [53] . Therefore, glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs.
Experimental results suggested that glycyrrhizin might be able to affect seasonal influenza A virus disease by antiviral and immunomodulatory effects [26, 27] . Mice were prevented from lethal H2N2 infection by glycyrrhizin although no influence on virus replication was detected. The mechanism was suggested to be induction of interferon-c in T-cells by glycyrrhizin [54] . Moreover, glycyrrhizin was shown to influence seasonal influenza A virus replication through interaction with the cell membrane [25, 28] . However, these effects were observed only in concentrations $200 mg/ml when glycyrrhizin was added during the virus adsorption period. Since glycyrrhizin addition during the adsorption period did not influence H5N1 replication in our experiments it appears not likely that membrane effects contribute to anti-H5N1 effects detected here in lower concentrations.
Our results rather suggest that glycyrrhizin interferes with H5N1-induced oxidative stress. Influenza A virus (including H5N1) infection induces ROS formation. Antioxidants were found to inhibit influenza A virus replication and influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] and glycyrrhizin is known to exert antioxidative effects [26] . Here, glycyrrhizin interfered with H5N1-induced activation of NFkB, p38, and JNK representing redox-sensitive signalling events [48] [49] [50] [51] involved in influenza A virus replication and influenza A virusinduced cellular cytokine/chemokine production [34, [43] [44] [45] [46] 55] . Glycyrrhizin 50 mg/ml significantly reduced H5N1-induced activation of NFkB. In addition, glycyrrhizin concentrations as low as 25 mg/ml effectively interfered with H5N1-induced ROS formation and with phosphorylation of the redox-sensitive MAPKs p38 and JNK. In our model, activation of p38 appears to be critical for H5N1-associated redox signalling since p38 inhibition had been shown before to mimick effects of the antioxidant N-acetyl-cysteine (NAC) [34] . Interestingly and in contrast to glycyrrhizin, NAC failed to inhibit H5N1 replication or H5N1-induced cytokine/chemokine expression in therapeutically relevant concentrations.
Glycyrrhizin diminished H5N1-induced cellular cytokine/ chemokine production in concentrations (#50 mg/ml) that did not interfere with H5N1 replication although redox-sensitive signalling pathways have been described to be involved in both processes. Therefore, H5N1-induced proinflammatory gene expression appears to be more sensitive to inhibition of ROS formation than H5N1 replication. Indeed, influenza viruses had been shown to induce cellular pathways through replicationdependent and -independent events [56] . In a previous report, we could show that similar glycyrrhizin concentrations like those investigated here interfered with H5N1-induced pro-inflammatory gene expression but not with H5N1 replication in human monocyte-derived macrophages [57] . In addition, other immunomodulatory treatment regimens that did not influence H5N1 replication reduced mortality in H5N1-infected mice [31, 58] . Therefore, glycyrrhizin represents a potential additional treatment option that interfers with both H5N1 replication and H5N1induced expression of pro-inflammatory cytokines in lung cells.
Interference with immune responses may also result in the loss of control of virus replication by cytotoxic immune cells including natural killer cells and cytotoxic CD8 + T-lymphocytes. Global immunosuppressants like corticosteroids failed to protect from lethal influenza virus infection [59] . Moreover, antiviral drugs may interfere with cytotoxic cells that control virus replication as demonstrated for ribavirin that was shown to hamper NK cell cytolytic activity [60] . In this context, glycyrrhizin had already been shown not to affect natural killer cell activity in the concentrations used here [57] .
In conclusion, we show in this report that therapeutic concentrations of glycyrrhizin (used as clinically approved parenteral preparation SNMC) interfere with highly pathogenic H5N1 influenza A virus replication and H5N1-induced proinflammatory gene expression at least in part through interference with H5N1-induced ROS formation and in turn reduced activation of p38, JNK, and NFkB in lung cells. Since we used the clinical formulation SNMC effects of other ingredients like glycin or cystein cannot be excluded. Vaccines and antiviral agents will fail to meet global needs at least at the beginning of a severe influenza A virus pandemic [61] . Anti-inflammatory and immunomodulatory agents are considered to be important candidates as constituents of anti-influenza treatment strategies that may save lives in an influenza pandemic situation [61] . Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1-caused disease.
|
What is another word for hypercytokinaemia?
| 5,239 |
cytokine storm'
| 3,853 |
1,596 |
Glycyrrhizin Exerts Antioxidative Effects in H5N1 Influenza A Virus-Infected Cells and Inhibits Virus Replication and Pro-Inflammatory Gene Expression
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3096629/
SHA: f3b7f4469ac01f1ce916d24172570c43c537627e
Authors: Michaelis, Martin; Geiler, Janina; Naczk, Patrizia; Sithisarn, Patchima; Leutz, Anke; Doerr, Hans Wilhelm; Cinatl, Jindrich
Date: 2011-05-17
DOI: 10.1371/journal.pone.0019705
License: cc-by
Abstract: Glycyrrhizin is known to exert antiviral and anti-inflammatory effects. Here, the effects of an approved parenteral glycyrrhizin preparation (Stronger Neo-Minophafen C) were investigated on highly pathogenic influenza A H5N1 virus replication, H5N1-induced apoptosis, and H5N1-induced pro-inflammatory responses in lung epithelial (A549) cells. Therapeutic glycyrrhizin concentrations substantially inhibited H5N1-induced expression of the pro-inflammatory molecules CXCL10, interleukin 6, CCL2, and CCL5 (effective glycyrrhizin concentrations 25 to 50 µg/ml) but interfered with H5N1 replication and H5N1-induced apoptosis to a lesser extent (effective glycyrrhizin concentrations 100 µg/ml or higher). Glycyrrhizin also diminished monocyte migration towards supernatants of H5N1-infected A549 cells. The mechanism by which glycyrrhizin interferes with H5N1 replication and H5N1-induced pro-inflammatory gene expression includes inhibition of H5N1-induced formation of reactive oxygen species and (in turn) reduced activation of NFκB, JNK, and p38, redox-sensitive signalling events known to be relevant for influenza A virus replication. Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1 disease.
Text: Highly pathogenic H5N1 influenza A viruses are considered to be potential influenza pandemic progenitors [1] [2] [3] [4] [5] [6] . At least for the first wave of an H5N1 pandemic, no sufficient amounts of adequate vaccines will be available [1] [2] [3] [4] [6] [7] [8] . Therefore, antiviral therapy for influenza A viruses including highly pathogenic H5N1 virus strains remains of great importance for the first line defense against the virus [1] [2] [3] [4] 6, 9] .
The neuraminidase inhibitors oseltamivir and zanamivir as well as the adamantanes amantadin and rimantadin that interfere with the influenza M2 protein are licensed for the treament of influenza [1] [2] [3] [4] 6] . However, the use of both drug classes is limited by the emergence of resistant virus strains. In seasonal influenza strains, the majority of H3N2 viruses and a great proportion of H1N1 viruses in humans are now considered to be amantadine-and rimantadine-resistant [10] [11] [12] [13] . Moreover, a drastic increase in oseltamivir-resistant H1N1 viruses has been reported during the 2007/2008 influenza season in the northern hemisphere [14] [15] [16] [17] . Preliminary data from the United States predict a further rise for the 2008/2009 season, possibly resulting in more than 90% of the circulating H1N1 strains to be oseltamivir resistant [14] .
H5N1 virus strains appear to be generally less sensitive to antiviral treatment than seasonal influenza A virus strains and treatment-resistant H5N1 strains emerge [1] [2] [3] [4] 6, [18] [19] [20] [21] . More-over, parenteral agents for the treatment of seriously ill patients are missing. Glycyrrhizin, a triterpene saponine, is a constituent of licorice root. It has been found to interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses including respiratory viruses such as respiratory syncytial virus, SARS coronavirus, HIV, and influenza viruses [22] [23] [24] [25] [26] [27] [28] . Moreover, antiinflammatory and immunomodulatory properties were attributed to glycyrrhizin [26] . The severity of human H5N1 disease has been associated with hypercytokinaemia (''cytokine storm'') [29, 30] . Delayed antiviral plus immunomodulator treatment reduced H5N1-induced mortality in mice [31] . Therefore, antiinflammatory and immunomodulatory effects exerted by glycyrrhizin may be beneficial for treatment of H5N1. Also, glycyrrhizin is a known antioxidant [26] and antioxidants were already shown to interfere with influenza A virus replication and virus-induced pro-inflammatory responses [32] [33] [34] .
Stronger Neo-Minophagen C (SNMC) is a glycyrrhizin preparation (available as tablets or parenteral formulation) that is approved in Japan for the treatment of chronic hepatic diseases and is marketed in Japan, China, Korea, Taiwan, Indonesia, India, and Mongolia. Here, we investigated the influence of SNMC on H5N1 replication, on H5N1-induced cytokine expression, on H5N1-induced cellular oxidative stress, and on critical H5N1-induced cellular signalling events in human pneumocytes (A549 cell line).
Glycyrrhizin (Stronger Neo Minophagen C) was obtained from Minophagen Pharmaceuticals Co., Ltd. (Tokyo, Japan).
The influenza strain A/Vietnam/1203/04 (H5N1) was received from the WHO Influenza Centre (National Institute for Medical Research, London, UK). The H5N1 influenza strain A/Thailand/ 1(Kan-1)/04 was obtained from Prof. Pilaipan Puthavathana (Mahidol University, Bangkok, Thailand).
Virus stocks were prepared by infecting Vero cells (African green monkey kidney; ATCC, Manassas, VA) and aliquots were stored at 280uC. Virus titres were determined as 50% tissue culture infectious dose (TCID 50 /ml) in confluent Vero cells in 96-well microtiter plates.
A549 cells (human lung carcinoma; ATCC: CCL-185, obtained from LGC Standards GmbH, Wesel, Germany) were grown at 37uC in minimal essential medium (MEM) supplemented with 10% FBS, 100 IU/ml of penicillin and 100 mg/ml streptomycin.
Human monocytes were isolated from buffy coats of healthy donors, obtained from Institute of Transfusion Medicine and Immune Haematology, German Red Cross Blood Donor Center, Johann Wolfgang Goethe-University, Frankfurt am Main. After centrifugation on Ficoll (Biocoll)-Hypaque density gradient (Biochrom AG, Berlin, Germany), mononuclear cells were collected from the interface and washed with PBS. Then, monocytes were isolated using magnetically labeled CD14 MicroBeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) following the manufacturer's instructions. Monocytes were cultivated in IMDM supplemented with 10% pooled human serum, 100 IU/ml of penicillin, and 100 mg/ml streptomycin.
The cellular viability was assessed on confluent cell layers with CellTiter-GloH Luminescent Cell Viability Assay (Promega GmbH, Mannheim, Germany) according to the manufacturers' protocol. Cell viability was expressed as percentage of non-treated control.
To determine intracellular NP localisation, H5N1-infected A549 were fixed 8 hours p.i. for 15 min with ice-cold acetone/ methanol (40:60, Mallinckrodt Baker B.V., Deventer, The Netherlands) and stained with a mouse monoclonal antibody (1 h incubation, 1:1000 in PBS) directed against the influenza A virus nucleoprotein (NP) (Millipore, Molsheim, France). An Alexa Fluor 488 goat anti-mouse IgG (H&L) (Invitrogen, Eugene, Oregon, USA) was used (1 h incubation, 1:1000 in PBS) as secondary antibody. Nuclei were stained using 49,6-diamidino-2phenylindole (DAPI) (Sigma-Aldrich Chemie GmbH, Munich, Germany). Fluorescence was visualised using Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
For flow cytometric analysis, the same antibodies were used.
The cytopathogenic effect (CPE) reduction assay was performed as described before [34] . Confluent A549 cell monolayers grown in 96-well microtitre plates were infected with influenza A strains at the indicated multiplicities of infection (MOIs). After a one hour adsorption period, cells were washed to remove non-detached virus. The virus-induced CPE was recorded at 24 h post infection (p.i.).
Unless otherwise stated, A549 cells were continuously treated with glycyrrhizin starting with a 1 h pre-incubation period. For time-ofaddition experiments, glycyrrhizin was added exclusively during the 1 h pre-incubation period, exclusively during the 1 h adsorption period, or after exclusively after the wash-out of input virus.
Total RNA was isolated from cell cultures using TRI reagent (Sigma-Aldrich, Munich, Germany). Real time PCR for H5 was performed using described methods [35] . The following primers were used: sense 59 acg tat gac tac ccg cag tat tca g 39; antisense 59 aga cca gcy acc atg att gc 39; probe 6-FAM-tca aca gtg gcg agt tcc cta gca-TAMRA.
The fraction of cells with fractional DNA content (''sub-G1'' cell subpopulation) indicates cytotoxicity. Sub-G1 cells are considered to be dead (usually apoptotic) cells. Cells were fixed with 70% ethanol for two hours at 220uC. The cellular DNA was stained using propidium iodide (20 mg/ml) and analysed by flow cytometry (FacsCalibur, BD Biosciences, Heidelberg, Germany).
Caspase activation was measured using the Caspase-Glo 8, 9, or 3/7 Assays (Promega, Mannheim, Germany) following the manufacturer's instructions.
Cell culture supernatants were collected and frozen at 280uC. Cytokines/chemokines were quantified by specific ELISA Duo Sets (R&D Systems GmbH, Wiesbaden, Germany) following the manufacturer's instructions.
NFkB activity was investigated in H5N1 (MOI 0.01)-infected cells by quantification of the NFkB subunits Rel A (p65) and NFkB1 (p50) from nuclear extracts using the TransAM TM transcription factor DNA-binding ELISAs (Active Motif, Rixensart, Belgium). Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction.
Cell culture supernatants were investigated for chemotactic activity by measurement of the activity to induce monocyte migration through membrane inserts in 24-well plates (pore size 8 mm; BD Biosciences, Heidelberg, Germany). Monocytes (1610 6 in 100 ml of IMDM with 10% pooled human serum) were added into the cell culture inserts (upper chamber) and cell culture supernatants (300 ml), were added to the lower chamber of the well. After a 48 h incubation period, cells were fixed with 4% paraformaldehyde and permeabilised with PBS containing 0.3% Tritron X-100. Then, nuclei were stained with 49,6-diamidino-2phenylindole (DAPI). The upper side of the membrane was wiped with a wet swab to remove the cells, while the lower side of the membrane was rinsed with PBS. The number of cells at the lower side of each membrane was quantified by counting of cells from three randomly chosen sections (3.7 mm 2 ) using an Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
Cells were lysed in Triton X-sample buffer and separated by SDS-PAGE. Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction. Proteins were detected using specific antibodies against bactin (Sigma-Aldrich Chemie GmbH, Munich, Germany), JNK, phosphorylated JNK, p38, or phosphorylated p38, (all purchased from New England Biolabs GmbH, Frankfurt am Main, Germany) and were visualised by enhanced chemiluminescence using a commercially available kit (Amersham, Freiburg, Germany).
Reactive oxygen species (ROS) were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes, distributed by Invitrogen, Karlsruhe, Germany).
Two groups were compared by t-test. More groups were compared by ANOVA with subsequent Student-Newman-Keuls test.
The A549 cell line, derived from a human pulmonary adenocarcinoma, is an established model for type II pneumocytes [36] , and commonly used for the investigation of the effect of influenza viruses on this cell type [see e.g. 6,37,38]. If not otherwise stated, glycyrrhizin was continuously present in cell culture media starting with a 1 h preinfection period. Glycyrrhizin 200 mg/ml (the maximum tested concentration) did not affect A549 cell viability (data not shown) but clearly decreased CPE formation in A549 cells infected with the H5N1 influenza strain A/Thailand/1(Kan-1)/04 at MOIs of 0.01, 0.1 or 1 ( Figure 1A ). Similar results were obtained in A549 cells infected with strain A/Vietnam/1203/04 (H5N1) (Suppl. Figure 1A) . Staining of A549 cells for influenza A nucleoprotein 24 h after infection with strain H5N1 A/Thailand/1(Kan-1)/04 indicated that glycyrrhizin 200 mg/ml significantly reduces the number of influenza A nucleoprotein positive cells ( Figure 1B) .
To examine the influence of glycyrrhizin on virus progeny, A549 cells were infected with the H5N1 influenza strain A/ Thailand/1(Kan-1)/04 at MOI 0.01 or MOI 1 and infectious virus titres were determined 24 h post infection ( Figure 1C ). While glycyrrhizin in concentrations up to 50 mg/ml did not affect H5N1 replication, moderate effects were exerted by glycyrrhizin 100 mg/ ml and more pronounced effects by glycyrrhizin 200 mg/ml (MOI 0.01: 13-fold reduction, MOI 1: 10-fold reduction). Next, influence of glycyrrhizin on H5N1 replication was confirmed by the detection of viral (H5) RNA using quantitative PCR. Only glycyrrhizin concentrations $100 mg/ml significantly reduced Figure 1B) or H5N1 A/Vietnam/1203/04-infected (Suppl. Figure 1C ) A549 cells (MOI 0.01) 24 h post infection.
Time-of-addition experiments revealed that maximal effects were achieved when glycyrrhizin was continuously present starting with a 1 h pre-incubation period ( Figure 1D ). Addition of glycyrrhizin post infection showed reduced antiviral effects while pre-incubation alone or glycyrrhizin addition during the adsorption period did not significantly affect H5N1 replication.
For investigation of H5N1-induced cytokine expression, five pro-inflammatory genes were chosen that had been correlated to severity of influenza disease: CXCL10 (also known as interferon-cinducible protein 10, IP-10), interleukin 6 (IL6), interleukin 8, (IL8; also known as CXCL8), CCL2 (also known as monocyte chemoattractant protein 1, MCP-1), and CCL5 (also known as RANTES). A549 cells were infected with H5N1 A/Thailand/ 1(Kan-1)/04 or H5N1 A/Vietnam/1203/04 at MOI 0.01, 0.1, or 1. Glycyrrhizin treatment was performed with 25, 50, 100, or 200 mg/ml. Cytokine expression was detected 24 h post infection by ELISA. Glycyrrhizin did not affect cytokine expression of noninfected cells (data not shown) but inhibited expression of all cytokines investigated in H5N1-infected cells in a dose-dependent manner (Figure 2, Figure 3A ). Effects were more pronounced at lower MOIs. Notably, expression of all cytokines except IL8 was significantly inhibited after treatment with glycyrrhizin 50 mg/ml Figure 3A ) although these glycyrrhizin concentrations had no effect on H5N1 replication in A549 cells (Figure 1, Figure S1 ).
Cytokine expression by influenza A virus-infected respiratory cells causes recruitment of peripheral blood monocytes into the lungs of patients where they differentiate to macrophages which are thought to contribute to influenza A virus pathogenicity [5, 39] . In a chemotaxis assay, the influence of glycyrrhizin was investigated on migration of monocytes towards supernatants of H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.1)-infected A549 cells through 8 mm filters. Monocyte migration towards supernatants of H5N1-infected cells was strongly increased relative to migration towards supernatants of non-infected cells. Treatment of H5N1- infected cells with glycyrrhizin 100 mg/ml clearly suppressed chemoattraction activity of supernatants ( Figure 3B ).
Influenza viruses including H5N1 have been shown to induce caspase-dependent apoptosis in airway cells and this apoptosis has been correlated to the virus pathogenicity [40, 41] . Glycyrrhizin concentrations up to 200 mg/ml did not affect caspase activation in non-infected cells ( Figure 4A-C) . Glycyrrhizin concentrations $100 mg/ml inhibited H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.01)-induced activation of the initiator caspases 8 and 9 as well as of the effector caspases 3/7 in A549 cells as determined 24 h post infection ( Figure 4A-C) . Lower glycyrrhizin concentrations did not affect H5N1-induced apoptosis. The detection of cells in sub-G1 phase resulted in similar findings ( Figure 4D ).
Substances that inhibit H5N1-induced caspase 3 activation including caspase 3 inhibitors cause nuclear retention of RNP complexes [34, 42] . In accordance, glycyrrhizin also interfered with nuclear export RNP at MOI 1 ( Figure S2 ). Similar results were obtained in MOI 0.01 H5N1 A/Thailand/1(Kan-1)/04infected cells ( Figure S3 ).
Influence of glycyrrhizin on H5N1-induced activation of nuclear factor kB (NFkB), p38, and on H5N1-induced cellular reactive oxygen species (ROS) formation Activation of NFkB, p38, and JNK have been associated with influenza A virus replication and virus-induced pro-inflammatory gene expression [34, [43] [44] [45] [46] [47] . While glycyrrhizin did not influence NFkB activity in non-infected A549 cells in the tested concentra-tions (data not shown), glycyrrhizin inhibited NFkB activation in H5N1-infected cells ( Figure 5A ). Moreover, glycyrrhizin inhibited H5N1-induced phosphorylation of the MAPKs p38 and JNK ( Figure 5B ).
In addition to their roles during influenza A virus replication and virus-induced cytokine/chemokine expression, NFkB, p38, and JNK are constituents of redox-sensitive signalling pathways [48] [49] [50] [51] . Antioxidants had been already found to interfere with influenza A virus-induced signalling through NFkB, p38, and JNK, with influenza A virus replication, and with influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] . Since glycyrrhizin is known to exert antioxidative effects [26] we speculated that glycyrrhizin may interfere with H5N1-induced ROS formation. Indeed glycyrrhizin exerted clear antioxidative effects in H5N1 (MOI 0.01)-infected cells ( Figure 5C ) causing significant reduction of ROS formation already at a concentration of 25 mg/ml ( Figure 5D ).
Here, we show that glycyrrhizin inhibits the replication of highly pathogenic H5N1 influenza A virus, H5N1-induced apoptosis, and H5N1-induced expression of pro-inflammatory cytokines in lung-derived A549 cells. After intravenous administration, achievable plasma concentrations of glycyrrhizin have been described to be about 100 mg/ml [52] . Therefore, the glycyrrhizin concentrations found to interfere with H5N1 replication and H5N1-induced pro-inflammatory gene expression in the present report are in the range of therapeutic plasma levels. Notably, although higher glycyrrhizin concentrations were needed to interfere with SARS coronavirus replication [22] than with H5N1 replication, beneficial results were reported in glycyrrhizin (SNMC)-treated SARS patients in comparison to SARS patients who did not receive glycyrrhizin [23] . Notably, investigation of different glycyrrhizin derivatives against SARS coronavirus led to the identification of compounds with enhanced antiviral activity [53] . Therefore, glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs.
Experimental results suggested that glycyrrhizin might be able to affect seasonal influenza A virus disease by antiviral and immunomodulatory effects [26, 27] . Mice were prevented from lethal H2N2 infection by glycyrrhizin although no influence on virus replication was detected. The mechanism was suggested to be induction of interferon-c in T-cells by glycyrrhizin [54] . Moreover, glycyrrhizin was shown to influence seasonal influenza A virus replication through interaction with the cell membrane [25, 28] . However, these effects were observed only in concentrations $200 mg/ml when glycyrrhizin was added during the virus adsorption period. Since glycyrrhizin addition during the adsorption period did not influence H5N1 replication in our experiments it appears not likely that membrane effects contribute to anti-H5N1 effects detected here in lower concentrations.
Our results rather suggest that glycyrrhizin interferes with H5N1-induced oxidative stress. Influenza A virus (including H5N1) infection induces ROS formation. Antioxidants were found to inhibit influenza A virus replication and influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] and glycyrrhizin is known to exert antioxidative effects [26] . Here, glycyrrhizin interfered with H5N1-induced activation of NFkB, p38, and JNK representing redox-sensitive signalling events [48] [49] [50] [51] involved in influenza A virus replication and influenza A virusinduced cellular cytokine/chemokine production [34, [43] [44] [45] [46] 55] . Glycyrrhizin 50 mg/ml significantly reduced H5N1-induced activation of NFkB. In addition, glycyrrhizin concentrations as low as 25 mg/ml effectively interfered with H5N1-induced ROS formation and with phosphorylation of the redox-sensitive MAPKs p38 and JNK. In our model, activation of p38 appears to be critical for H5N1-associated redox signalling since p38 inhibition had been shown before to mimick effects of the antioxidant N-acetyl-cysteine (NAC) [34] . Interestingly and in contrast to glycyrrhizin, NAC failed to inhibit H5N1 replication or H5N1-induced cytokine/chemokine expression in therapeutically relevant concentrations.
Glycyrrhizin diminished H5N1-induced cellular cytokine/ chemokine production in concentrations (#50 mg/ml) that did not interfere with H5N1 replication although redox-sensitive signalling pathways have been described to be involved in both processes. Therefore, H5N1-induced proinflammatory gene expression appears to be more sensitive to inhibition of ROS formation than H5N1 replication. Indeed, influenza viruses had been shown to induce cellular pathways through replicationdependent and -independent events [56] . In a previous report, we could show that similar glycyrrhizin concentrations like those investigated here interfered with H5N1-induced pro-inflammatory gene expression but not with H5N1 replication in human monocyte-derived macrophages [57] . In addition, other immunomodulatory treatment regimens that did not influence H5N1 replication reduced mortality in H5N1-infected mice [31, 58] . Therefore, glycyrrhizin represents a potential additional treatment option that interfers with both H5N1 replication and H5N1induced expression of pro-inflammatory cytokines in lung cells.
Interference with immune responses may also result in the loss of control of virus replication by cytotoxic immune cells including natural killer cells and cytotoxic CD8 + T-lymphocytes. Global immunosuppressants like corticosteroids failed to protect from lethal influenza virus infection [59] . Moreover, antiviral drugs may interfere with cytotoxic cells that control virus replication as demonstrated for ribavirin that was shown to hamper NK cell cytolytic activity [60] . In this context, glycyrrhizin had already been shown not to affect natural killer cell activity in the concentrations used here [57] .
In conclusion, we show in this report that therapeutic concentrations of glycyrrhizin (used as clinically approved parenteral preparation SNMC) interfere with highly pathogenic H5N1 influenza A virus replication and H5N1-induced proinflammatory gene expression at least in part through interference with H5N1-induced ROS formation and in turn reduced activation of p38, JNK, and NFkB in lung cells. Since we used the clinical formulation SNMC effects of other ingredients like glycin or cystein cannot be excluded. Vaccines and antiviral agents will fail to meet global needs at least at the beginning of a severe influenza A virus pandemic [61] . Anti-inflammatory and immunomodulatory agents are considered to be important candidates as constituents of anti-influenza treatment strategies that may save lives in an influenza pandemic situation [61] . Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1-caused disease.
|
What has been correlated with the pathogenicity of the H5N1 infection?
| 5,240 |
caspase-dependent apoptosis in airway cells
| 15,524 |
1,596 |
Glycyrrhizin Exerts Antioxidative Effects in H5N1 Influenza A Virus-Infected Cells and Inhibits Virus Replication and Pro-Inflammatory Gene Expression
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3096629/
SHA: f3b7f4469ac01f1ce916d24172570c43c537627e
Authors: Michaelis, Martin; Geiler, Janina; Naczk, Patrizia; Sithisarn, Patchima; Leutz, Anke; Doerr, Hans Wilhelm; Cinatl, Jindrich
Date: 2011-05-17
DOI: 10.1371/journal.pone.0019705
License: cc-by
Abstract: Glycyrrhizin is known to exert antiviral and anti-inflammatory effects. Here, the effects of an approved parenteral glycyrrhizin preparation (Stronger Neo-Minophafen C) were investigated on highly pathogenic influenza A H5N1 virus replication, H5N1-induced apoptosis, and H5N1-induced pro-inflammatory responses in lung epithelial (A549) cells. Therapeutic glycyrrhizin concentrations substantially inhibited H5N1-induced expression of the pro-inflammatory molecules CXCL10, interleukin 6, CCL2, and CCL5 (effective glycyrrhizin concentrations 25 to 50 µg/ml) but interfered with H5N1 replication and H5N1-induced apoptosis to a lesser extent (effective glycyrrhizin concentrations 100 µg/ml or higher). Glycyrrhizin also diminished monocyte migration towards supernatants of H5N1-infected A549 cells. The mechanism by which glycyrrhizin interferes with H5N1 replication and H5N1-induced pro-inflammatory gene expression includes inhibition of H5N1-induced formation of reactive oxygen species and (in turn) reduced activation of NFκB, JNK, and p38, redox-sensitive signalling events known to be relevant for influenza A virus replication. Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1 disease.
Text: Highly pathogenic H5N1 influenza A viruses are considered to be potential influenza pandemic progenitors [1] [2] [3] [4] [5] [6] . At least for the first wave of an H5N1 pandemic, no sufficient amounts of adequate vaccines will be available [1] [2] [3] [4] [6] [7] [8] . Therefore, antiviral therapy for influenza A viruses including highly pathogenic H5N1 virus strains remains of great importance for the first line defense against the virus [1] [2] [3] [4] 6, 9] .
The neuraminidase inhibitors oseltamivir and zanamivir as well as the adamantanes amantadin and rimantadin that interfere with the influenza M2 protein are licensed for the treament of influenza [1] [2] [3] [4] 6] . However, the use of both drug classes is limited by the emergence of resistant virus strains. In seasonal influenza strains, the majority of H3N2 viruses and a great proportion of H1N1 viruses in humans are now considered to be amantadine-and rimantadine-resistant [10] [11] [12] [13] . Moreover, a drastic increase in oseltamivir-resistant H1N1 viruses has been reported during the 2007/2008 influenza season in the northern hemisphere [14] [15] [16] [17] . Preliminary data from the United States predict a further rise for the 2008/2009 season, possibly resulting in more than 90% of the circulating H1N1 strains to be oseltamivir resistant [14] .
H5N1 virus strains appear to be generally less sensitive to antiviral treatment than seasonal influenza A virus strains and treatment-resistant H5N1 strains emerge [1] [2] [3] [4] 6, [18] [19] [20] [21] . More-over, parenteral agents for the treatment of seriously ill patients are missing. Glycyrrhizin, a triterpene saponine, is a constituent of licorice root. It has been found to interfere with replication and/or cytopathogenic effect (CPE) induction of many viruses including respiratory viruses such as respiratory syncytial virus, SARS coronavirus, HIV, and influenza viruses [22] [23] [24] [25] [26] [27] [28] . Moreover, antiinflammatory and immunomodulatory properties were attributed to glycyrrhizin [26] . The severity of human H5N1 disease has been associated with hypercytokinaemia (''cytokine storm'') [29, 30] . Delayed antiviral plus immunomodulator treatment reduced H5N1-induced mortality in mice [31] . Therefore, antiinflammatory and immunomodulatory effects exerted by glycyrrhizin may be beneficial for treatment of H5N1. Also, glycyrrhizin is a known antioxidant [26] and antioxidants were already shown to interfere with influenza A virus replication and virus-induced pro-inflammatory responses [32] [33] [34] .
Stronger Neo-Minophagen C (SNMC) is a glycyrrhizin preparation (available as tablets or parenteral formulation) that is approved in Japan for the treatment of chronic hepatic diseases and is marketed in Japan, China, Korea, Taiwan, Indonesia, India, and Mongolia. Here, we investigated the influence of SNMC on H5N1 replication, on H5N1-induced cytokine expression, on H5N1-induced cellular oxidative stress, and on critical H5N1-induced cellular signalling events in human pneumocytes (A549 cell line).
Glycyrrhizin (Stronger Neo Minophagen C) was obtained from Minophagen Pharmaceuticals Co., Ltd. (Tokyo, Japan).
The influenza strain A/Vietnam/1203/04 (H5N1) was received from the WHO Influenza Centre (National Institute for Medical Research, London, UK). The H5N1 influenza strain A/Thailand/ 1(Kan-1)/04 was obtained from Prof. Pilaipan Puthavathana (Mahidol University, Bangkok, Thailand).
Virus stocks were prepared by infecting Vero cells (African green monkey kidney; ATCC, Manassas, VA) and aliquots were stored at 280uC. Virus titres were determined as 50% tissue culture infectious dose (TCID 50 /ml) in confluent Vero cells in 96-well microtiter plates.
A549 cells (human lung carcinoma; ATCC: CCL-185, obtained from LGC Standards GmbH, Wesel, Germany) were grown at 37uC in minimal essential medium (MEM) supplemented with 10% FBS, 100 IU/ml of penicillin and 100 mg/ml streptomycin.
Human monocytes were isolated from buffy coats of healthy donors, obtained from Institute of Transfusion Medicine and Immune Haematology, German Red Cross Blood Donor Center, Johann Wolfgang Goethe-University, Frankfurt am Main. After centrifugation on Ficoll (Biocoll)-Hypaque density gradient (Biochrom AG, Berlin, Germany), mononuclear cells were collected from the interface and washed with PBS. Then, monocytes were isolated using magnetically labeled CD14 MicroBeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) following the manufacturer's instructions. Monocytes were cultivated in IMDM supplemented with 10% pooled human serum, 100 IU/ml of penicillin, and 100 mg/ml streptomycin.
The cellular viability was assessed on confluent cell layers with CellTiter-GloH Luminescent Cell Viability Assay (Promega GmbH, Mannheim, Germany) according to the manufacturers' protocol. Cell viability was expressed as percentage of non-treated control.
To determine intracellular NP localisation, H5N1-infected A549 were fixed 8 hours p.i. for 15 min with ice-cold acetone/ methanol (40:60, Mallinckrodt Baker B.V., Deventer, The Netherlands) and stained with a mouse monoclonal antibody (1 h incubation, 1:1000 in PBS) directed against the influenza A virus nucleoprotein (NP) (Millipore, Molsheim, France). An Alexa Fluor 488 goat anti-mouse IgG (H&L) (Invitrogen, Eugene, Oregon, USA) was used (1 h incubation, 1:1000 in PBS) as secondary antibody. Nuclei were stained using 49,6-diamidino-2phenylindole (DAPI) (Sigma-Aldrich Chemie GmbH, Munich, Germany). Fluorescence was visualised using Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
For flow cytometric analysis, the same antibodies were used.
The cytopathogenic effect (CPE) reduction assay was performed as described before [34] . Confluent A549 cell monolayers grown in 96-well microtitre plates were infected with influenza A strains at the indicated multiplicities of infection (MOIs). After a one hour adsorption period, cells were washed to remove non-detached virus. The virus-induced CPE was recorded at 24 h post infection (p.i.).
Unless otherwise stated, A549 cells were continuously treated with glycyrrhizin starting with a 1 h pre-incubation period. For time-ofaddition experiments, glycyrrhizin was added exclusively during the 1 h pre-incubation period, exclusively during the 1 h adsorption period, or after exclusively after the wash-out of input virus.
Total RNA was isolated from cell cultures using TRI reagent (Sigma-Aldrich, Munich, Germany). Real time PCR for H5 was performed using described methods [35] . The following primers were used: sense 59 acg tat gac tac ccg cag tat tca g 39; antisense 59 aga cca gcy acc atg att gc 39; probe 6-FAM-tca aca gtg gcg agt tcc cta gca-TAMRA.
The fraction of cells with fractional DNA content (''sub-G1'' cell subpopulation) indicates cytotoxicity. Sub-G1 cells are considered to be dead (usually apoptotic) cells. Cells were fixed with 70% ethanol for two hours at 220uC. The cellular DNA was stained using propidium iodide (20 mg/ml) and analysed by flow cytometry (FacsCalibur, BD Biosciences, Heidelberg, Germany).
Caspase activation was measured using the Caspase-Glo 8, 9, or 3/7 Assays (Promega, Mannheim, Germany) following the manufacturer's instructions.
Cell culture supernatants were collected and frozen at 280uC. Cytokines/chemokines were quantified by specific ELISA Duo Sets (R&D Systems GmbH, Wiesbaden, Germany) following the manufacturer's instructions.
NFkB activity was investigated in H5N1 (MOI 0.01)-infected cells by quantification of the NFkB subunits Rel A (p65) and NFkB1 (p50) from nuclear extracts using the TransAM TM transcription factor DNA-binding ELISAs (Active Motif, Rixensart, Belgium). Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction.
Cell culture supernatants were investigated for chemotactic activity by measurement of the activity to induce monocyte migration through membrane inserts in 24-well plates (pore size 8 mm; BD Biosciences, Heidelberg, Germany). Monocytes (1610 6 in 100 ml of IMDM with 10% pooled human serum) were added into the cell culture inserts (upper chamber) and cell culture supernatants (300 ml), were added to the lower chamber of the well. After a 48 h incubation period, cells were fixed with 4% paraformaldehyde and permeabilised with PBS containing 0.3% Tritron X-100. Then, nuclei were stained with 49,6-diamidino-2phenylindole (DAPI). The upper side of the membrane was wiped with a wet swab to remove the cells, while the lower side of the membrane was rinsed with PBS. The number of cells at the lower side of each membrane was quantified by counting of cells from three randomly chosen sections (3.7 mm 2 ) using an Olympus IX 1 fluorescence microscope (Olympus, Planegg, Germany).
Cells were lysed in Triton X-sample buffer and separated by SDS-PAGE. Nuclear extract were prepared using the Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA) following the manufacturer's instruction. Proteins were detected using specific antibodies against bactin (Sigma-Aldrich Chemie GmbH, Munich, Germany), JNK, phosphorylated JNK, p38, or phosphorylated p38, (all purchased from New England Biolabs GmbH, Frankfurt am Main, Germany) and were visualised by enhanced chemiluminescence using a commercially available kit (Amersham, Freiburg, Germany).
Reactive oxygen species (ROS) were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes, distributed by Invitrogen, Karlsruhe, Germany).
Two groups were compared by t-test. More groups were compared by ANOVA with subsequent Student-Newman-Keuls test.
The A549 cell line, derived from a human pulmonary adenocarcinoma, is an established model for type II pneumocytes [36] , and commonly used for the investigation of the effect of influenza viruses on this cell type [see e.g. 6,37,38]. If not otherwise stated, glycyrrhizin was continuously present in cell culture media starting with a 1 h preinfection period. Glycyrrhizin 200 mg/ml (the maximum tested concentration) did not affect A549 cell viability (data not shown) but clearly decreased CPE formation in A549 cells infected with the H5N1 influenza strain A/Thailand/1(Kan-1)/04 at MOIs of 0.01, 0.1 or 1 ( Figure 1A ). Similar results were obtained in A549 cells infected with strain A/Vietnam/1203/04 (H5N1) (Suppl. Figure 1A) . Staining of A549 cells for influenza A nucleoprotein 24 h after infection with strain H5N1 A/Thailand/1(Kan-1)/04 indicated that glycyrrhizin 200 mg/ml significantly reduces the number of influenza A nucleoprotein positive cells ( Figure 1B) .
To examine the influence of glycyrrhizin on virus progeny, A549 cells were infected with the H5N1 influenza strain A/ Thailand/1(Kan-1)/04 at MOI 0.01 or MOI 1 and infectious virus titres were determined 24 h post infection ( Figure 1C ). While glycyrrhizin in concentrations up to 50 mg/ml did not affect H5N1 replication, moderate effects were exerted by glycyrrhizin 100 mg/ ml and more pronounced effects by glycyrrhizin 200 mg/ml (MOI 0.01: 13-fold reduction, MOI 1: 10-fold reduction). Next, influence of glycyrrhizin on H5N1 replication was confirmed by the detection of viral (H5) RNA using quantitative PCR. Only glycyrrhizin concentrations $100 mg/ml significantly reduced Figure 1B) or H5N1 A/Vietnam/1203/04-infected (Suppl. Figure 1C ) A549 cells (MOI 0.01) 24 h post infection.
Time-of-addition experiments revealed that maximal effects were achieved when glycyrrhizin was continuously present starting with a 1 h pre-incubation period ( Figure 1D ). Addition of glycyrrhizin post infection showed reduced antiviral effects while pre-incubation alone or glycyrrhizin addition during the adsorption period did not significantly affect H5N1 replication.
For investigation of H5N1-induced cytokine expression, five pro-inflammatory genes were chosen that had been correlated to severity of influenza disease: CXCL10 (also known as interferon-cinducible protein 10, IP-10), interleukin 6 (IL6), interleukin 8, (IL8; also known as CXCL8), CCL2 (also known as monocyte chemoattractant protein 1, MCP-1), and CCL5 (also known as RANTES). A549 cells were infected with H5N1 A/Thailand/ 1(Kan-1)/04 or H5N1 A/Vietnam/1203/04 at MOI 0.01, 0.1, or 1. Glycyrrhizin treatment was performed with 25, 50, 100, or 200 mg/ml. Cytokine expression was detected 24 h post infection by ELISA. Glycyrrhizin did not affect cytokine expression of noninfected cells (data not shown) but inhibited expression of all cytokines investigated in H5N1-infected cells in a dose-dependent manner (Figure 2, Figure 3A ). Effects were more pronounced at lower MOIs. Notably, expression of all cytokines except IL8 was significantly inhibited after treatment with glycyrrhizin 50 mg/ml Figure 3A ) although these glycyrrhizin concentrations had no effect on H5N1 replication in A549 cells (Figure 1, Figure S1 ).
Cytokine expression by influenza A virus-infected respiratory cells causes recruitment of peripheral blood monocytes into the lungs of patients where they differentiate to macrophages which are thought to contribute to influenza A virus pathogenicity [5, 39] . In a chemotaxis assay, the influence of glycyrrhizin was investigated on migration of monocytes towards supernatants of H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.1)-infected A549 cells through 8 mm filters. Monocyte migration towards supernatants of H5N1-infected cells was strongly increased relative to migration towards supernatants of non-infected cells. Treatment of H5N1- infected cells with glycyrrhizin 100 mg/ml clearly suppressed chemoattraction activity of supernatants ( Figure 3B ).
Influenza viruses including H5N1 have been shown to induce caspase-dependent apoptosis in airway cells and this apoptosis has been correlated to the virus pathogenicity [40, 41] . Glycyrrhizin concentrations up to 200 mg/ml did not affect caspase activation in non-infected cells ( Figure 4A-C) . Glycyrrhizin concentrations $100 mg/ml inhibited H5N1 A/Thailand/1(Kan-1)/04 (MOI 0.01)-induced activation of the initiator caspases 8 and 9 as well as of the effector caspases 3/7 in A549 cells as determined 24 h post infection ( Figure 4A-C) . Lower glycyrrhizin concentrations did not affect H5N1-induced apoptosis. The detection of cells in sub-G1 phase resulted in similar findings ( Figure 4D ).
Substances that inhibit H5N1-induced caspase 3 activation including caspase 3 inhibitors cause nuclear retention of RNP complexes [34, 42] . In accordance, glycyrrhizin also interfered with nuclear export RNP at MOI 1 ( Figure S2 ). Similar results were obtained in MOI 0.01 H5N1 A/Thailand/1(Kan-1)/04infected cells ( Figure S3 ).
Influence of glycyrrhizin on H5N1-induced activation of nuclear factor kB (NFkB), p38, and on H5N1-induced cellular reactive oxygen species (ROS) formation Activation of NFkB, p38, and JNK have been associated with influenza A virus replication and virus-induced pro-inflammatory gene expression [34, [43] [44] [45] [46] [47] . While glycyrrhizin did not influence NFkB activity in non-infected A549 cells in the tested concentra-tions (data not shown), glycyrrhizin inhibited NFkB activation in H5N1-infected cells ( Figure 5A ). Moreover, glycyrrhizin inhibited H5N1-induced phosphorylation of the MAPKs p38 and JNK ( Figure 5B ).
In addition to their roles during influenza A virus replication and virus-induced cytokine/chemokine expression, NFkB, p38, and JNK are constituents of redox-sensitive signalling pathways [48] [49] [50] [51] . Antioxidants had been already found to interfere with influenza A virus-induced signalling through NFkB, p38, and JNK, with influenza A virus replication, and with influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] . Since glycyrrhizin is known to exert antioxidative effects [26] we speculated that glycyrrhizin may interfere with H5N1-induced ROS formation. Indeed glycyrrhizin exerted clear antioxidative effects in H5N1 (MOI 0.01)-infected cells ( Figure 5C ) causing significant reduction of ROS formation already at a concentration of 25 mg/ml ( Figure 5D ).
Here, we show that glycyrrhizin inhibits the replication of highly pathogenic H5N1 influenza A virus, H5N1-induced apoptosis, and H5N1-induced expression of pro-inflammatory cytokines in lung-derived A549 cells. After intravenous administration, achievable plasma concentrations of glycyrrhizin have been described to be about 100 mg/ml [52] . Therefore, the glycyrrhizin concentrations found to interfere with H5N1 replication and H5N1-induced pro-inflammatory gene expression in the present report are in the range of therapeutic plasma levels. Notably, although higher glycyrrhizin concentrations were needed to interfere with SARS coronavirus replication [22] than with H5N1 replication, beneficial results were reported in glycyrrhizin (SNMC)-treated SARS patients in comparison to SARS patients who did not receive glycyrrhizin [23] . Notably, investigation of different glycyrrhizin derivatives against SARS coronavirus led to the identification of compounds with enhanced antiviral activity [53] . Therefore, glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs.
Experimental results suggested that glycyrrhizin might be able to affect seasonal influenza A virus disease by antiviral and immunomodulatory effects [26, 27] . Mice were prevented from lethal H2N2 infection by glycyrrhizin although no influence on virus replication was detected. The mechanism was suggested to be induction of interferon-c in T-cells by glycyrrhizin [54] . Moreover, glycyrrhizin was shown to influence seasonal influenza A virus replication through interaction with the cell membrane [25, 28] . However, these effects were observed only in concentrations $200 mg/ml when glycyrrhizin was added during the virus adsorption period. Since glycyrrhizin addition during the adsorption period did not influence H5N1 replication in our experiments it appears not likely that membrane effects contribute to anti-H5N1 effects detected here in lower concentrations.
Our results rather suggest that glycyrrhizin interferes with H5N1-induced oxidative stress. Influenza A virus (including H5N1) infection induces ROS formation. Antioxidants were found to inhibit influenza A virus replication and influenza A virus-induced pro-inflammatory gene expression [32] [33] [34] and glycyrrhizin is known to exert antioxidative effects [26] . Here, glycyrrhizin interfered with H5N1-induced activation of NFkB, p38, and JNK representing redox-sensitive signalling events [48] [49] [50] [51] involved in influenza A virus replication and influenza A virusinduced cellular cytokine/chemokine production [34, [43] [44] [45] [46] 55] . Glycyrrhizin 50 mg/ml significantly reduced H5N1-induced activation of NFkB. In addition, glycyrrhizin concentrations as low as 25 mg/ml effectively interfered with H5N1-induced ROS formation and with phosphorylation of the redox-sensitive MAPKs p38 and JNK. In our model, activation of p38 appears to be critical for H5N1-associated redox signalling since p38 inhibition had been shown before to mimick effects of the antioxidant N-acetyl-cysteine (NAC) [34] . Interestingly and in contrast to glycyrrhizin, NAC failed to inhibit H5N1 replication or H5N1-induced cytokine/chemokine expression in therapeutically relevant concentrations.
Glycyrrhizin diminished H5N1-induced cellular cytokine/ chemokine production in concentrations (#50 mg/ml) that did not interfere with H5N1 replication although redox-sensitive signalling pathways have been described to be involved in both processes. Therefore, H5N1-induced proinflammatory gene expression appears to be more sensitive to inhibition of ROS formation than H5N1 replication. Indeed, influenza viruses had been shown to induce cellular pathways through replicationdependent and -independent events [56] . In a previous report, we could show that similar glycyrrhizin concentrations like those investigated here interfered with H5N1-induced pro-inflammatory gene expression but not with H5N1 replication in human monocyte-derived macrophages [57] . In addition, other immunomodulatory treatment regimens that did not influence H5N1 replication reduced mortality in H5N1-infected mice [31, 58] . Therefore, glycyrrhizin represents a potential additional treatment option that interfers with both H5N1 replication and H5N1induced expression of pro-inflammatory cytokines in lung cells.
Interference with immune responses may also result in the loss of control of virus replication by cytotoxic immune cells including natural killer cells and cytotoxic CD8 + T-lymphocytes. Global immunosuppressants like corticosteroids failed to protect from lethal influenza virus infection [59] . Moreover, antiviral drugs may interfere with cytotoxic cells that control virus replication as demonstrated for ribavirin that was shown to hamper NK cell cytolytic activity [60] . In this context, glycyrrhizin had already been shown not to affect natural killer cell activity in the concentrations used here [57] .
In conclusion, we show in this report that therapeutic concentrations of glycyrrhizin (used as clinically approved parenteral preparation SNMC) interfere with highly pathogenic H5N1 influenza A virus replication and H5N1-induced proinflammatory gene expression at least in part through interference with H5N1-induced ROS formation and in turn reduced activation of p38, JNK, and NFkB in lung cells. Since we used the clinical formulation SNMC effects of other ingredients like glycin or cystein cannot be excluded. Vaccines and antiviral agents will fail to meet global needs at least at the beginning of a severe influenza A virus pandemic [61] . Anti-inflammatory and immunomodulatory agents are considered to be important candidates as constituents of anti-influenza treatment strategies that may save lives in an influenza pandemic situation [61] . Therefore, glycyrrhizin may complement the arsenal of potential drugs for the treatment of H5N1-caused disease.
|
What is the conclusion of the study?
| 5,241 |
glycyrrhizin might also serve as lead structure for the development of novel anti-influenza drugs
| 18,948 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the mean duration of time from single lobe consolidation to bilateral multilobar lung infiltrates in human adenovirus type 55 (HAdV-55)?
| 3,240 |
2 days
| 1,760 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the mean duration of time from first positive chest x-ray to bilateral multilobar lung infiltrates in human adenovirus type 55 (HAdV-55)?
| 3,241 |
4.8 days
| 1,771 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What are the most frequent clinical manifestations of human adenovirus type 55 (HAdV-55) induced ARDS?
| 3,242 |
Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates
| 2,285 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What do we know about the genomics of human adenovirus type 55 (HAdV-55)?
| 3,243 |
This pathogen was fully characterized by whole-genome sequencing
| 3,601 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What are the clinical symptoms of human adenovirus type 55 (HAdV-55)?
| 3,244 |
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness
| 8,509 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the mean time from onset of symptoms to dyspnea in human adenovirus type 55 (HAdV-55)?
| 3,245 |
5 days
| 8,958 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the mean time of onset of symptoms to ICU admission in human adenovirus type 55 (HAdV-55)?
| 3,246 |
9.6 days
| 9,127 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the mean rate of respiration upon admission to the ICU when admitted for human adenovirus type 55 (HAdV-55)?
| 3,247 |
43 breaths per minute
| 9,244 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the white blood cell count in severe cases of human adenovirus type 55 (HAdV-55)?
| 3,248 |
low or in the normal range
| 9,441 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What does a chest x-ray look like for a patient with a severe case of human adenovirus type 55 (HAdV-55)?
| 3,249 |
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission
| 9,825 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What are the high resolution pulmonary CT scan findings for patients with severe cases of human adenovirus type 55 (HAdV-55)?
| 3,250 |
Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT
| 10,189 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
Where could a clinician acquire a positive viral sample in severe cases of human adenovirus type 55 (HAdV-55)?
| 3,251 |
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples
| 10,659 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
How long did it take for patients with positive human adenovirus type 55 (HAdV-55) endotracheal aspirates to develop a measurable viremia?
| 3,252 |
1 to 10 days
| 10,911 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
Does blood type play a role in the severity of human adenovirus type 55 (HAdV-55) infection?
| 3,253 |
HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B
| 12,906 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What are the most common clinical manifestations of severe human adenovirus type 55 (HAdV-55) induced ARDS?
| 3,254 |
Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations
| 13,068 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What is the mortality rate of severe ARDS from human adenovirus type 55 (HAdV-55)?
| 3,255 |
HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
| 13,514 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
What role does T-cell count play in severe human adenovirus type 55 (HAdV-55) infection?
| 3,256 |
a lower T-cell count may be a risk factor for HAdV-55 infection in young adults
| 15,873 |
1,604 |
Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4243941/
SHA: f5b706d0529bfcf7e2d1dfc037df5b6f95fc5ec0
Authors: Sun, Bing; He, Hangyong; Wang, Zheng; Qu, Jiuxin; Li, Xuyan; Ban, Chengjun; Wan, Jun; Cao, Bin; Tong, Zhaohui; Wang, Chen
Date: 2014-08-12
DOI: 10.1186/s13054-014-0456-6
License: cc-by
Abstract: INTRODUCTION: Since 2008, severe cases of emerging human adenovirus type 55 (HAdV-55) in immunocompetent adults have been reported sporadically in China. The clinical features and outcomes of the most critically ill patients with severe acute respiratory distress syndrome (ARDS) caused by HAdV-55 requiring invasive mechanical ventilation (IMV) and/or extracorporeal membrane oxygenation (ECMO) are lacking. METHODS: We conducted a prospective, single-center observational study of pneumonia with ARDS in immunocompetent adults admitted to our respiratory ICU. We prospectively collected and analyzed clinical, laboratory, radiological characteristics, sequential tests of viral load in respiratory tract and blood, treatments and outcomes. RESULTS: The results for a total of five consecutive patients with severe ARDS with confirmed HAdV-55 infection were included. All five patients were immunocompetent young men with a median age of 32 years. The mean time from onset to dyspnea was 5 days. Arterial blood gas analysis at ICU admission revealed profound hypoxia. Mean partial oxygen pressure/fraction of inspired oxygen was 58.1. Mean durations from onset to a single-lobe consolidation shown on chest X-rays (CXRs) and, from the first positive CXR to bilateral multilobar lung infiltrates, were 2 days and 4.8 days, respectively. The viral load was higher than 1 × 10(8) copies in three patients and was 1 × 10(4) in one patient. It was negative in the only patient who survived. The mean duration for noninvasive positive pressure ventilation (NPPV) failure and IMV failure were 30.8 hours and 6.2 days, respectively. Four patients received venovenous ECMO. Four (80%) of the five patients died despite receiving appropriate respiratory support. CONCLUSIONS: HAdV-55 may cause severe ARDS in immunocompetent young men. Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates, are the most frequent clinical manifestations of HAdV-55-induced severe ARDS. Viral load monitoring may help predict disease severity and outcome. The NPPV and IMV failure rates were very high, but ECMO may still be the respiratory support therapy of choice. TRIAL REGISTRATION: Clinicaltrials.gov NCT01585922. Registered 20 April 2012
Text: Human adenoviruses (HAdVs) are notorious pathogens in people with compromised immune function and a frequent cause of outbreaks of acute respiratory disease among young children. Life-threatening adenoviral pneumonia has previously been documented among military trainees, patients with AIDS and transplant recipients [1] [2] [3] [4] [5] . Human adenovirus type 55 (HAdV-55), which is emerging as a highly virulent pathogen for acute fatal adenoviral pneumonia among immunocompetent adults in China, has gained increasing attention [6] . HAdV-55 is a newly identified, emergent acute respiratory disease pathogen causing two recent outbreaks in China in 2006 [7] and in Singapore in 2005 [8] . In 2011, this pathogen apparently re-emerged in Beijing, China, causing several cases of severe community-acquired pneumonia [9] . This pathogen was fully characterized by whole-genome sequencing [10] . Comparative studies showed that the ability of HAdV to cause severe disease may relate to the serotypes of HAdVs. Severe adenoviral pneumonia induced by HAdV-55 has been reported to be more closely related to severe cases compared to other serotypes (HAdV-3, HAdV-7 and HAdV-14) [6] .
Current knowledge of HAdV-55-induced severe acute respiratory distress syndrome (ARDS) requiring invasive mechanical ventilation and/or extracorporeal membrane oxygenation (ECMO) support in immunocompetent adults is derived from single case reports or relatively small, single-center series. As a result, little information is available on HAdV-55 pneumonia complicated with severe ARDS, the frequency of which is expected to increase in the coming years. Here we describe the clinical features and outcomes of five prospective cases of HAdV-55 pneumonia complicated with severe ARDS in immunocompetent adults in our ICU.
Beginning in May 2012, a randomized trial of noninvasive positive pressure ventilation (NPPV) in ARDS patients was carried out in our center (ClinicalTrials.gov ID: NCT01585922). From May 2012 to April 2014, all adult patients with ARDS caused by pneumonia who were admitted to the respiratory ICU of Beijing Chao-Yang Hospital were prospectively enrolled. Severe ARDS was diagnosed according to the Berlin definition: (1) developing within 1 week of a known clinical insult or new or worsening respiratory symptoms; (2) bilateral opacities not fully explained by effusions, lobar and/or lung collapse, or nodules; (3) respiratory failure not fully explained by cardiac failure or fluid overload; (4) partial oxygen pressure/ fraction of inspired oxygen (PaO 2 /FiO 2 ) ≤100 mmHg with positive end-expiratory pressure (PEEP) ≥5 cmH 2 O; and (5) a chest radiograph with three or four quadrants with opacities. Patients with HAdV-55 infection and severe ARDS who failed conventional NPPV and invasive mechanical ventilation (IMV) were included in the analysis. This study was approved by the Institutional Review Board of Beijing Chao-Yang Hospital (LLKYPJ2012031). Data were analyzed anonymously. Each patient gave written informed consent for their data to be used for research and publication.
Clinical information collected by investigators with a standardized data form included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms (fever, cough, sputum, dyspnea, chest pain, rash, nausea, vomiting, abdominal pain, diarrhea and headache), signs (body temperature, heart rate, respiratory frequency, blood pressure and crackles in the lungs), laboratory tests (whole-blood cell count and blood chemistry) and microbiological findings and images of the lung (chest X-ray (CXR) and computed tomography). Concomitant medications, respiratory support, complications and outcomes were also recorded.
Patients' specimens, including sputum, whole blood and serum samples, were collected upon admission and during hospitalization. Microbiological tests were performed at the Department of Infectious Disease and Clinical Microbiology in our center, and the detection methods used were described in our previous report [6] . Common viruses causing respiratory illness were screened using a kit with 15 different viral assays. Serum samples were used for Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila antibodies. All patients had their HAdV-55 infection confirmed by RT-PCR assay. Partial sequences of the hexon gene were analyzed to type the phylogeny of HAdV-55 strains. The adenoviral load was also performed on both respiratory specimens and blood by multiplex RT-PCR assay.
Viral pneumonia was diagnosed based on the presence of HAdV detected in sputum or throat swab samples by molecular methods.
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range).
During the study period, a total of eight patients diagnosed with HAdV infection and respiratory failure were admitted to our ICU, and seven of them received a diagnosis of ARDS. Five consecutive patients with severe ARDS with confirmed HAdV-55 infection were admitted to our ICU between April and July 2013. They were included in the analysis. The other two patients had mild ARDS and were infected with other types of HAdVs.
All five patients were immunocompetent young men with a median age of 32 years (range, 28 to 40 years). All of the patients shared a B blood type and came from the same city: Baoding city, Hebei province, northern China. All patients had no exposure to farm animals, corn or hay. Patient 3 had tuberculosis pleuritis and received antituberculosis therapy at ICU admission. His blood tests, including the T-SPOT tuberculosis assay (Oxford Immunotec, Marlborough, MA, USA) and antibody of Mycobacterium tuberculosis, were negative.
Flulike symptoms, such as fever, cough and little sputum, were commonly observed at the onset of illness. All patients presented with a high fever, with a mean body temperature of 39.5°C (range, 39.0°C to 40.0°C), which persisted for 8 days (range, 6 to 11 days). Productive cough was observed in two patients. Dull substernal chest pain and rash were also observed in two patients. All patients had dyspnea. The mean time from onset to dyspnea was 5 days (range, 1 to 10 days). After the onset of dyspnea, patients usually progressed to respiratory failure or hypoxemia. The mean time from onset to ICU admission was 9.6 days (range, 8 to 11 days) ( Table 1) . All patients had tachypnea when admitted to the ICU, with a mean rate of 43 breaths per minute (range = 38 to 52). Arterial blood gas analysis at ICU admission revealed profound hypoxia, with a mean PaO 2 /FiO 2 of 58.1 (range = 49 to 62.5). White blood cell counts were low or in the normal range. All patients had elevated serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBDH) ( Table 1) . At admission, all patients' levels of immunoglobulin (serum immunoglobulins G and M) and components C3 and C4 were in the normal range.
Four patients had lower than normal T-cell subset counts (Table 2) .
CXRs revealed multiple bilateral lobar or segment consolidation in the lungs of all five patients, and radiographic lesions progressed rapidly after ICU admission ( Figure 1 ). Three patients were examined by highresolution computed tomography (HRCT). Unilateral or bilateral consolidations and infiltrates were found on HRCT scans of all three of these patients. Consolidations within a single lobe or several lobes with a clear border and air bronchogram were the most common findings on HRCT scans. Nodules, patches, pleural effusion, abscess and a cavity were also seen visualized by HRCT (Figure 2 ). The mean duration from onset to a single-lobe consolidation on CXRs was 2 days (range = 1 to 5 days). The mean duration from the first positive CXR to bilaterally multilobar lung infiltrates was 4.8 days (range = 4 to 7 days).
All patients had HAdV-55 viremia. In four of the five patients, it was first detected in endotracheal aspirate (ETA) samples. The time between initial ETA sample collection of adenoviruses and positive results for HAdV-55 nucleic acid in the blood was 1 to 10 days (Table 3) . Virus DNA copies in ETAs were determined for all patients during their ICU stay. The viral load was higher than 1 × 10 8 copies in three patients and 1 × 10 4 in one patient. The viral load became negative in the only patient who survived. In the four patients who did not survive, DNA copies did not decrease, even with antiviral therapy (Figure 3 ).
Oxygenation was not maintained with conventional NPPV or IMV support in any of the patients. The mean duration until NPPV failure was 30.8 hours (range = 22 to 48 hours), and the mean time until IMV failure was 6.2 days (range 2 = to 13 days) ( Table 1) . Four patients received venovenous ECMO to maintain oxygen saturation, and one patient refused ECMO support and received high-frequency oscillatory ventilation instead. Table 4 gives the oxygenation data of patients before and after venovenous ECMO support.
All patients received antiviral therapy, including acyclovir (10 mg/kg, every 8 hours, intravenous drip), ganciclovir (5 mg/kg, every 12 hours, intravenous drip) and ribavirin (250 mg, twice daily, intravenous drip). Considering that bacterial coinfection may combine with a severe viral infection, broad-spectrum intravenous antibiotics were given to all patients. Tests for bacterial pathogens were negative for only one patient (Table 3) . Four (80%) of the five patients died. Among the four patients receiving venovenous ECMO, only one patient survived. The other four patients died due to ARDS, Aspergillus fumigatus coinfection, septic shock and catheter-related bloodstream infection due to Acinetobacter baumannii, respectively.
To the best of our knowledge, this is the first cohort observational study on the clinical characteristics of patients with severe ARDS caused by emergent HAdV-55 infection and also the first on the evaluation of a viral load test for monitoring the reaction to therapy and for prediction of patient outcome. The following are the main findings of this study. (1) HAdV-55 may cause severe ARDS in immunocompetent young men with blood type B. All of our patients were from the same city of Hebei province, northern China. (2) Persistent high fever, dyspnea and rapid progression to respiratory failure within 2 weeks, together with bilateral consolidations and infiltrates at the same time, are the most frequent clinical manifestations of severe HAdV-55induced ARDS. (3) Viral load monitoring may help predict disease severity and patient outcome. (4) The NPPV and IMV failure rates were very high, and ECMO may be the last support method for this group of patients. (5) HAdV-55-induced severe ARDS has a very high mortality rate (80%) despite appropriate respiratory support.
Sporadic severe adenoviral infection in healthy adults has historically been described for serotype 4 [11] , serotype 7 [4, 12] and, more recently, serotype 14 in the general population and in military trainees [13, 14] . HAdV-55 was first completely characterized in Shaanxi, China [7] and then reemerged in Hebei, a province close to Beijing, where it caused several cases of acute respiratory disease [9] . It was presumed that HAdV-55 was a recombinant form of the B2 species of HAdV-14 and HAdV-11 [7, 15] due to its sharing a hexon gene with the HAdV-11 and HAdV-14 chassis [16] . The results of our study show that HAdV-55, as an emerging pathogen among immunocompetent adults, may cause severe ARDS.
The prevalence of severe fatal adenoviral pneumonia induced by HAdV-55 in our study is somewhat similar to that described by Cao and colleagues [6] . All cases of reported HAdV-55 in our study were from the same city: Baoding, Hebei province, northern China. They occurred between April and July 2013, just partly overlapping or following the influenza epidemic. The patients with severe disease also came from the same region and were treated during a similar time period, which suggests that HAdV-55 may be an important viral pathogen derived from this region.
Our study results suggest that the following may be clinical features of ARDS caused by HAdV-55: persistent high fever, rapid progression of dyspnea, need for mechanical ventilation support, elevated AST level and rapid progression from unilateral infiltrates to bilateral consolidations. These clinical features are highly similar to those of ARDS caused by other types of HAdV described in previous reports [6, 9] .
Recent studies have shown that the immune system plays a crucial role in the clearance of HAdV viremia and survival of the host [17] . Chen et al. reported that, in the acute phase of HAdV-55 infection, patients with severe disease may have high levels of dendritic cells and Th17 cells [18] . In our study, the only patient who recovered from severe infection had higher T-cell counts. Three of the five patients had relatively low T-cell counts when admitted. Our results suggest that these three patients may have been relatively immunocompromised and that a lower T-cell count may be a risk factor for HAdV-55 infection in young adults. HAdV-55 DNA was previously reported in 41.2% of patients with severe infection [18] . In our study, HAdV-55 DNA was detected and monitored in all patients with severe ARDS. The initial, and trend of, viral load that presented as HAdV-55 DNA copies in the respiratory tract samples and blood may suggest the severity of infection and may predict both the reaction to therapy and patient outcome.
The use of mechanical ventilation and ECMO in patients with ARDS caused by HAdV-55 has not been detailed in previous studies. In our cohort, we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV. This failure rate may be a result of the large area of consolidation that induced a severe shunt in the lung, which may lead to lack of response to positive pressure ventilation. For patients with severe ARDS, ECMO should be considered a better choice for oxygenation.
Our study has limitations. It is an observational study with no comparison group, so the difference between the severe and modest infections could not be clarified in terms of immune status, clinical features, radiological findings, viral load and treatment effects on respiratory support and antiviral therapy. Sequential dynamic analysis is needed to determine the relationship between HAdV-55 viremia and treatment response.
|
How successful are the use of invasive mechanical ventilation (IMV) and non-invasive positive pressure ventilation (NPPV) in the treatment of severe ARDS from human adenovirus type 55 infection?
| 3,257 |
we found that severe HAdV-55 infection could cause a rapid progression of respiratory failure, with a very high failure rate for NPPV and IMV
| 16,491 |
1,599 |
Diagnostic accuracy of C-reactive protein and procalcitonin in suspected community-acquired pneumonia adults visiting emergency department and having a systematic thoracic CT scan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4608327/
SHA: f3d150545162ff3cc253c235011a02a91ee676cb
Authors: Le Bel, Josselin; Hausfater, Pierre; Chenevier-Gobeaux, Camille; Blanc, François-Xavier; Benjoar, Mikhael; Ficko, Cécile; Ray, Patrick; Choquet, Christophe; Duval, Xavier; Claessens, Yann-Erick
Date: 2015-10-16
DOI: 10.1186/s13054-015-1083-6
License: cc-by
Abstract: INTRODUCTION: Community-acquired pneumonia (CAP) requires prompt treatment, but its diagnosis is complex. Improvement of bacterial CAP diagnosis by biomarkers has been evaluated using chest X-ray infiltrate as the CAP gold standard, producing conflicting results. We analyzed the diagnostic accuracy of biomarkers in suspected CAP adults visiting emergency departments for whom CAP diagnosis was established by an adjudication committee which founded its judgment on a systematic multidetector thoracic CT scan. METHODS: In an ancillary study of a multi-center prospective study evaluating the impact of systematic thoracic CT scan on CAP diagnosis, sensitivity and specificity of C-reactive protein (CRP) and procalcitonin (PCT) were evaluated. Systematic nasopharyngeal multiplex respiratory virus PCR was performed at inclusion. An adjudication committee classified CAP diagnostic probability on a 4-level Likert scale, based on all available data. RESULTS: Two hundred patients with suspected CAP were analyzed. The adjudication committee classified 98 patients (49.0 %) as definite CAP, 8 (4.0 %) as probable, 23 (11.5 %) as possible and excluded in 71 (35.5 %, including 29 patients with pulmonary infiltrates on chest X-ray). Among patients with radiological pulmonary infiltrate, 23 % were finally classified as excluded. Viruses were identified by PCR in 29 % of patients classified as definite. Area under the curve was 0.787 [95 % confidence interval (95 % CI), 0.717 to 0.857] for CRP and 0.655 (95 % CI, 0.570 to 0.739) for PCT to detect definite CAP. CRP threshold at 50 mg/L resulted in a positive predictive value of 0.76 and a negative predictive value of 0.75. No PCT cut-off resulted in satisfactory positive or negative predictive values. CRP and PCT accuracy was not improved by exclusion of the 25 (25.5 %) definite viral CAP cases. CONCLUSIONS: For patients with suspected CAP visiting emergency departments, diagnostic accuracy of CRP and PCT are insufficient to confirm the CAP diagnosis established using a gold standard that includes thoracic CT scan. Diagnostic accuracy of these biomarkers is also insufficient to distinguish bacterial CAP from viral CAP. TRIAL REGISTRATION: ClinicalTrials.gov registry NCT01574066 (February 7, 2012) ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13054-015-1083-6) contains supplementary material, which is available to authorized users.
Text: Community-acquired pneumonia (CAP) is a frequently seen disease, with high morbidity and mortality, accounting for 600,000 hospitalizations each year. It represents the seventh leading cause of death in the USA [1] . CAP prognosis depends on the rapidity of specific treatment, which should ideally be initiated within four hours and no later than eight hours after diagnosis [2, 3] . CAP diagnosis is based on the clustering of non-specific pulmonary and general symptoms [4, 5] , an increase in biomarkers reflecting systemic inflammatory response syndrome (SIRS), and the presence of new parenchymal infiltrates on chest X-ray. However, CAP diagnosis remains uncertain in many cases with alternative diagnoses, such as cardiac failure, acute bronchitis, chronic obstructive pulmonary disease (COPD) exacerbations, pulmonary embolism, neoplasia, and sepsis [6, 7] .
Part of the uncertainty of CAP diagnosis may be due to the high rate of chest X-ray misdiagnosis [8, 9] ; over diagnosis of CAP is frequent when infiltrates of noninfectious origin coexist with pulmonary or general symptoms, and the diagnosis of CAP is often ignored when the lung infiltrates are at the limit of visibility or are hidden due to superposition [10] . We recently published a study in which thoracic CT scan was systematically performed in a population of clinically suspected CAP patients visiting the emergency department for CAP (the ESCAPED study) [11] . We showed that CAP diagnosis based on chest X-ray led to a false CAP diagnosis in many patients: among CAP suspected patients with radiological pulmonary infiltrate, CAP diagnosis was excluded in around 30 % of patients based on CT scan results; on the contrary, among patients without radiological pulmonary infiltrate, one-third had a pulmonary infiltrate on thoracic CT-scan. We also reported the isolation of viruses in one-third of patients [11, 12] .
Several attempts have been made to improve CAP diagnosis based on biomarkers, such as C-reactive protein (CRP) and procalcitonin (PCT); however, there are conflicting data on their reliability [13] [14] [15] [16] [17] . This could be due to the consideration of CAP diagnosis based on chest X-ray as establishing pulmonary infection. In the present study, we aimed to analyze CRP and PCT values in the population of the ESCAPED study reported above for whom CAP diagnosis was established by an adjudication committee which founded its judgment on all usual available data, systematic multidetector thoracic CT scan performed at inclusion, and results from a day-28 follow-up. We also analyzed whether the viral etiology of definite CAP based on polymerase chain reaction (PCR) multiplex naso-pharyngeal swab interfered with the accuracy of the biomarkers.
Setting ESCAPED was a multicenter, prospective, interventional study, entitled "Early Thoracic CT-Scan for Community-Acquired Pneumonia at the Emergency Department (ESCAPED)" [11] , conducted from November 2011 to January 2013, in four emergency departments (EDs) of four tertiary teaching hospitals in Paris, France, designed to measure the impact of thoracic CT scan on clinical decision. The study was sponsored and monitored by the Paris public health hospitals, and funded by the French Ministry of Health. The French health authorities (Agence nationale de sécurité des medicaments et produits de santé, ANSM) and the institutional review board for the protection of human subjects approved the study protocol and patient informed consent procedures. All enrolled patients provided written informed consent for inclusion. The protocol was registered in the clinicaltrial.gov website under the PACSCAN acronym, the French translation of the English ESCAPED acronym (NCT01574066). The Ethics Committee of Ile de France (Comité de Protection des Personnes. Paris N°2 011-oct-12749) approved the study protocol.
The primary objective was to compare CRP and PCT values in the four different categories of CAP level of certainty using the day-28 adjudication committee classification. The four categories were: 1) absence of CAP hereafter referred to as excluded CAP diagnosis; 2) possible CAP; 3) probable CAP; and 4) definite CAP. The secondary objectives were to assess whether CRP and PCT were associated with CAP diagnosis using sensitivity analyses in three successive subgroups chosen a priori; 1) when specifically considering patients classified as having excluded CAP diagnosis and definite CAP (i.e., the patients for whom the level of certainty was the highest); 2) when patients with excluded CAP diagnosis and diagnosed extra-pulmonary infectious disease (which may increase biomarker values) were not taken into account, in the excluded CAP group; and 3) when patients classified as viral CAP were not taken into account in the definite CAP group, as PCT has been reported to be lower in viral infections as compared to bacterial infections [18] .
Consecutive adults ( [19] .
Multidetector thoracic CT-scan was performed after chest X-ray, ideally within the four hours following inclusion. Chest X-ray and thoracic CT-scan were performed using a standardized protocol. The four levels of CAP probability according to CT scan were defined as definite (systematic alveolar condensation, alveolar condensation with peripheral and localized ground glass opacities, bronchiolar focal or multifocal micronodules), probable (peripheral alveolar condensation, retractile systematic alveolar condensation, or diffuse ground glass opacities), possible (pulmonary infarct), or excluded (pulmonary mass, other abnormalities, or normal images). Scan views were recorded on a DVD.
Based on data collected from baseline standardized case report forms, DVD recorded pictures of X-ray and CTscan, and blinded to local interpretations, an adjudication committee consisting of three independent senior experts in infectious diseases, pneumology and radiology retrospectively assigned the probability of CAP diagnosis using the same 4-level Likert scale, with all available data including patients' discharge summary, and follow-up data obtained by assistant investigators who contacted by phone either the patient, relatives or general practitioners at day 28. For this study, the gold standard of CAP was the diagnosis assessed by this adjudication committee. Alternative diagnoses were established for excluded CAP and classified as non-CAP pulmonary diseases and extra-pulmonary infectious diseases and others.
Blood samples were collected at inclusion in sodium heparin-treated tubes, centrifuged, and stored at −40°C until completion of the study. CRP and PCT concentrations were measured a posteriori on plasma collection (see Additional file 1 for methodology), except for patients in whom marker dosage was performed by the emergency practitioner on his own initiative.
Naso-pharyngeal swabs were collected at enrollment and placed in a Middle Virocult MWE (Sigma®) transport medium. Samples were kept at room temperature and sent to the virology laboratory of Bichat -Claude Bernard Hospital (Paris) as soon as possible after collection. The samples were not frozen and thawed. Multiplex PCR (RespiFinder-19 assay (Pathofinder®, Maastricht, Netherlands)) was performed on naso-pharyngeal swabs to detect 15 respiratory viruses -coronavirus 229E, NL63, OC43, human metapneumovirus (hMPV), influenza A, A (H1N1) pdm2009 and B viruses, parainfluenza viruses 1, 2, 3, and 4, respiratory syncytial virus (RSV) A and B, rhinovirus, adenovirus, and 4 intracellular bacteria -Bordetella pertussis, Chlamydophila pneumoniae, Legionella pneumophila, Mycoplasma pneumoniae, in one reaction. The multiplex PCR results were not available to the adjudication committee. Routine microbiological examinations were also performed at the discretion of the emergency physicians and included blood culture, sputum culture, and antigenuria (see Additional file 1 for methodology). CAP, classified as definite, was considered as being of viral origin when multiplex PCR was positive for at least one of the 15 respiratory viruses and no bacteria were found using PCR and routine bacterial microbiological samples (sputum, blood culture, antigenuria) when performed.
Baseline and follow-up characteristics were described by means and standard deviations (SD) or by median and interquartile range (IQR) for continuous variables normally distributed or with skewed distribution, respectively, and by percentages for categorical variables, for the total study population and for the study groups. We performed chi-square or Fisher exact tests when appropriate for qualitative variables, and the Student or Mann-Whitney tests for continuous variables with skewed distributions to compare baseline patient characteristics and study outcomes between study groups.
The distribution values of the biomarkers were determined in the different populations of patients using boxplots. The performances of CRP and PCT in predicting definite CAP were evaluated by sensitivity analysis (definite CAP vs excluded CAP). CRP was evaluated at several cut-off points of 20 mg/L, 30 mg/L, 50 mg/L, 70 mg/L, and 100 mg/L, values used in previous studies [15, 20, 21] . Several cut-off points for PCT were chosen at the level of 0.10 μg/L [18] , and at the two levels for suspected bacterial infection as stated by the manufacturer, i.e., 0.25 μg/L and 0.50 μg/L. Sensitivities, specificities, positive predictive values (PPVs), negative predictive values (NPVs), and likelihood ratio were calculated. Receiver operating characteristic (ROC) curves were drawn, area under the curve AUC was computed and optimal cut-off was identified by the maximization of the Youden's index, comparing biomarker values in patients with excluded CAP and definite CAP. From these optimal cut-offs for CRP and PCT, sensitivity analyses were performed combining the CRP and PCT cut-offs.
A multivariate logistic regression model was built to identify factors associated with having definite CAP as compared to having an excluded CAP diagnosis. We excluded from the excluded CAP diagnosis group, patients with an extra-pulmonary infectious disease. All variables with a p value of < 0.25 in the bivariate analysis were entered into a multivariate logistic regression with a backward stepwise approach; the discrimination was evaluated by the C-index and its 95 % confidence interval (95 % CI) and the calibration was evaluated by the Hosmer Lemeshow goodness-of-fit test.
All tests were two-sided, and p-values below 0.05 were considered to denote statistical significance. All statistical analyses were performed using SPSS statistical software version 21.0 (SPSS Inc., Chicago, IL, USA).
Two hundred patients with suspected CAP out of the 319 in the ESCAPED study were included in the present study, for which CRP and PCT assays and nasopharyngeal swab for multiplex PCR were available (Fig. 1) . Characteristics of the 200 patients (age, age more than 65, gender, probability of CAP diagnosis by adjudication committee) were not significantly different from those of the 119 other patients of the ESCAPED study and are summarized in Table 1 . CRP and PCT assays were performed based on the emergency practitioner's own initiative in 70 patients for CRP and 131 for PCT, or performed a posteriori on plasma samples of the remaining patients. Sex ratio was approximately 1. More than half of the patients (54 %) were 65 years of age or older. The
Pulmonary infiltrates were seen on chest X-ray in 127 (63.5 %) patients. Thoracic CT-scan excluded a CAP diagnosis in 16.5 % of these 127 patients; on the contrary, thoracic CT-scan revealed a parenchymal infiltrate in 27 % of the 73 patients without infiltrate on chest X-ray.
Based on all available data including multidetector CT scan results (but excluding PCR results), the adjudication
The CRP and PCT distributions in the 200 patients are presented in Fig. 2 A statistically significant difference between the two groups (excluded CAP vs definite CAP) was demonstrated for several cut-off points for CRP and PCT ( Table 2 ). For CRP, the value of 50 mg/L resulted in a PPV of 0.76 and a NPV of 0.75. For PCT, no value resulted in a satisfactory PPV or NPV. For these two biochemical markers, the ability to predict CAP was evaluated by a ROC curve. The AUC was 0.787 (95 % CI 0.717-0.857), optimal cut-off = 45.9 mg/L for CRP (Fig. 3 ) and 0.655 (95 % CI 0.570-0.739), optimal cut-off = 0.13 μg/ L for PCT (Fig. 4) .
Sensitivity analyses for the combination of CRP and PCT, using these optimal cut-offs, resulted in a PPV of 0.74 and a NPV of 0.58. Use of the other PCT cut-offs did not result in better PPV or NPV ( Table 2) .
The present study is novel as patients prospectively benefited from extensive investigation to determine the diagnosis of CAP in the ED, including both early multidetector thoracic CT-scan and day-28 adjudication committee. This led to the correction of CAP diagnosis previously based on chest X-ray in a high number of patients. In these extensively characterized patients, both CRP and PCT lacked operational precision to allow the decisionmaking process to rule out or confirm diagnosis of CAP even in selected subgroups.
The clinical characteristics of the patients included in this sub-study are consistent with those in the current literature. As previously reported, patients frequently had a history of respiratory disorders, cancer and congestive heart failure [21, 22] . The design of the ESCAPED study required exclusion of patients within the highest CRB 65 categories, which limited the inclusion of patients older than 65. This may explain why the mean age of our patients (64 years) falls within the lower values of those reported elsewhere [19] . Data to identify the microbial agent responsible for the disease were collected by the usual techniques and multiplex PCR. Viral identification using naso-pharyngeal PCR that revealed viral respiratory infection in approximately one-third of cases was concordant with values reported in the literature [23] . Therefore, we believe that our results can be extrapolated to most emergency patients suffering from CAP.
In the present study, patients were recruited on the basis of initial clinical assessment for the diagnosis of CAP. Therefore, we believe that the characteristics of the patients closely correspond to those that lead practitioners to consider a possible diagnosis of CAP. In these patients, the design of our study allowed us to confirm or refute CAP diagnosis with a high level of certainty. Results confirmed the poor predictive value of clinical symptoms (new onset of systemic features and symptoms of an acute lower respiratory tract illness) in identifying CAP patients [21] . Indeed, clinical presentation of excluded CAP patients was similar to that of definite CAP patients except for fever and cough that were more frequent in definite CAP patients. Furthermore, the design also revealed that the combination of clinical symptoms and chest X-ray results led to CAP misdiagnosis in a high number of patients, including the 98 whose CAP diagnosis was excluded by the adjudication committee and who would have been considered as possible, probable or definite CAP without the use of the CT scan. This low specificity of clinical-standard radiological evaluation led to the consideration of either non-infectious pulmonary diseases (such as, cardiac failure, pulmonary embolism, pulmonary neoplasia or bronchitis) or extra-pulmonary infectious diseases as CAP. Of note, some of these diseases are also associated with increased biomarker values. This raises concerns about previous evaluations of biomarkers in CAP-suspected patients, which used clinical and standard radiological (chest X-ray) evaluations as the gold standard for CAP diagnosis [15] . The use of biomarkers has been advocated to improve diagnosis and management of patients with lower respiratory tract infections [14] . However, this issue is still unresolved [24] , with conflicting positions [14, 15, 25, 26] . In our study, while median values of both biomarkers did increase with level of certainty for CAP diagnosis, we were unable to establish discriminating values for PCT. Recent data suggested that CRP could be of more help in assisting in the diagnosis of lower respiratory tract infections (LRTI) [15, 27, 28] . In our study, although CRP seems more discriminating than PCT, neither the experimental exclusion of extra-pulmonary bacterial infections from the excluded CAP group, nor the exclusion of viral CAP from the definite CAP patients group, made possible the determination of a discriminant cutoff. The combination of CRP and PCT was not more discriminating than each biomarker separately. An operational algorithm has been released to assist physicians in prescribing antimicrobial therapy [14, 26, 29] . According to this strategy, a PCT concentration higher than 0.25 μg/L should prompt administration of antibiotics to patients with suspected LRTI. In our study, this value was associated with poor performance. Additionally, mean PCT levels remained above this threshold both in excluded CAP patients without infectious disorders and in definite CAP presumably related to virus. Therefore, the gold standard for the diagnosis of CAP may influence the performance and utility of PCT in this setting.
This study has some limitations. First, the adjudication committee was not blinded to the value of biomarkers measured at bedside in some patients (70 for CRP and 131 for PCT) and its CAP classification could thus have been influenced by these results. However, the lack of statistically significant differences in the mean CRP and PCT values in the definite CAP cases, whether or not these biomarkers were available for the adjudication committee, argues against a major impact of these results on adjudication committee classification. Second, another critical point is the prescription of antibiotic therapy (34 %) previous to inclusion. We cannot exclude that these previously-treated CAP patients may have altered biomarker performance and reduced the yield of bacterial cultures, although such a population reflects the usual emergency department practice. Third, multiplex PCR was performed on naso-pharyngeal sampling and not on lower respiratory tract samples, which does not allow definite confirmation of the viral origin of CAP. However, a recent large study on CAP patients which reported a viral etiology of CAP at a comparable rate, did not find upper respiratory tract shedding in a control population without CAP explored during the same year and season [30] . Finally, even if multidetector thoracic CT scan is a better imaging examination than X-ray to explore the chest, only invasive local microbiological samples would have provided a diagnosis with certainty.
Given the diversity of the clinical and radiological CAP presentations, CAP diagnosis is often uncertain. In our population of patients treated in the emergency room with clinical symptoms evoking CAP, neither CRP nor PCT cut-off values carried sufficient weight to confirm or refute CAP diagnosis at bedside; this underlines that these biomarkers are telltales of the host inflammatory response to the intrusion of microorganisms independent of the site of infection. These results, based on a systematic thoracic CT scan evaluation of CAP-suspected patients, do not argue for the use of CRP and PCT in routine care to diagnose CAP with certainty in patients visiting the ED for suspected CAP.
|
How many patients were analyzed in the study?
| 5,252 |
Two hundred
| 1,521 |
1,599 |
Diagnostic accuracy of C-reactive protein and procalcitonin in suspected community-acquired pneumonia adults visiting emergency department and having a systematic thoracic CT scan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4608327/
SHA: f3d150545162ff3cc253c235011a02a91ee676cb
Authors: Le Bel, Josselin; Hausfater, Pierre; Chenevier-Gobeaux, Camille; Blanc, François-Xavier; Benjoar, Mikhael; Ficko, Cécile; Ray, Patrick; Choquet, Christophe; Duval, Xavier; Claessens, Yann-Erick
Date: 2015-10-16
DOI: 10.1186/s13054-015-1083-6
License: cc-by
Abstract: INTRODUCTION: Community-acquired pneumonia (CAP) requires prompt treatment, but its diagnosis is complex. Improvement of bacterial CAP diagnosis by biomarkers has been evaluated using chest X-ray infiltrate as the CAP gold standard, producing conflicting results. We analyzed the diagnostic accuracy of biomarkers in suspected CAP adults visiting emergency departments for whom CAP diagnosis was established by an adjudication committee which founded its judgment on a systematic multidetector thoracic CT scan. METHODS: In an ancillary study of a multi-center prospective study evaluating the impact of systematic thoracic CT scan on CAP diagnosis, sensitivity and specificity of C-reactive protein (CRP) and procalcitonin (PCT) were evaluated. Systematic nasopharyngeal multiplex respiratory virus PCR was performed at inclusion. An adjudication committee classified CAP diagnostic probability on a 4-level Likert scale, based on all available data. RESULTS: Two hundred patients with suspected CAP were analyzed. The adjudication committee classified 98 patients (49.0 %) as definite CAP, 8 (4.0 %) as probable, 23 (11.5 %) as possible and excluded in 71 (35.5 %, including 29 patients with pulmonary infiltrates on chest X-ray). Among patients with radiological pulmonary infiltrate, 23 % were finally classified as excluded. Viruses were identified by PCR in 29 % of patients classified as definite. Area under the curve was 0.787 [95 % confidence interval (95 % CI), 0.717 to 0.857] for CRP and 0.655 (95 % CI, 0.570 to 0.739) for PCT to detect definite CAP. CRP threshold at 50 mg/L resulted in a positive predictive value of 0.76 and a negative predictive value of 0.75. No PCT cut-off resulted in satisfactory positive or negative predictive values. CRP and PCT accuracy was not improved by exclusion of the 25 (25.5 %) definite viral CAP cases. CONCLUSIONS: For patients with suspected CAP visiting emergency departments, diagnostic accuracy of CRP and PCT are insufficient to confirm the CAP diagnosis established using a gold standard that includes thoracic CT scan. Diagnostic accuracy of these biomarkers is also insufficient to distinguish bacterial CAP from viral CAP. TRIAL REGISTRATION: ClinicalTrials.gov registry NCT01574066 (February 7, 2012) ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13054-015-1083-6) contains supplementary material, which is available to authorized users.
Text: Community-acquired pneumonia (CAP) is a frequently seen disease, with high morbidity and mortality, accounting for 600,000 hospitalizations each year. It represents the seventh leading cause of death in the USA [1] . CAP prognosis depends on the rapidity of specific treatment, which should ideally be initiated within four hours and no later than eight hours after diagnosis [2, 3] . CAP diagnosis is based on the clustering of non-specific pulmonary and general symptoms [4, 5] , an increase in biomarkers reflecting systemic inflammatory response syndrome (SIRS), and the presence of new parenchymal infiltrates on chest X-ray. However, CAP diagnosis remains uncertain in many cases with alternative diagnoses, such as cardiac failure, acute bronchitis, chronic obstructive pulmonary disease (COPD) exacerbations, pulmonary embolism, neoplasia, and sepsis [6, 7] .
Part of the uncertainty of CAP diagnosis may be due to the high rate of chest X-ray misdiagnosis [8, 9] ; over diagnosis of CAP is frequent when infiltrates of noninfectious origin coexist with pulmonary or general symptoms, and the diagnosis of CAP is often ignored when the lung infiltrates are at the limit of visibility or are hidden due to superposition [10] . We recently published a study in which thoracic CT scan was systematically performed in a population of clinically suspected CAP patients visiting the emergency department for CAP (the ESCAPED study) [11] . We showed that CAP diagnosis based on chest X-ray led to a false CAP diagnosis in many patients: among CAP suspected patients with radiological pulmonary infiltrate, CAP diagnosis was excluded in around 30 % of patients based on CT scan results; on the contrary, among patients without radiological pulmonary infiltrate, one-third had a pulmonary infiltrate on thoracic CT-scan. We also reported the isolation of viruses in one-third of patients [11, 12] .
Several attempts have been made to improve CAP diagnosis based on biomarkers, such as C-reactive protein (CRP) and procalcitonin (PCT); however, there are conflicting data on their reliability [13] [14] [15] [16] [17] . This could be due to the consideration of CAP diagnosis based on chest X-ray as establishing pulmonary infection. In the present study, we aimed to analyze CRP and PCT values in the population of the ESCAPED study reported above for whom CAP diagnosis was established by an adjudication committee which founded its judgment on all usual available data, systematic multidetector thoracic CT scan performed at inclusion, and results from a day-28 follow-up. We also analyzed whether the viral etiology of definite CAP based on polymerase chain reaction (PCR) multiplex naso-pharyngeal swab interfered with the accuracy of the biomarkers.
Setting ESCAPED was a multicenter, prospective, interventional study, entitled "Early Thoracic CT-Scan for Community-Acquired Pneumonia at the Emergency Department (ESCAPED)" [11] , conducted from November 2011 to January 2013, in four emergency departments (EDs) of four tertiary teaching hospitals in Paris, France, designed to measure the impact of thoracic CT scan on clinical decision. The study was sponsored and monitored by the Paris public health hospitals, and funded by the French Ministry of Health. The French health authorities (Agence nationale de sécurité des medicaments et produits de santé, ANSM) and the institutional review board for the protection of human subjects approved the study protocol and patient informed consent procedures. All enrolled patients provided written informed consent for inclusion. The protocol was registered in the clinicaltrial.gov website under the PACSCAN acronym, the French translation of the English ESCAPED acronym (NCT01574066). The Ethics Committee of Ile de France (Comité de Protection des Personnes. Paris N°2 011-oct-12749) approved the study protocol.
The primary objective was to compare CRP and PCT values in the four different categories of CAP level of certainty using the day-28 adjudication committee classification. The four categories were: 1) absence of CAP hereafter referred to as excluded CAP diagnosis; 2) possible CAP; 3) probable CAP; and 4) definite CAP. The secondary objectives were to assess whether CRP and PCT were associated with CAP diagnosis using sensitivity analyses in three successive subgroups chosen a priori; 1) when specifically considering patients classified as having excluded CAP diagnosis and definite CAP (i.e., the patients for whom the level of certainty was the highest); 2) when patients with excluded CAP diagnosis and diagnosed extra-pulmonary infectious disease (which may increase biomarker values) were not taken into account, in the excluded CAP group; and 3) when patients classified as viral CAP were not taken into account in the definite CAP group, as PCT has been reported to be lower in viral infections as compared to bacterial infections [18] .
Consecutive adults ( [19] .
Multidetector thoracic CT-scan was performed after chest X-ray, ideally within the four hours following inclusion. Chest X-ray and thoracic CT-scan were performed using a standardized protocol. The four levels of CAP probability according to CT scan were defined as definite (systematic alveolar condensation, alveolar condensation with peripheral and localized ground glass opacities, bronchiolar focal or multifocal micronodules), probable (peripheral alveolar condensation, retractile systematic alveolar condensation, or diffuse ground glass opacities), possible (pulmonary infarct), or excluded (pulmonary mass, other abnormalities, or normal images). Scan views were recorded on a DVD.
Based on data collected from baseline standardized case report forms, DVD recorded pictures of X-ray and CTscan, and blinded to local interpretations, an adjudication committee consisting of three independent senior experts in infectious diseases, pneumology and radiology retrospectively assigned the probability of CAP diagnosis using the same 4-level Likert scale, with all available data including patients' discharge summary, and follow-up data obtained by assistant investigators who contacted by phone either the patient, relatives or general practitioners at day 28. For this study, the gold standard of CAP was the diagnosis assessed by this adjudication committee. Alternative diagnoses were established for excluded CAP and classified as non-CAP pulmonary diseases and extra-pulmonary infectious diseases and others.
Blood samples were collected at inclusion in sodium heparin-treated tubes, centrifuged, and stored at −40°C until completion of the study. CRP and PCT concentrations were measured a posteriori on plasma collection (see Additional file 1 for methodology), except for patients in whom marker dosage was performed by the emergency practitioner on his own initiative.
Naso-pharyngeal swabs were collected at enrollment and placed in a Middle Virocult MWE (Sigma®) transport medium. Samples were kept at room temperature and sent to the virology laboratory of Bichat -Claude Bernard Hospital (Paris) as soon as possible after collection. The samples were not frozen and thawed. Multiplex PCR (RespiFinder-19 assay (Pathofinder®, Maastricht, Netherlands)) was performed on naso-pharyngeal swabs to detect 15 respiratory viruses -coronavirus 229E, NL63, OC43, human metapneumovirus (hMPV), influenza A, A (H1N1) pdm2009 and B viruses, parainfluenza viruses 1, 2, 3, and 4, respiratory syncytial virus (RSV) A and B, rhinovirus, adenovirus, and 4 intracellular bacteria -Bordetella pertussis, Chlamydophila pneumoniae, Legionella pneumophila, Mycoplasma pneumoniae, in one reaction. The multiplex PCR results were not available to the adjudication committee. Routine microbiological examinations were also performed at the discretion of the emergency physicians and included blood culture, sputum culture, and antigenuria (see Additional file 1 for methodology). CAP, classified as definite, was considered as being of viral origin when multiplex PCR was positive for at least one of the 15 respiratory viruses and no bacteria were found using PCR and routine bacterial microbiological samples (sputum, blood culture, antigenuria) when performed.
Baseline and follow-up characteristics were described by means and standard deviations (SD) or by median and interquartile range (IQR) for continuous variables normally distributed or with skewed distribution, respectively, and by percentages for categorical variables, for the total study population and for the study groups. We performed chi-square or Fisher exact tests when appropriate for qualitative variables, and the Student or Mann-Whitney tests for continuous variables with skewed distributions to compare baseline patient characteristics and study outcomes between study groups.
The distribution values of the biomarkers were determined in the different populations of patients using boxplots. The performances of CRP and PCT in predicting definite CAP were evaluated by sensitivity analysis (definite CAP vs excluded CAP). CRP was evaluated at several cut-off points of 20 mg/L, 30 mg/L, 50 mg/L, 70 mg/L, and 100 mg/L, values used in previous studies [15, 20, 21] . Several cut-off points for PCT were chosen at the level of 0.10 μg/L [18] , and at the two levels for suspected bacterial infection as stated by the manufacturer, i.e., 0.25 μg/L and 0.50 μg/L. Sensitivities, specificities, positive predictive values (PPVs), negative predictive values (NPVs), and likelihood ratio were calculated. Receiver operating characteristic (ROC) curves were drawn, area under the curve AUC was computed and optimal cut-off was identified by the maximization of the Youden's index, comparing biomarker values in patients with excluded CAP and definite CAP. From these optimal cut-offs for CRP and PCT, sensitivity analyses were performed combining the CRP and PCT cut-offs.
A multivariate logistic regression model was built to identify factors associated with having definite CAP as compared to having an excluded CAP diagnosis. We excluded from the excluded CAP diagnosis group, patients with an extra-pulmonary infectious disease. All variables with a p value of < 0.25 in the bivariate analysis were entered into a multivariate logistic regression with a backward stepwise approach; the discrimination was evaluated by the C-index and its 95 % confidence interval (95 % CI) and the calibration was evaluated by the Hosmer Lemeshow goodness-of-fit test.
All tests were two-sided, and p-values below 0.05 were considered to denote statistical significance. All statistical analyses were performed using SPSS statistical software version 21.0 (SPSS Inc., Chicago, IL, USA).
Two hundred patients with suspected CAP out of the 319 in the ESCAPED study were included in the present study, for which CRP and PCT assays and nasopharyngeal swab for multiplex PCR were available (Fig. 1) . Characteristics of the 200 patients (age, age more than 65, gender, probability of CAP diagnosis by adjudication committee) were not significantly different from those of the 119 other patients of the ESCAPED study and are summarized in Table 1 . CRP and PCT assays were performed based on the emergency practitioner's own initiative in 70 patients for CRP and 131 for PCT, or performed a posteriori on plasma samples of the remaining patients. Sex ratio was approximately 1. More than half of the patients (54 %) were 65 years of age or older. The
Pulmonary infiltrates were seen on chest X-ray in 127 (63.5 %) patients. Thoracic CT-scan excluded a CAP diagnosis in 16.5 % of these 127 patients; on the contrary, thoracic CT-scan revealed a parenchymal infiltrate in 27 % of the 73 patients without infiltrate on chest X-ray.
Based on all available data including multidetector CT scan results (but excluding PCR results), the adjudication
The CRP and PCT distributions in the 200 patients are presented in Fig. 2 A statistically significant difference between the two groups (excluded CAP vs definite CAP) was demonstrated for several cut-off points for CRP and PCT ( Table 2 ). For CRP, the value of 50 mg/L resulted in a PPV of 0.76 and a NPV of 0.75. For PCT, no value resulted in a satisfactory PPV or NPV. For these two biochemical markers, the ability to predict CAP was evaluated by a ROC curve. The AUC was 0.787 (95 % CI 0.717-0.857), optimal cut-off = 45.9 mg/L for CRP (Fig. 3 ) and 0.655 (95 % CI 0.570-0.739), optimal cut-off = 0.13 μg/ L for PCT (Fig. 4) .
Sensitivity analyses for the combination of CRP and PCT, using these optimal cut-offs, resulted in a PPV of 0.74 and a NPV of 0.58. Use of the other PCT cut-offs did not result in better PPV or NPV ( Table 2) .
The present study is novel as patients prospectively benefited from extensive investigation to determine the diagnosis of CAP in the ED, including both early multidetector thoracic CT-scan and day-28 adjudication committee. This led to the correction of CAP diagnosis previously based on chest X-ray in a high number of patients. In these extensively characterized patients, both CRP and PCT lacked operational precision to allow the decisionmaking process to rule out or confirm diagnosis of CAP even in selected subgroups.
The clinical characteristics of the patients included in this sub-study are consistent with those in the current literature. As previously reported, patients frequently had a history of respiratory disorders, cancer and congestive heart failure [21, 22] . The design of the ESCAPED study required exclusion of patients within the highest CRB 65 categories, which limited the inclusion of patients older than 65. This may explain why the mean age of our patients (64 years) falls within the lower values of those reported elsewhere [19] . Data to identify the microbial agent responsible for the disease were collected by the usual techniques and multiplex PCR. Viral identification using naso-pharyngeal PCR that revealed viral respiratory infection in approximately one-third of cases was concordant with values reported in the literature [23] . Therefore, we believe that our results can be extrapolated to most emergency patients suffering from CAP.
In the present study, patients were recruited on the basis of initial clinical assessment for the diagnosis of CAP. Therefore, we believe that the characteristics of the patients closely correspond to those that lead practitioners to consider a possible diagnosis of CAP. In these patients, the design of our study allowed us to confirm or refute CAP diagnosis with a high level of certainty. Results confirmed the poor predictive value of clinical symptoms (new onset of systemic features and symptoms of an acute lower respiratory tract illness) in identifying CAP patients [21] . Indeed, clinical presentation of excluded CAP patients was similar to that of definite CAP patients except for fever and cough that were more frequent in definite CAP patients. Furthermore, the design also revealed that the combination of clinical symptoms and chest X-ray results led to CAP misdiagnosis in a high number of patients, including the 98 whose CAP diagnosis was excluded by the adjudication committee and who would have been considered as possible, probable or definite CAP without the use of the CT scan. This low specificity of clinical-standard radiological evaluation led to the consideration of either non-infectious pulmonary diseases (such as, cardiac failure, pulmonary embolism, pulmonary neoplasia or bronchitis) or extra-pulmonary infectious diseases as CAP. Of note, some of these diseases are also associated with increased biomarker values. This raises concerns about previous evaluations of biomarkers in CAP-suspected patients, which used clinical and standard radiological (chest X-ray) evaluations as the gold standard for CAP diagnosis [15] . The use of biomarkers has been advocated to improve diagnosis and management of patients with lower respiratory tract infections [14] . However, this issue is still unresolved [24] , with conflicting positions [14, 15, 25, 26] . In our study, while median values of both biomarkers did increase with level of certainty for CAP diagnosis, we were unable to establish discriminating values for PCT. Recent data suggested that CRP could be of more help in assisting in the diagnosis of lower respiratory tract infections (LRTI) [15, 27, 28] . In our study, although CRP seems more discriminating than PCT, neither the experimental exclusion of extra-pulmonary bacterial infections from the excluded CAP group, nor the exclusion of viral CAP from the definite CAP patients group, made possible the determination of a discriminant cutoff. The combination of CRP and PCT was not more discriminating than each biomarker separately. An operational algorithm has been released to assist physicians in prescribing antimicrobial therapy [14, 26, 29] . According to this strategy, a PCT concentration higher than 0.25 μg/L should prompt administration of antibiotics to patients with suspected LRTI. In our study, this value was associated with poor performance. Additionally, mean PCT levels remained above this threshold both in excluded CAP patients without infectious disorders and in definite CAP presumably related to virus. Therefore, the gold standard for the diagnosis of CAP may influence the performance and utility of PCT in this setting.
This study has some limitations. First, the adjudication committee was not blinded to the value of biomarkers measured at bedside in some patients (70 for CRP and 131 for PCT) and its CAP classification could thus have been influenced by these results. However, the lack of statistically significant differences in the mean CRP and PCT values in the definite CAP cases, whether or not these biomarkers were available for the adjudication committee, argues against a major impact of these results on adjudication committee classification. Second, another critical point is the prescription of antibiotic therapy (34 %) previous to inclusion. We cannot exclude that these previously-treated CAP patients may have altered biomarker performance and reduced the yield of bacterial cultures, although such a population reflects the usual emergency department practice. Third, multiplex PCR was performed on naso-pharyngeal sampling and not on lower respiratory tract samples, which does not allow definite confirmation of the viral origin of CAP. However, a recent large study on CAP patients which reported a viral etiology of CAP at a comparable rate, did not find upper respiratory tract shedding in a control population without CAP explored during the same year and season [30] . Finally, even if multidetector thoracic CT scan is a better imaging examination than X-ray to explore the chest, only invasive local microbiological samples would have provided a diagnosis with certainty.
Given the diversity of the clinical and radiological CAP presentations, CAP diagnosis is often uncertain. In our population of patients treated in the emergency room with clinical symptoms evoking CAP, neither CRP nor PCT cut-off values carried sufficient weight to confirm or refute CAP diagnosis at bedside; this underlines that these biomarkers are telltales of the host inflammatory response to the intrusion of microorganisms independent of the site of infection. These results, based on a systematic thoracic CT scan evaluation of CAP-suspected patients, do not argue for the use of CRP and PCT in routine care to diagnose CAP with certainty in patients visiting the ED for suspected CAP.
|
How many patients with community-acquired pneumonia are hospitalized each year?
| 5,253 |
600,000
| 3,121 |
1,599 |
Diagnostic accuracy of C-reactive protein and procalcitonin in suspected community-acquired pneumonia adults visiting emergency department and having a systematic thoracic CT scan
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4608327/
SHA: f3d150545162ff3cc253c235011a02a91ee676cb
Authors: Le Bel, Josselin; Hausfater, Pierre; Chenevier-Gobeaux, Camille; Blanc, François-Xavier; Benjoar, Mikhael; Ficko, Cécile; Ray, Patrick; Choquet, Christophe; Duval, Xavier; Claessens, Yann-Erick
Date: 2015-10-16
DOI: 10.1186/s13054-015-1083-6
License: cc-by
Abstract: INTRODUCTION: Community-acquired pneumonia (CAP) requires prompt treatment, but its diagnosis is complex. Improvement of bacterial CAP diagnosis by biomarkers has been evaluated using chest X-ray infiltrate as the CAP gold standard, producing conflicting results. We analyzed the diagnostic accuracy of biomarkers in suspected CAP adults visiting emergency departments for whom CAP diagnosis was established by an adjudication committee which founded its judgment on a systematic multidetector thoracic CT scan. METHODS: In an ancillary study of a multi-center prospective study evaluating the impact of systematic thoracic CT scan on CAP diagnosis, sensitivity and specificity of C-reactive protein (CRP) and procalcitonin (PCT) were evaluated. Systematic nasopharyngeal multiplex respiratory virus PCR was performed at inclusion. An adjudication committee classified CAP diagnostic probability on a 4-level Likert scale, based on all available data. RESULTS: Two hundred patients with suspected CAP were analyzed. The adjudication committee classified 98 patients (49.0 %) as definite CAP, 8 (4.0 %) as probable, 23 (11.5 %) as possible and excluded in 71 (35.5 %, including 29 patients with pulmonary infiltrates on chest X-ray). Among patients with radiological pulmonary infiltrate, 23 % were finally classified as excluded. Viruses were identified by PCR in 29 % of patients classified as definite. Area under the curve was 0.787 [95 % confidence interval (95 % CI), 0.717 to 0.857] for CRP and 0.655 (95 % CI, 0.570 to 0.739) for PCT to detect definite CAP. CRP threshold at 50 mg/L resulted in a positive predictive value of 0.76 and a negative predictive value of 0.75. No PCT cut-off resulted in satisfactory positive or negative predictive values. CRP and PCT accuracy was not improved by exclusion of the 25 (25.5 %) definite viral CAP cases. CONCLUSIONS: For patients with suspected CAP visiting emergency departments, diagnostic accuracy of CRP and PCT are insufficient to confirm the CAP diagnosis established using a gold standard that includes thoracic CT scan. Diagnostic accuracy of these biomarkers is also insufficient to distinguish bacterial CAP from viral CAP. TRIAL REGISTRATION: ClinicalTrials.gov registry NCT01574066 (February 7, 2012) ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13054-015-1083-6) contains supplementary material, which is available to authorized users.
Text: Community-acquired pneumonia (CAP) is a frequently seen disease, with high morbidity and mortality, accounting for 600,000 hospitalizations each year. It represents the seventh leading cause of death in the USA [1] . CAP prognosis depends on the rapidity of specific treatment, which should ideally be initiated within four hours and no later than eight hours after diagnosis [2, 3] . CAP diagnosis is based on the clustering of non-specific pulmonary and general symptoms [4, 5] , an increase in biomarkers reflecting systemic inflammatory response syndrome (SIRS), and the presence of new parenchymal infiltrates on chest X-ray. However, CAP diagnosis remains uncertain in many cases with alternative diagnoses, such as cardiac failure, acute bronchitis, chronic obstructive pulmonary disease (COPD) exacerbations, pulmonary embolism, neoplasia, and sepsis [6, 7] .
Part of the uncertainty of CAP diagnosis may be due to the high rate of chest X-ray misdiagnosis [8, 9] ; over diagnosis of CAP is frequent when infiltrates of noninfectious origin coexist with pulmonary or general symptoms, and the diagnosis of CAP is often ignored when the lung infiltrates are at the limit of visibility or are hidden due to superposition [10] . We recently published a study in which thoracic CT scan was systematically performed in a population of clinically suspected CAP patients visiting the emergency department for CAP (the ESCAPED study) [11] . We showed that CAP diagnosis based on chest X-ray led to a false CAP diagnosis in many patients: among CAP suspected patients with radiological pulmonary infiltrate, CAP diagnosis was excluded in around 30 % of patients based on CT scan results; on the contrary, among patients without radiological pulmonary infiltrate, one-third had a pulmonary infiltrate on thoracic CT-scan. We also reported the isolation of viruses in one-third of patients [11, 12] .
Several attempts have been made to improve CAP diagnosis based on biomarkers, such as C-reactive protein (CRP) and procalcitonin (PCT); however, there are conflicting data on their reliability [13] [14] [15] [16] [17] . This could be due to the consideration of CAP diagnosis based on chest X-ray as establishing pulmonary infection. In the present study, we aimed to analyze CRP and PCT values in the population of the ESCAPED study reported above for whom CAP diagnosis was established by an adjudication committee which founded its judgment on all usual available data, systematic multidetector thoracic CT scan performed at inclusion, and results from a day-28 follow-up. We also analyzed whether the viral etiology of definite CAP based on polymerase chain reaction (PCR) multiplex naso-pharyngeal swab interfered with the accuracy of the biomarkers.
Setting ESCAPED was a multicenter, prospective, interventional study, entitled "Early Thoracic CT-Scan for Community-Acquired Pneumonia at the Emergency Department (ESCAPED)" [11] , conducted from November 2011 to January 2013, in four emergency departments (EDs) of four tertiary teaching hospitals in Paris, France, designed to measure the impact of thoracic CT scan on clinical decision. The study was sponsored and monitored by the Paris public health hospitals, and funded by the French Ministry of Health. The French health authorities (Agence nationale de sécurité des medicaments et produits de santé, ANSM) and the institutional review board for the protection of human subjects approved the study protocol and patient informed consent procedures. All enrolled patients provided written informed consent for inclusion. The protocol was registered in the clinicaltrial.gov website under the PACSCAN acronym, the French translation of the English ESCAPED acronym (NCT01574066). The Ethics Committee of Ile de France (Comité de Protection des Personnes. Paris N°2 011-oct-12749) approved the study protocol.
The primary objective was to compare CRP and PCT values in the four different categories of CAP level of certainty using the day-28 adjudication committee classification. The four categories were: 1) absence of CAP hereafter referred to as excluded CAP diagnosis; 2) possible CAP; 3) probable CAP; and 4) definite CAP. The secondary objectives were to assess whether CRP and PCT were associated with CAP diagnosis using sensitivity analyses in three successive subgroups chosen a priori; 1) when specifically considering patients classified as having excluded CAP diagnosis and definite CAP (i.e., the patients for whom the level of certainty was the highest); 2) when patients with excluded CAP diagnosis and diagnosed extra-pulmonary infectious disease (which may increase biomarker values) were not taken into account, in the excluded CAP group; and 3) when patients classified as viral CAP were not taken into account in the definite CAP group, as PCT has been reported to be lower in viral infections as compared to bacterial infections [18] .
Consecutive adults ( [19] .
Multidetector thoracic CT-scan was performed after chest X-ray, ideally within the four hours following inclusion. Chest X-ray and thoracic CT-scan were performed using a standardized protocol. The four levels of CAP probability according to CT scan were defined as definite (systematic alveolar condensation, alveolar condensation with peripheral and localized ground glass opacities, bronchiolar focal or multifocal micronodules), probable (peripheral alveolar condensation, retractile systematic alveolar condensation, or diffuse ground glass opacities), possible (pulmonary infarct), or excluded (pulmonary mass, other abnormalities, or normal images). Scan views were recorded on a DVD.
Based on data collected from baseline standardized case report forms, DVD recorded pictures of X-ray and CTscan, and blinded to local interpretations, an adjudication committee consisting of three independent senior experts in infectious diseases, pneumology and radiology retrospectively assigned the probability of CAP diagnosis using the same 4-level Likert scale, with all available data including patients' discharge summary, and follow-up data obtained by assistant investigators who contacted by phone either the patient, relatives or general practitioners at day 28. For this study, the gold standard of CAP was the diagnosis assessed by this adjudication committee. Alternative diagnoses were established for excluded CAP and classified as non-CAP pulmonary diseases and extra-pulmonary infectious diseases and others.
Blood samples were collected at inclusion in sodium heparin-treated tubes, centrifuged, and stored at −40°C until completion of the study. CRP and PCT concentrations were measured a posteriori on plasma collection (see Additional file 1 for methodology), except for patients in whom marker dosage was performed by the emergency practitioner on his own initiative.
Naso-pharyngeal swabs were collected at enrollment and placed in a Middle Virocult MWE (Sigma®) transport medium. Samples were kept at room temperature and sent to the virology laboratory of Bichat -Claude Bernard Hospital (Paris) as soon as possible after collection. The samples were not frozen and thawed. Multiplex PCR (RespiFinder-19 assay (Pathofinder®, Maastricht, Netherlands)) was performed on naso-pharyngeal swabs to detect 15 respiratory viruses -coronavirus 229E, NL63, OC43, human metapneumovirus (hMPV), influenza A, A (H1N1) pdm2009 and B viruses, parainfluenza viruses 1, 2, 3, and 4, respiratory syncytial virus (RSV) A and B, rhinovirus, adenovirus, and 4 intracellular bacteria -Bordetella pertussis, Chlamydophila pneumoniae, Legionella pneumophila, Mycoplasma pneumoniae, in one reaction. The multiplex PCR results were not available to the adjudication committee. Routine microbiological examinations were also performed at the discretion of the emergency physicians and included blood culture, sputum culture, and antigenuria (see Additional file 1 for methodology). CAP, classified as definite, was considered as being of viral origin when multiplex PCR was positive for at least one of the 15 respiratory viruses and no bacteria were found using PCR and routine bacterial microbiological samples (sputum, blood culture, antigenuria) when performed.
Baseline and follow-up characteristics were described by means and standard deviations (SD) or by median and interquartile range (IQR) for continuous variables normally distributed or with skewed distribution, respectively, and by percentages for categorical variables, for the total study population and for the study groups. We performed chi-square or Fisher exact tests when appropriate for qualitative variables, and the Student or Mann-Whitney tests for continuous variables with skewed distributions to compare baseline patient characteristics and study outcomes between study groups.
The distribution values of the biomarkers were determined in the different populations of patients using boxplots. The performances of CRP and PCT in predicting definite CAP were evaluated by sensitivity analysis (definite CAP vs excluded CAP). CRP was evaluated at several cut-off points of 20 mg/L, 30 mg/L, 50 mg/L, 70 mg/L, and 100 mg/L, values used in previous studies [15, 20, 21] . Several cut-off points for PCT were chosen at the level of 0.10 μg/L [18] , and at the two levels for suspected bacterial infection as stated by the manufacturer, i.e., 0.25 μg/L and 0.50 μg/L. Sensitivities, specificities, positive predictive values (PPVs), negative predictive values (NPVs), and likelihood ratio were calculated. Receiver operating characteristic (ROC) curves were drawn, area under the curve AUC was computed and optimal cut-off was identified by the maximization of the Youden's index, comparing biomarker values in patients with excluded CAP and definite CAP. From these optimal cut-offs for CRP and PCT, sensitivity analyses were performed combining the CRP and PCT cut-offs.
A multivariate logistic regression model was built to identify factors associated with having definite CAP as compared to having an excluded CAP diagnosis. We excluded from the excluded CAP diagnosis group, patients with an extra-pulmonary infectious disease. All variables with a p value of < 0.25 in the bivariate analysis were entered into a multivariate logistic regression with a backward stepwise approach; the discrimination was evaluated by the C-index and its 95 % confidence interval (95 % CI) and the calibration was evaluated by the Hosmer Lemeshow goodness-of-fit test.
All tests were two-sided, and p-values below 0.05 were considered to denote statistical significance. All statistical analyses were performed using SPSS statistical software version 21.0 (SPSS Inc., Chicago, IL, USA).
Two hundred patients with suspected CAP out of the 319 in the ESCAPED study were included in the present study, for which CRP and PCT assays and nasopharyngeal swab for multiplex PCR were available (Fig. 1) . Characteristics of the 200 patients (age, age more than 65, gender, probability of CAP diagnosis by adjudication committee) were not significantly different from those of the 119 other patients of the ESCAPED study and are summarized in Table 1 . CRP and PCT assays were performed based on the emergency practitioner's own initiative in 70 patients for CRP and 131 for PCT, or performed a posteriori on plasma samples of the remaining patients. Sex ratio was approximately 1. More than half of the patients (54 %) were 65 years of age or older. The
Pulmonary infiltrates were seen on chest X-ray in 127 (63.5 %) patients. Thoracic CT-scan excluded a CAP diagnosis in 16.5 % of these 127 patients; on the contrary, thoracic CT-scan revealed a parenchymal infiltrate in 27 % of the 73 patients without infiltrate on chest X-ray.
Based on all available data including multidetector CT scan results (but excluding PCR results), the adjudication
The CRP and PCT distributions in the 200 patients are presented in Fig. 2 A statistically significant difference between the two groups (excluded CAP vs definite CAP) was demonstrated for several cut-off points for CRP and PCT ( Table 2 ). For CRP, the value of 50 mg/L resulted in a PPV of 0.76 and a NPV of 0.75. For PCT, no value resulted in a satisfactory PPV or NPV. For these two biochemical markers, the ability to predict CAP was evaluated by a ROC curve. The AUC was 0.787 (95 % CI 0.717-0.857), optimal cut-off = 45.9 mg/L for CRP (Fig. 3 ) and 0.655 (95 % CI 0.570-0.739), optimal cut-off = 0.13 μg/ L for PCT (Fig. 4) .
Sensitivity analyses for the combination of CRP and PCT, using these optimal cut-offs, resulted in a PPV of 0.74 and a NPV of 0.58. Use of the other PCT cut-offs did not result in better PPV or NPV ( Table 2) .
The present study is novel as patients prospectively benefited from extensive investigation to determine the diagnosis of CAP in the ED, including both early multidetector thoracic CT-scan and day-28 adjudication committee. This led to the correction of CAP diagnosis previously based on chest X-ray in a high number of patients. In these extensively characterized patients, both CRP and PCT lacked operational precision to allow the decisionmaking process to rule out or confirm diagnosis of CAP even in selected subgroups.
The clinical characteristics of the patients included in this sub-study are consistent with those in the current literature. As previously reported, patients frequently had a history of respiratory disorders, cancer and congestive heart failure [21, 22] . The design of the ESCAPED study required exclusion of patients within the highest CRB 65 categories, which limited the inclusion of patients older than 65. This may explain why the mean age of our patients (64 years) falls within the lower values of those reported elsewhere [19] . Data to identify the microbial agent responsible for the disease were collected by the usual techniques and multiplex PCR. Viral identification using naso-pharyngeal PCR that revealed viral respiratory infection in approximately one-third of cases was concordant with values reported in the literature [23] . Therefore, we believe that our results can be extrapolated to most emergency patients suffering from CAP.
In the present study, patients were recruited on the basis of initial clinical assessment for the diagnosis of CAP. Therefore, we believe that the characteristics of the patients closely correspond to those that lead practitioners to consider a possible diagnosis of CAP. In these patients, the design of our study allowed us to confirm or refute CAP diagnosis with a high level of certainty. Results confirmed the poor predictive value of clinical symptoms (new onset of systemic features and symptoms of an acute lower respiratory tract illness) in identifying CAP patients [21] . Indeed, clinical presentation of excluded CAP patients was similar to that of definite CAP patients except for fever and cough that were more frequent in definite CAP patients. Furthermore, the design also revealed that the combination of clinical symptoms and chest X-ray results led to CAP misdiagnosis in a high number of patients, including the 98 whose CAP diagnosis was excluded by the adjudication committee and who would have been considered as possible, probable or definite CAP without the use of the CT scan. This low specificity of clinical-standard radiological evaluation led to the consideration of either non-infectious pulmonary diseases (such as, cardiac failure, pulmonary embolism, pulmonary neoplasia or bronchitis) or extra-pulmonary infectious diseases as CAP. Of note, some of these diseases are also associated with increased biomarker values. This raises concerns about previous evaluations of biomarkers in CAP-suspected patients, which used clinical and standard radiological (chest X-ray) evaluations as the gold standard for CAP diagnosis [15] . The use of biomarkers has been advocated to improve diagnosis and management of patients with lower respiratory tract infections [14] . However, this issue is still unresolved [24] , with conflicting positions [14, 15, 25, 26] . In our study, while median values of both biomarkers did increase with level of certainty for CAP diagnosis, we were unable to establish discriminating values for PCT. Recent data suggested that CRP could be of more help in assisting in the diagnosis of lower respiratory tract infections (LRTI) [15, 27, 28] . In our study, although CRP seems more discriminating than PCT, neither the experimental exclusion of extra-pulmonary bacterial infections from the excluded CAP group, nor the exclusion of viral CAP from the definite CAP patients group, made possible the determination of a discriminant cutoff. The combination of CRP and PCT was not more discriminating than each biomarker separately. An operational algorithm has been released to assist physicians in prescribing antimicrobial therapy [14, 26, 29] . According to this strategy, a PCT concentration higher than 0.25 μg/L should prompt administration of antibiotics to patients with suspected LRTI. In our study, this value was associated with poor performance. Additionally, mean PCT levels remained above this threshold both in excluded CAP patients without infectious disorders and in definite CAP presumably related to virus. Therefore, the gold standard for the diagnosis of CAP may influence the performance and utility of PCT in this setting.
This study has some limitations. First, the adjudication committee was not blinded to the value of biomarkers measured at bedside in some patients (70 for CRP and 131 for PCT) and its CAP classification could thus have been influenced by these results. However, the lack of statistically significant differences in the mean CRP and PCT values in the definite CAP cases, whether or not these biomarkers were available for the adjudication committee, argues against a major impact of these results on adjudication committee classification. Second, another critical point is the prescription of antibiotic therapy (34 %) previous to inclusion. We cannot exclude that these previously-treated CAP patients may have altered biomarker performance and reduced the yield of bacterial cultures, although such a population reflects the usual emergency department practice. Third, multiplex PCR was performed on naso-pharyngeal sampling and not on lower respiratory tract samples, which does not allow definite confirmation of the viral origin of CAP. However, a recent large study on CAP patients which reported a viral etiology of CAP at a comparable rate, did not find upper respiratory tract shedding in a control population without CAP explored during the same year and season [30] . Finally, even if multidetector thoracic CT scan is a better imaging examination than X-ray to explore the chest, only invasive local microbiological samples would have provided a diagnosis with certainty.
Given the diversity of the clinical and radiological CAP presentations, CAP diagnosis is often uncertain. In our population of patients treated in the emergency room with clinical symptoms evoking CAP, neither CRP nor PCT cut-off values carried sufficient weight to confirm or refute CAP diagnosis at bedside; this underlines that these biomarkers are telltales of the host inflammatory response to the intrusion of microorganisms independent of the site of infection. These results, based on a systematic thoracic CT scan evaluation of CAP-suspected patients, do not argue for the use of CRP and PCT in routine care to diagnose CAP with certainty in patients visiting the ED for suspected CAP.
|
What chest X-ray findings are typically indicative of community-acquired pneumonia?
| 5,254 |
the presence of new parenchymal infiltrates
| 3,577 |
1,601 |
Pandemic Influenza Due to pH1N1/2009 Virus: Estimation of Infection Burden in Reunion Island through a Prospective Serosurvey, Austral Winter 2009
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3183080/
SHA: ee6d70a53e3262cea6f85bd8b226f6b4c8b5f64b
Authors: Dellagi, Koussay; Rollot, Olivier; Temmam, Sarah; Salez, Nicolas; Guernier, Vanina; Pascalis, Hervé; Gérardin, Patrick; Fianu, Adrian; Lapidus, Nathanael; Naty, Nadège; Tortosa, Pablo; Boussaïd, Karim; Jaffar-Banjee, Marie-Christine; Filleul, Laurent; Flahault, Antoine; Carrat, Fabrice; Favier, Francois; de Lamballerie, Xavier
Date: 2011-09-29
DOI: 10.1371/journal.pone.0025738
License: cc-by
Abstract: BACKGROUND: To date, there is little information that reflects the true extent of spread of the pH1N1/2009v influenza pandemic at the community level as infection often results in mild or no clinical symptoms. This study aimed at assessing through a prospective study, the attack rate of pH1N1/2009 virus in Reunion Island and risk factors of infection, during the 2009 season. METHODOLOGY/PRINCIPAL FINDINGS: A serosurvey was conducted during the 2009 austral winter, in the frame of a prospective population study. Pairs of sera were collected from 1687 individuals belonging to 772 households, during and after passage of the pandemic wave. Antibodies to pH1N1/2009v were titered using the hemagglutination inhibition assay (HIA) with titers ≥1/40 being considered positive. Seroprevalence during the first two weeks of detection of pH1N1/2009v in Reunion Island was 29.8% in people under 20 years of age, 35.6% in adults (20–59 years) and 73.3% in the elderly (≥60 years) (P<0.0001). Baseline corrected cumulative incidence rates, were 42.9%, 13.9% and 0% in these age groups respectively (P<0.0001). A significant decline in antibody titers occurred soon after the passage of the epidemic wave. Seroconversion rates to pH1N1/2009 correlated negatively with age: 63.2%, 39.4% and 16.7%, in each age group respectively (P<0.0001). Seroconversion occurred in 65.2% of individuals who were seronegative at inclusion compared to 6.8% in those who were initially seropositive. CONCLUSIONS: Seroincidence of pH1N1/2009v infection was three times that estimated from clinical surveillance, indicating that almost two thirds of infections occurring at the community level have escaped medical detection. People under 20 years of age were the most affected group. Pre-epidemic titers ≥1/40 prevented seroconversion and are likely protective against infection. A concern was raised about the long term stability of the antibody responses.
Text: In April 2009, the first cases of acute respiratory infections caused by a novel triple-reassortant influenza virus, pH1N1/ 2009v, occurred in Mexico and the United States [1] . The rapid spread of infection to other continents led the World Health Organization (WHO) to declare on 11 June 2009 that a pandemic of pH1N1/2009v influenza was under way, which raised major international concern about the risk of high morbidity and lethality and the potential for severe socio-economic impact. Actually, the potential impact of this first third-millenium influenza pandemic has been revisited downwards as morbidity and case-fatality rates were less severe than initially anticipated [2] . Illness surveillance data do not allow to an accurate estimate of the true influenza infection rate, as a substantial proportion of infections are asymptomatic or mild [3] . Serological surveys can overcome this limitation, but must take into account that a significant proportion of the population that exhibited crossprotective antibody titers before circulation of the pH1N1/2009v [4] . This so-called ''baseline immunity'' has to be subtracted from the seroprevalence observed after the pandemic wave, to determine seroincidence in serosurveys [5] [6] [7] [8] . However, except for few studies [9] [10] [11] , most of these serosurveys did not use serial measurements in the same person, which allows for a better understanding of antibody kinetics and the dynamics of infection within individuals and communities.
Reunion Island (805,500 inhabitants) is a French overseas department located in the southwestern Indian Ocean, 700 km east of Madagascar and 200 km southwest of Mauritius. The first imported case of pH1N1/2009v was identified on 5 th July 2009 (week 29) in a traveller returning from Australia. The first case indicating community transmission was detected on 21 st July (week 30). pH1N1/2009v became the predominant circulating influenza virus within four weeks of its first detection, its activity peaked during week 35 (24) (25) (26) (27) (28) (29) (30) and ended at week 38 [12] . Contrary to initial fears, the health care system was not overwhelmed, as morbidity and mortality rates were lower than predicted [12] [13] [14] .
In order to assess at the community level, the actual magnitude of the pH1N1/2009v pandemic and the extent of the herd immunity acquired after passage of the epidemic wave, a prospective population serosurvey was conducted in Reunion Island during the passage of the epidemic wave in the 2009 austral winter season (July-December 2009): prevalence of infection was assessed on a weekly basis and seroconversion rates were measured using paired sera.
The CoPanFLu-RUN was part of the CoPanFLu international project, a consortium between the French National Institute of Health and Medical Research (INSERM), the Institute of Research for Development (IRD) and the Mérieux Fondation under the promotion of the School of Advanced Studies in Public Health (EHESP). To enable the rapid implementation of the study in anticipation of the imminent spread of the pandemic wave, we used a pre-existing sample of 2442 households established in October 2006 for the investigation of the Chikungunya outbreak (SEROCHIK) and updated in May 2008 throughout a follow-up telephone survey (TELECHIK) on a basis of 1148 households [15, 16] . We took special attention to select households representing a wide range of geographic locations in order to minimize the repartition bias.
The inclusion phase started on July 21 st (week 30) and was continued up to week 44, throughout the epidemic wave and beyond. A first serum sample (sample 1) was obtained from each household member. An active telephonic inquiry was then conducted twice a week to record symptoms compatible with influenza-like illness (ILI) occurring in households. Report of ILI (fever $37.8uC associated with any respiratory or systemic symptom) led to three consecutive visits of a nurse to the incident case-dwelling (on day 0, +3 and +8 post-report) to record symptoms and collect nasal swabs from all family members (for qRT-PCR detection of pH1N1/2009v. At week 45, the active inquiry was discontinued and a second (post-epidemic) serum sample (sample 2) was obtained (weeks 45-52) to determine seroconversion rates. Sera were aliquoted and stored at 280uC.
The protocol was conducted in accordance with the Declaration of Helsinki and French law for biomedical research (Nu ID RCB AFSSAPS: 2009-A00689-48) and was approved by the local Ethics Committee (Comité de Protection des Personnes of Bordeaux 2 University). Every eligible person for participation was asked for giving their written informed consent.
Viral genome detection by RT-PCR. Viral RNA was extracted from 140 mL of nasal swab eluate using the QIAamp Viral RNA kit (Qiagen) and processed for detection by TaqMan qRT-PCR targeting the heamagglutinin HA gene (SuperScript III Platinum one-step qRT-PCR system, Invitrogen) according to the recommendations of the Pasteur Institute (Van der Werf S. & Enouf V., SOP/FluA/130509). Confirmed pH1N1/2009v infection was defined as a positive qRT-PCR detection of the HA gene in at least one nasal swab.
Hemagglutination inhibition assay (HIA). A standard hemagglutination inhibition technique was adapted to detect and quantify pH1N1/2009v antibodies [17] . The antigen was prepared by diluting a non-inactivated cell culture supernatant producing a pdm H1N1v strain (strain OPYFLU-1 isolated from a young patient returning from Mexico in early May 2009) [18] . Briefly, the virus was propagated onto MDCK cells under standard conditions. The last passage (used for antigen preparation) was performed in the absence of trypsin and ht-FBS. The supernatant was collected at day seven p.i. clarified by centrifugation at 8006 g for 10 min at room temperature, aliquoted and conserved at 280uC. The hemagglutinating titer of the non inactivated viral antigen was immediately determined under the HIA format described below. The dilution providing 5.33 hemagglutinating units in a volume of 25 mL was used for subsequent HIA. Sera were heat-inactivated at 56uC for 30 min prior to use. Sequential twofold dilutions in PBS (1/10 to 1/1280) in volumes of 25 mL were performed and distributed in V-bottom 96 well microplates. Human red blood cells (RBC) were used for hemagglutination experiments. Detection and quantification of antibody to pH1N1/2009v was performed as follows: 25 mL of virus suspension was added to the serum dilution (25 mL) and incubated for 1 hour at room temperature. Each well was then filled with 25 mL of a 1% RBC suspension in PBS (v/v: 0.33%), followed by another 30 min incubation at room temperature. The HIA titer was determined as the last dilution providing clear inhibition of hemagglutination. All experiments were performed in the presence of the same negative and positive controls, the latter including sera with 1/40, 1/80, 1/160 and 1/320 antibody titers.
The results reported in this study were based only on serological analysis of paired sera. For the sake of analysis, four successive phases were identified throughout the pandemic wave: phase A (weeks 30-31) corresponded to early epidemic time, phase B (W32-39) to the epidemic unfolding, phase C (W40-44) to the immediate post-epidemic stage and phase D (W45-52) to the late post-epidemic stage. Seropositivity was defined as a HIA titer of 1/ 40 or more. The baseline-proxy seroprevalence rate was estimated on serum samples collected in phase A. The cumulative incidence rate of infection measured the raise between the raw seroprevalence rate at any given time during the epidemic phases (S2pi) and the age-specific baseline-proxy seroprevalence rate (S1pA) (s2 pi -s1 pA ). Seroconversion was defined as a shift from seronegative at inclusion (sample 1: HIA ,1/40) to seropositive on follow-up (sample 2: HIA $1/40), or for sera tested seropositive on inclusion as a four-fold increase of HIA titers between sample 1 and sample 2 paired sera. We also calculated the proportion of sera that tested seropositive in sample 1 for which the HIA titer decreased fourfold and passed under the cut-off value of 1/40 in sample 2. We considered this proportion as a ''seronegation'' rate.
The sample size was calculated for identifying risk factors in the prospective cohort study. Considering on average three individuals per household, an intra-household correlation of 0.3, a power greater than 80% could be obtained with a sample size of 840 comprising 2500 individuals, assuming exposure levels ranging from 10% to 90% and a relative risk greater than 1.3. With 2,500 subjects, the study allowed 1-2% absolute precision around the estimated values for seroconversion rates. Data entry used EpiData version 3.1 (The Epidata Association, Odense, Denmark). SAS version 9.1 (SAS Inc., Cary, NC, USA) was used for statistical analysis. The characteristics of the study cohort were compared to those of the population of Reunion Island and a Chi2 test (or Fisher's exact test when non applicable) was used to analyse differences in age, sex and geographic location. Cumulative incidence rates of infection (i.e. seroincidence) and seroconversion rates were standardized according to the age structure of the community (French National Institute for Statistics and Economical Studies (INSEE) source).
Baseline-proxy seroprevalence, cumulative incidence rates of infection, as well as seroconversion and seronegation rates, were expressed as percentages. Cumulative reverse distribution curves were used to show the distribution of antibody titers. In all tests, a P value,0.05 was considered significant.
We estimated 95% confidence intervals (CIs) of proportions by using a cluster bootstrap technique with 1000 re-samples [19] . After bootstraping, we used an ANOVA model to compare mean cumulative incidence proportions between pandemic phases, within each age group. We used an alternating logistic regression model (ALR) with an exchangeable log Odds Ratio (OR) to test the intra-household correlation-adjusted association between factors and the seroconversion outcome.
Data were analysed with respect to subject age. Initially, four age groups were considered: the children and adolescents (,20 yrs), young adults (20-39 yrs), middle-age adults (40-59 yrs), and elderly adults ($60 yrs). As the cumulative incidence of infection of the second and third groups were very close, both groups were merged into one adults group (20-59 yrs). Therefore we refer further in our study to three age groups: children and adolescents (,20 yrs), adults (20-59 yrs), elderly ($60 yrs).
A total of 2,164 individuals from 772 households were enrolled between weeks 30 and 44 in the CoPanFlu-RUN cohort, allowing the collection of 1,932 sera at inclusion (sample 1). During this period, 136 households (17.7% of households) containing 464 individuals (21.4% of individuals) reported at least one case of ILI. Sixty subjects among the 464 individuals (12.9%, belonging to 33 households [24.3%]) were qRT-PCR positive, which documented the pH1N1/2009v infection. No positive qRT-PCR could be detected after week 37 and no ILI was reported after week 40, the end of the epidemic wave. The second follow up serum sample (sample 2) was obtained for 1,759 subjects at least five weeks after the end of the epidemic wave (weeks 45-52) which allowed the constitution of a serobank of 1,687 paired-sera. The profile of the cohort and the major outcomes are displayed in Figure 1 . Details on inclusions and serum sample timing with respect to the circulation of pH1N1/2009v over the island are provided in figure 2 . The socio-demographic and space-time characteristics of the cohort are detailed in Table 1 . Compared to the community of Reunion Island, the sample of 1,687 individuals for whom pairedsera were available, was older (,20 yrs: 27% vs 35%, and $60 yrs: 17,9% vs 11,3%) and composed of a slight excess of females (54.1% vs 51.5%). The imbalance was due to a deficit in subjects aged under 40 years, reflecting men at work and the fact that parents declined the second serum for children younger than five.
Baseline-proxy (,pre-epidemic) HIA titers to the pH1N1/ 2009v were measured on sample 1 ( Table 2) , obtained from 249 subjects (103 households) recruited at the very beginning of the investigation during weeks 30 and 31 (phase A, Figure 2 ), when the epidemic activity in the cohort was still very low. Age distribution in this group was similar to that of the whole cohort (data not shown). The overall, the baseline-proxy seroprevalence rate (HIA $1/40), over all ages, was 43.4% (95%CI: 37.4%-49.6%). However the majority of positive sera had low antibody titers, at the cut off value for confirmation (i.e. = 1/40). The proportions of sera with HIA titer .1/40 were 0%, 3.0% and 24.6% in the young, middle-aged and older age groups respectively. These results indicate that pre-epidemic baseline antibody cross reactivity was stronger in the elderly ($60 yrs) and weaker in children and adolescents (,20 yrs) and adults (20-59 yrs), with highly significant differences between age groups (P,0.0001).
The reverse cumulative distribution curves of HIA titers are displayed for each age group and for the whole cohort on Figure 3 . The proportion of seropositive sera (HI $1/40) steadily increased during the epidemic unfolding (phase B, W32-39) and in immediate post epidemic period (phase C, W40-44) when it reached its maximum level, then declined in the late post epidemic period (phase D, W45-52). This decline was significant enough to return the reverse cumulative distribution curve to baseline levels in the elderly. The cumulative incidence rates, obtained after subtraction of the age-specific baseline-proxy seroprevalence from the raw seroprevalence at each phase of the epidemic are shown in Table 2 (note that the cumulative incidence rates of infection represented for the group ''all ages'' were standardized according to age structure of the community). The cumulative incidence rates were much higher in children and adolescents (,20 yrs), indicating very active transmission of infection within this age group. As mentioned earlier, cumulative incidence rates peaked in phase C (W40-44), and then declined indicating some lability of the humoral immune response against the pH1N1/2009v. The age-related difference observed in the incidence rates was highly statistically significant (P,0.0001).
To estimate more appropriately the decline of antibody titers occurring after the peak of the humoral response to the pH1N1/ 2009v, we considered paired-sera from the group of 264 subjects for whom the first serum sample (sample 1) was obtained just after the epidemic wave (phase C, W40-44), and the corresponding second sample was collected at the end of the survey (phase D, W45-52). Seronegation rates were 27.0% (61/226) for all age groups, 17.4% (12/69) in children and adolescents (,20 yrs), 32.3% (41/127) in adults (20-59 yrs) and 26.7% (8/30) in the elderly ($60 yrs). Differences between the seronegation rates according to age were statistically weakly significant (P = 0.0671).
We then considered the 1687 individuals for whom paired sera were available and we measured the seroconversion rates according to age and to the time of first serum sample collection (phase A, B or C). Criteria of seroconversion were defined in the method section. As shown in table 3, there was a sharp decline in seroconversion rates across all the age groups, depending on whether participants were enrolled during phase A, phase B, or phase C (P,0.0001). To interpret these data, one should remember that antibodies at seroprotective levels (HIA $1/40), in serum samples 1 collected during the per epidemic phase B or early post epidemic phase C could represent either base line cross reactive antibodies or rising pH1N1/2009 specific antibodies due to a recent or ongoing infection. This ambiguity could lead to underestimation of the seroconversion rate for subjects enrolled in phases B and C. In order to solve this ambiguity, we specifically considered the group of 249 subjects in whom cross reactive antibodies were detected at the time of phase A (W30-31). The seroconversion rate of this group is the most indicative of the exposure of individuals to the whole epidemic wave. It was the highest (63,2%, P,0.0001) in children and adolescents (,20 yrs), and still significantly high in adults (39.4%, P,0.0001).
We then tested in this particular group, the impact of (baseline) pre-epidemic cross reactive antibodies on the rate of seroconversion to pH1N1/2009 (Table 4) . No subject with HIA titer superior to 1/40 had evidence of seroconversion to pH1N1/2009. The seroconversion rate in individuals with a HIA titer equal to 1/40 was linked with age, being more important in children and adolescents (,20 yrs). The highest seroconversion rate (.56%) was registered in subjects with HIA titers inferior to 1/40, particularly for the under 20 years where it reached 85%. Hence, the risk of seroconversion decreased when pre-epidemic HIA titer was high after controlling for age (P,0.0001) (Figure 4) . The multivariate adjusted odds ratio for seroconversion were 0.15 (95%CI: 0.06-0.37, P,0.0001) per two-fold increase in baseline titer, 1.79 (95%CI: 1.23-2.59, P,0.003) per other household members who seroconverted, 5.33 (95%CI: 1.56-19.27, P,0.008) Figure 1 . The cohort profile and major outcomes. Figure 1 details the three phases of the protocol: i) inclusion (weeks 30-44) and serum samples S1 collection; ii) follow up for detection of ILI in households, qRT-PCR on nasal swabs and estimation of cumulative seroincidence rates; iii) end of the study (weeks 45-52) and samples S2 collection. HIA on paired sera (S1+S2) allowed estimating seroconversion rates. doi:10.1371/journal.pone.0025738.g001 Bp (baseline-proxy) seroprevalence rates were estimated on weeks 30-31 in each age group. b Cumulative incidence rates measured the raise between raw seroprevalence rates and age-specific baseline-proxy seroprevalence rate. In the group ''All ages'', cumulative incidence rates were standardized according to age structure of the community. doi:10.1371/journal.pone.0025738.t002 Data are numbers, percentages (95% confidence intervals) and ALR parameter test P value for comparison of seroconversion proportions according to time of first sample (S1) collection at inclusion, in each age group, after controlling for household selection. In the group ''All ages'', rates of seroconversion were standardized according to age structure of the community. NA: not assessed. Seroconversion was defined as a shift from seronegative at inclusion (i.e. HIA titer ,1/40) to seropositive on follow-up sample, or as a 4-fold increase of reciprocal HIA titer between first and second paired samples for sera tested seropositive on inclusion (i.e. HIA titer $1/40). for age ,20 years (vs age $60 years) and 11.35 (95%CI: 0.41-4.47, P = 0.62) for age 20-60 years (vs age $60 years). The observed and predicted seroconversion rates according to age and baseline HIA titer are displayed Figure 4 .
Finally, we considered the 46 subjects who had been infected by the pandemic virus over the course of the study, verified by a positive qRT-PCR nasal swab, and for whom paired sera were available. Initial HIA antibody titers in this group were ,1/40,
The CoPanFlu-RUN cohort was set up to conduct a prospective population-based study investigating the herd immunity induced by the 2009 pandemic influenza virus and identifying risk factors for pH1N1/2009v infection from paired sera collected in an entire community. Most works published to date have used either extensive cross-sectional serosurveys on pre-and post-epidemic independent serum samples, the baseline immunity being assessed from stored frozen samples [5, 7, 8] , or non representative adult cohorts (military, health care workers, long-stay patients). Antibody titers were measured by HIA using a cut-off value set at 1/40 as classically recommended. This HIA titer at 1/40 is considered protective, i.e. conferring 50% protection against a viral challenge [20] . Our assay has introduced some changes in the experimental protocol compared to the classic one. The use of a non-inactivated viral antigen, i.e. a native virus, with nondenatured epitopes probably allows detection of antibodies to epitopes of the hemagglutinin not detected in the classic HIA test. This can induce slight differences in the sensitivity of detection of cross-reacting antibodies, but this does not modify the kinetics of Ab and the epidemiological evolution of seroprevalence and does not jeopardize the global comparability of serological results. This is confirmed by the fact that our HI assay detected seroprotective antibody titers in 93.5% and gave evidence seroconversion in 73.9% of qRT-PCR confirmed pH1N1/2009 influenza, all figures close to those reported in the literature [5, 21] .
We considered that titers of .1/40, in sera collected from individuals enrolled during weeks 30 and 31 were cross reactive antibodies and not de novo antibodies triggered by the pandemic virus and hence used them as a proxy for baseline pre epidemic immunity. Several arguments support this assumption: i) the first case indicating autochthonous transmission in Reunion Island was reported by the epidemiological surveillance department of La Réunion on 21st July (week 30), i.e. the same day when inclusion started in our study cohort; ii) 7 to 15 days are required to develop an antibody response after viral infection; iii) On weeks 30 and 31, the epidemic activity due to the pandemic virus was very low in our study cohort and it became significant only after week 32. Hence, during weeks 30-31, 103 households were recruited and only 2 households reported ILI cases. Nasal swabs collected from these 2 individuals were tested qRT-PCR negative to the pandemic virus whereas one had evidence of coronavirus and rhinovirus using a multiplex RT-PCR to respiratory viruses (H. Pascalis, manuscript in preparation). In contrast, during weeks 32 to 39, 199 individuals belonging to 99 households reported ILI, among whom 60 individuals had documented infection by the pandemic virus.
Our study shows that a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus with the highest baseline-proxy seroprevalence rate observed among adults aged of 60 years or more. Other studies from all continents had also reported high pre-epidemic seropositivity rates among the elderly [5, 6, 8, [22] [23] [24] [25] [26] , though large variations do exist between countries [10, 11, 23, 27, 28] . These cross reactive antibodies have been interpreted as being the residual signature of the remote exposure of these individuals to H1N1 viruses circulating before 1957 [24, 25, 29, 30] . Baseline seropositivity rates that we report in children and in younger adults (i.e. 30%-35%) were notably higher than those reported from other parts of the world [6, 8, 22, 23, [31] [32] [33] . However one should note that these baseline antibodies were of low titer, just at the level of the HIA threshold (i.e. 1/40). Several factors could have contributed to this comparatively high baseline rates found in our study: i) It may reflect the fact that the HI test used in our study was marginally more sensitive than the classic one [17] ; ii) Some individuals may have already been infected with pH1N1/ 2009 virus at weeks 30 and 31 and may have triggered an antibody response to the virus. This hypothesis seems unlikely in view of the arguments presented above and of a similar high proportion of sera titering HIA = 1/40 among 122 sera from adult patients sent for diagnostic purposes to the Regional Hospital microbiology laboratory, during the first half of 2009 (i.e. before the 2009 pandemic) (data not shown). However we cannot formally exclude this hypothesis in view of a recently reported study from Taiwan [11] that showed evidence of subclinical community transmission with proved seroconversion several weeks before report of the first documented case in the island. A similar conclusion was also drawn from Australia [34] ; iii) our serological test might detect cross-reactive antibodies triggered by recent vaccination with trivalent seasonal influenza vaccine as reported [4, [35] [36] [37] [38] [39] . However, seasonal influenza vaccines were of rather limited use in Reunion Island, especially in children and young adults; iv) Finally the high baseline titers may reflect the infectious history of the individuals to seasonal influenza viruses cross antigenic with pH1N1/2009 virus as recently suggested for seasonal 2007 H1N1 infection [40] . This serosurvey indicates that a large fraction of the Reunion Island population was infected with the pandemic virus. Younger people, have paid the main tribute to the epidemic as almost two thirds show evidence of seroconversion, confirming earlier clinical reports from the island [12] and accumulating reports from other countries [17, 32, 41, 42] and suggesting that school children have likely played the central role in the epidemic diffusion of the pandemic virus. Lower infection rates were found in adults and the lowest rates were recorded in the elderly.
Based on clinical cases reported to the epidemiological surveillance services [12] , it was estimated that 66,915 persons in Reunion Island who consulted a physician were infected by the pH1N1/2009 virus during the 9 weeks of the epidemic, giving a cumulative attack rate of 8.26%. Taking into account those who did not consult a physician, the number of symptomatic infected persons was estimated to 104,067 (attack rate: 12.85%). In fact, the attack rate of pH1N1/2009 infection in our serosurvey was about 42%-44% at the peak of the antibody response (i.e., weeks 40-44), a figure which is at least 3 to 4 times higher than rates of infection based on clinical cases The wide gap between the two estimates indicates that a large fraction (almost two thirds) of those who got infected by pH1N1/2009 virus escaped medical detection, probably because they developed mild disease or asymptomatic infection, a further indication of the benign nature of the virus, at least at the community level. In England, Baguelin et al. [43] estimated that the cumulative incidence rates of infection by the pandemic virus in children were 20 to 40 times higher than that estimated from clinical surveillance.
Our study, as others [6] , indicates that pre-existing cross reactive antibodies to pH1N1/2009 at titers $1/40 prevented from seroconversion in response to the pandemic virus. This level of pre-existing cross reactive immunity likely confers true protection against infection as about two thirds and one third of documented infection (qRT-PCR positive) in our series have occurred in individuals with baseline HIA titers ,1/40 and = 1/ 40 respectively and less than 5% of documented infections occurred in individuals with base line titers .1/40. The protection was effective not only in older adults but also in younger persons. This indicates that protection was conferred not only by baseline cross reactive antibodies triggered by close pH1N1/2009 viruses that circulated before 1957 (as in the elderly), but also by antibodies likely resulting from recent exposure to seasonal influenza epidemics (as shown in younger persons) [40] . The observed seroconversion rates depend on age, after adjusting for baseline pH1N1/2009 titers. The protective role of increasing age might be explained by a stronger cross-immunity in adults and elderly or by a higher exposure of young subjects to the virus during the 2009 epidemic (due to social contacts and mixing patterns). It may also indicate that immune mechanisms other than cross reactive antibodies detected by HIA (i.e. immunity to neuraminidase and conserved T cells epitopes [44] might develop throughout life, providing additional protection from infection or severe disease, especially in the elderly. Interestingly, evidence is seen for a decline in antibody titers, which occurred soon after the passage of the epidemic wave. In paired sera, this decline was significant enough to bring, within a few weeks, almost 27% of sera that tested positive (i.e. HI titers $1/40) in the immediate post epidemic phase to levels under the cut-off value in the second serum sample. This decay accounts for the observation that older adults ($60 yrs) in the study cohort were apparently almost completely spared by the epidemic if one only considers cumulative incidence rates derived from IHA titration on samples 2 (weeks 45-52). In fact, the cumulative incidence rate in older adults measured just after the epidemic peak (i.e. weeks 40-44) was 20.4%. Similar results of early antibody decay were recently reported [10, 45] . More generally, these data show that serosurveys conducted months after passage of the epidemic, likely underestimate the real extent of pH1N1/2009 infection, compared to antibody titration performed earlier, when humoral responses are at their highest level. Whether the decline in antibody titers has functional immunologic consequence to individuals or within the communities warrants further investigation. However, one should note that there was no second epidemic wave in Reunion Island during the subsequent austral winter seasons in 2010 and 2011.
Influenza during the 2010 winter was at a level not higher than the usual passages of seasonal flu, though almost two thirds of documented cases in 2010 were also due to pH1N1/2009v [46] . In addition many fewer pandemic virus isolates were noted during the ongoing 2011 austral winter, strongly suggesting that the first epidemic wave had conferred a solid herd immunity, at the community level.
Our study has some limitations. The fact that the epidemic progression coincided with the implementation of the prospective study, we were not able to collect, strictly speaking, pre-epidemic sera from the cohort members. Therefore we used as proxy base line seroprevalence data from individuals recruited at the very beginning of the investigation when the epidemic activity in the cohort was very low. This may overestimate the base line immunity if subclinical community transmission had occurred before the first cases of pH1N1/2009 influenza were reported. Antibodies to the pandemic virus were detected by HIA, a test that has a good specificity but a rather low sensitivity [46] . Hence, the threshold of 1/40 may underestimate the number of infected individuals. However, rates of seroconversion, the serologic gold standard test based on paired sera, likely gave the most accurate picture of the pandemic in at the community level in Reunion Island.
|
When did WHO declare a pandemic of pH1N1/2009v influenza?
| 5,256 |
11 June 2009
| 2,887 |
1,601 |
Pandemic Influenza Due to pH1N1/2009 Virus: Estimation of Infection Burden in Reunion Island through a Prospective Serosurvey, Austral Winter 2009
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3183080/
SHA: ee6d70a53e3262cea6f85bd8b226f6b4c8b5f64b
Authors: Dellagi, Koussay; Rollot, Olivier; Temmam, Sarah; Salez, Nicolas; Guernier, Vanina; Pascalis, Hervé; Gérardin, Patrick; Fianu, Adrian; Lapidus, Nathanael; Naty, Nadège; Tortosa, Pablo; Boussaïd, Karim; Jaffar-Banjee, Marie-Christine; Filleul, Laurent; Flahault, Antoine; Carrat, Fabrice; Favier, Francois; de Lamballerie, Xavier
Date: 2011-09-29
DOI: 10.1371/journal.pone.0025738
License: cc-by
Abstract: BACKGROUND: To date, there is little information that reflects the true extent of spread of the pH1N1/2009v influenza pandemic at the community level as infection often results in mild or no clinical symptoms. This study aimed at assessing through a prospective study, the attack rate of pH1N1/2009 virus in Reunion Island and risk factors of infection, during the 2009 season. METHODOLOGY/PRINCIPAL FINDINGS: A serosurvey was conducted during the 2009 austral winter, in the frame of a prospective population study. Pairs of sera were collected from 1687 individuals belonging to 772 households, during and after passage of the pandemic wave. Antibodies to pH1N1/2009v were titered using the hemagglutination inhibition assay (HIA) with titers ≥1/40 being considered positive. Seroprevalence during the first two weeks of detection of pH1N1/2009v in Reunion Island was 29.8% in people under 20 years of age, 35.6% in adults (20–59 years) and 73.3% in the elderly (≥60 years) (P<0.0001). Baseline corrected cumulative incidence rates, were 42.9%, 13.9% and 0% in these age groups respectively (P<0.0001). A significant decline in antibody titers occurred soon after the passage of the epidemic wave. Seroconversion rates to pH1N1/2009 correlated negatively with age: 63.2%, 39.4% and 16.7%, in each age group respectively (P<0.0001). Seroconversion occurred in 65.2% of individuals who were seronegative at inclusion compared to 6.8% in those who were initially seropositive. CONCLUSIONS: Seroincidence of pH1N1/2009v infection was three times that estimated from clinical surveillance, indicating that almost two thirds of infections occurring at the community level have escaped medical detection. People under 20 years of age were the most affected group. Pre-epidemic titers ≥1/40 prevented seroconversion and are likely protective against infection. A concern was raised about the long term stability of the antibody responses.
Text: In April 2009, the first cases of acute respiratory infections caused by a novel triple-reassortant influenza virus, pH1N1/ 2009v, occurred in Mexico and the United States [1] . The rapid spread of infection to other continents led the World Health Organization (WHO) to declare on 11 June 2009 that a pandemic of pH1N1/2009v influenza was under way, which raised major international concern about the risk of high morbidity and lethality and the potential for severe socio-economic impact. Actually, the potential impact of this first third-millenium influenza pandemic has been revisited downwards as morbidity and case-fatality rates were less severe than initially anticipated [2] . Illness surveillance data do not allow to an accurate estimate of the true influenza infection rate, as a substantial proportion of infections are asymptomatic or mild [3] . Serological surveys can overcome this limitation, but must take into account that a significant proportion of the population that exhibited crossprotective antibody titers before circulation of the pH1N1/2009v [4] . This so-called ''baseline immunity'' has to be subtracted from the seroprevalence observed after the pandemic wave, to determine seroincidence in serosurveys [5] [6] [7] [8] . However, except for few studies [9] [10] [11] , most of these serosurveys did not use serial measurements in the same person, which allows for a better understanding of antibody kinetics and the dynamics of infection within individuals and communities.
Reunion Island (805,500 inhabitants) is a French overseas department located in the southwestern Indian Ocean, 700 km east of Madagascar and 200 km southwest of Mauritius. The first imported case of pH1N1/2009v was identified on 5 th July 2009 (week 29) in a traveller returning from Australia. The first case indicating community transmission was detected on 21 st July (week 30). pH1N1/2009v became the predominant circulating influenza virus within four weeks of its first detection, its activity peaked during week 35 (24) (25) (26) (27) (28) (29) (30) and ended at week 38 [12] . Contrary to initial fears, the health care system was not overwhelmed, as morbidity and mortality rates were lower than predicted [12] [13] [14] .
In order to assess at the community level, the actual magnitude of the pH1N1/2009v pandemic and the extent of the herd immunity acquired after passage of the epidemic wave, a prospective population serosurvey was conducted in Reunion Island during the passage of the epidemic wave in the 2009 austral winter season (July-December 2009): prevalence of infection was assessed on a weekly basis and seroconversion rates were measured using paired sera.
The CoPanFLu-RUN was part of the CoPanFLu international project, a consortium between the French National Institute of Health and Medical Research (INSERM), the Institute of Research for Development (IRD) and the Mérieux Fondation under the promotion of the School of Advanced Studies in Public Health (EHESP). To enable the rapid implementation of the study in anticipation of the imminent spread of the pandemic wave, we used a pre-existing sample of 2442 households established in October 2006 for the investigation of the Chikungunya outbreak (SEROCHIK) and updated in May 2008 throughout a follow-up telephone survey (TELECHIK) on a basis of 1148 households [15, 16] . We took special attention to select households representing a wide range of geographic locations in order to minimize the repartition bias.
The inclusion phase started on July 21 st (week 30) and was continued up to week 44, throughout the epidemic wave and beyond. A first serum sample (sample 1) was obtained from each household member. An active telephonic inquiry was then conducted twice a week to record symptoms compatible with influenza-like illness (ILI) occurring in households. Report of ILI (fever $37.8uC associated with any respiratory or systemic symptom) led to three consecutive visits of a nurse to the incident case-dwelling (on day 0, +3 and +8 post-report) to record symptoms and collect nasal swabs from all family members (for qRT-PCR detection of pH1N1/2009v. At week 45, the active inquiry was discontinued and a second (post-epidemic) serum sample (sample 2) was obtained (weeks 45-52) to determine seroconversion rates. Sera were aliquoted and stored at 280uC.
The protocol was conducted in accordance with the Declaration of Helsinki and French law for biomedical research (Nu ID RCB AFSSAPS: 2009-A00689-48) and was approved by the local Ethics Committee (Comité de Protection des Personnes of Bordeaux 2 University). Every eligible person for participation was asked for giving their written informed consent.
Viral genome detection by RT-PCR. Viral RNA was extracted from 140 mL of nasal swab eluate using the QIAamp Viral RNA kit (Qiagen) and processed for detection by TaqMan qRT-PCR targeting the heamagglutinin HA gene (SuperScript III Platinum one-step qRT-PCR system, Invitrogen) according to the recommendations of the Pasteur Institute (Van der Werf S. & Enouf V., SOP/FluA/130509). Confirmed pH1N1/2009v infection was defined as a positive qRT-PCR detection of the HA gene in at least one nasal swab.
Hemagglutination inhibition assay (HIA). A standard hemagglutination inhibition technique was adapted to detect and quantify pH1N1/2009v antibodies [17] . The antigen was prepared by diluting a non-inactivated cell culture supernatant producing a pdm H1N1v strain (strain OPYFLU-1 isolated from a young patient returning from Mexico in early May 2009) [18] . Briefly, the virus was propagated onto MDCK cells under standard conditions. The last passage (used for antigen preparation) was performed in the absence of trypsin and ht-FBS. The supernatant was collected at day seven p.i. clarified by centrifugation at 8006 g for 10 min at room temperature, aliquoted and conserved at 280uC. The hemagglutinating titer of the non inactivated viral antigen was immediately determined under the HIA format described below. The dilution providing 5.33 hemagglutinating units in a volume of 25 mL was used for subsequent HIA. Sera were heat-inactivated at 56uC for 30 min prior to use. Sequential twofold dilutions in PBS (1/10 to 1/1280) in volumes of 25 mL were performed and distributed in V-bottom 96 well microplates. Human red blood cells (RBC) were used for hemagglutination experiments. Detection and quantification of antibody to pH1N1/2009v was performed as follows: 25 mL of virus suspension was added to the serum dilution (25 mL) and incubated for 1 hour at room temperature. Each well was then filled with 25 mL of a 1% RBC suspension in PBS (v/v: 0.33%), followed by another 30 min incubation at room temperature. The HIA titer was determined as the last dilution providing clear inhibition of hemagglutination. All experiments were performed in the presence of the same negative and positive controls, the latter including sera with 1/40, 1/80, 1/160 and 1/320 antibody titers.
The results reported in this study were based only on serological analysis of paired sera. For the sake of analysis, four successive phases were identified throughout the pandemic wave: phase A (weeks 30-31) corresponded to early epidemic time, phase B (W32-39) to the epidemic unfolding, phase C (W40-44) to the immediate post-epidemic stage and phase D (W45-52) to the late post-epidemic stage. Seropositivity was defined as a HIA titer of 1/ 40 or more. The baseline-proxy seroprevalence rate was estimated on serum samples collected in phase A. The cumulative incidence rate of infection measured the raise between the raw seroprevalence rate at any given time during the epidemic phases (S2pi) and the age-specific baseline-proxy seroprevalence rate (S1pA) (s2 pi -s1 pA ). Seroconversion was defined as a shift from seronegative at inclusion (sample 1: HIA ,1/40) to seropositive on follow-up (sample 2: HIA $1/40), or for sera tested seropositive on inclusion as a four-fold increase of HIA titers between sample 1 and sample 2 paired sera. We also calculated the proportion of sera that tested seropositive in sample 1 for which the HIA titer decreased fourfold and passed under the cut-off value of 1/40 in sample 2. We considered this proportion as a ''seronegation'' rate.
The sample size was calculated for identifying risk factors in the prospective cohort study. Considering on average three individuals per household, an intra-household correlation of 0.3, a power greater than 80% could be obtained with a sample size of 840 comprising 2500 individuals, assuming exposure levels ranging from 10% to 90% and a relative risk greater than 1.3. With 2,500 subjects, the study allowed 1-2% absolute precision around the estimated values for seroconversion rates. Data entry used EpiData version 3.1 (The Epidata Association, Odense, Denmark). SAS version 9.1 (SAS Inc., Cary, NC, USA) was used for statistical analysis. The characteristics of the study cohort were compared to those of the population of Reunion Island and a Chi2 test (or Fisher's exact test when non applicable) was used to analyse differences in age, sex and geographic location. Cumulative incidence rates of infection (i.e. seroincidence) and seroconversion rates were standardized according to the age structure of the community (French National Institute for Statistics and Economical Studies (INSEE) source).
Baseline-proxy seroprevalence, cumulative incidence rates of infection, as well as seroconversion and seronegation rates, were expressed as percentages. Cumulative reverse distribution curves were used to show the distribution of antibody titers. In all tests, a P value,0.05 was considered significant.
We estimated 95% confidence intervals (CIs) of proportions by using a cluster bootstrap technique with 1000 re-samples [19] . After bootstraping, we used an ANOVA model to compare mean cumulative incidence proportions between pandemic phases, within each age group. We used an alternating logistic regression model (ALR) with an exchangeable log Odds Ratio (OR) to test the intra-household correlation-adjusted association between factors and the seroconversion outcome.
Data were analysed with respect to subject age. Initially, four age groups were considered: the children and adolescents (,20 yrs), young adults (20-39 yrs), middle-age adults (40-59 yrs), and elderly adults ($60 yrs). As the cumulative incidence of infection of the second and third groups were very close, both groups were merged into one adults group (20-59 yrs). Therefore we refer further in our study to three age groups: children and adolescents (,20 yrs), adults (20-59 yrs), elderly ($60 yrs).
A total of 2,164 individuals from 772 households were enrolled between weeks 30 and 44 in the CoPanFlu-RUN cohort, allowing the collection of 1,932 sera at inclusion (sample 1). During this period, 136 households (17.7% of households) containing 464 individuals (21.4% of individuals) reported at least one case of ILI. Sixty subjects among the 464 individuals (12.9%, belonging to 33 households [24.3%]) were qRT-PCR positive, which documented the pH1N1/2009v infection. No positive qRT-PCR could be detected after week 37 and no ILI was reported after week 40, the end of the epidemic wave. The second follow up serum sample (sample 2) was obtained for 1,759 subjects at least five weeks after the end of the epidemic wave (weeks 45-52) which allowed the constitution of a serobank of 1,687 paired-sera. The profile of the cohort and the major outcomes are displayed in Figure 1 . Details on inclusions and serum sample timing with respect to the circulation of pH1N1/2009v over the island are provided in figure 2 . The socio-demographic and space-time characteristics of the cohort are detailed in Table 1 . Compared to the community of Reunion Island, the sample of 1,687 individuals for whom pairedsera were available, was older (,20 yrs: 27% vs 35%, and $60 yrs: 17,9% vs 11,3%) and composed of a slight excess of females (54.1% vs 51.5%). The imbalance was due to a deficit in subjects aged under 40 years, reflecting men at work and the fact that parents declined the second serum for children younger than five.
Baseline-proxy (,pre-epidemic) HIA titers to the pH1N1/ 2009v were measured on sample 1 ( Table 2) , obtained from 249 subjects (103 households) recruited at the very beginning of the investigation during weeks 30 and 31 (phase A, Figure 2 ), when the epidemic activity in the cohort was still very low. Age distribution in this group was similar to that of the whole cohort (data not shown). The overall, the baseline-proxy seroprevalence rate (HIA $1/40), over all ages, was 43.4% (95%CI: 37.4%-49.6%). However the majority of positive sera had low antibody titers, at the cut off value for confirmation (i.e. = 1/40). The proportions of sera with HIA titer .1/40 were 0%, 3.0% and 24.6% in the young, middle-aged and older age groups respectively. These results indicate that pre-epidemic baseline antibody cross reactivity was stronger in the elderly ($60 yrs) and weaker in children and adolescents (,20 yrs) and adults (20-59 yrs), with highly significant differences between age groups (P,0.0001).
The reverse cumulative distribution curves of HIA titers are displayed for each age group and for the whole cohort on Figure 3 . The proportion of seropositive sera (HI $1/40) steadily increased during the epidemic unfolding (phase B, W32-39) and in immediate post epidemic period (phase C, W40-44) when it reached its maximum level, then declined in the late post epidemic period (phase D, W45-52). This decline was significant enough to return the reverse cumulative distribution curve to baseline levels in the elderly. The cumulative incidence rates, obtained after subtraction of the age-specific baseline-proxy seroprevalence from the raw seroprevalence at each phase of the epidemic are shown in Table 2 (note that the cumulative incidence rates of infection represented for the group ''all ages'' were standardized according to age structure of the community). The cumulative incidence rates were much higher in children and adolescents (,20 yrs), indicating very active transmission of infection within this age group. As mentioned earlier, cumulative incidence rates peaked in phase C (W40-44), and then declined indicating some lability of the humoral immune response against the pH1N1/2009v. The age-related difference observed in the incidence rates was highly statistically significant (P,0.0001).
To estimate more appropriately the decline of antibody titers occurring after the peak of the humoral response to the pH1N1/ 2009v, we considered paired-sera from the group of 264 subjects for whom the first serum sample (sample 1) was obtained just after the epidemic wave (phase C, W40-44), and the corresponding second sample was collected at the end of the survey (phase D, W45-52). Seronegation rates were 27.0% (61/226) for all age groups, 17.4% (12/69) in children and adolescents (,20 yrs), 32.3% (41/127) in adults (20-59 yrs) and 26.7% (8/30) in the elderly ($60 yrs). Differences between the seronegation rates according to age were statistically weakly significant (P = 0.0671).
We then considered the 1687 individuals for whom paired sera were available and we measured the seroconversion rates according to age and to the time of first serum sample collection (phase A, B or C). Criteria of seroconversion were defined in the method section. As shown in table 3, there was a sharp decline in seroconversion rates across all the age groups, depending on whether participants were enrolled during phase A, phase B, or phase C (P,0.0001). To interpret these data, one should remember that antibodies at seroprotective levels (HIA $1/40), in serum samples 1 collected during the per epidemic phase B or early post epidemic phase C could represent either base line cross reactive antibodies or rising pH1N1/2009 specific antibodies due to a recent or ongoing infection. This ambiguity could lead to underestimation of the seroconversion rate for subjects enrolled in phases B and C. In order to solve this ambiguity, we specifically considered the group of 249 subjects in whom cross reactive antibodies were detected at the time of phase A (W30-31). The seroconversion rate of this group is the most indicative of the exposure of individuals to the whole epidemic wave. It was the highest (63,2%, P,0.0001) in children and adolescents (,20 yrs), and still significantly high in adults (39.4%, P,0.0001).
We then tested in this particular group, the impact of (baseline) pre-epidemic cross reactive antibodies on the rate of seroconversion to pH1N1/2009 (Table 4) . No subject with HIA titer superior to 1/40 had evidence of seroconversion to pH1N1/2009. The seroconversion rate in individuals with a HIA titer equal to 1/40 was linked with age, being more important in children and adolescents (,20 yrs). The highest seroconversion rate (.56%) was registered in subjects with HIA titers inferior to 1/40, particularly for the under 20 years where it reached 85%. Hence, the risk of seroconversion decreased when pre-epidemic HIA titer was high after controlling for age (P,0.0001) (Figure 4) . The multivariate adjusted odds ratio for seroconversion were 0.15 (95%CI: 0.06-0.37, P,0.0001) per two-fold increase in baseline titer, 1.79 (95%CI: 1.23-2.59, P,0.003) per other household members who seroconverted, 5.33 (95%CI: 1.56-19.27, P,0.008) Figure 1 . The cohort profile and major outcomes. Figure 1 details the three phases of the protocol: i) inclusion (weeks 30-44) and serum samples S1 collection; ii) follow up for detection of ILI in households, qRT-PCR on nasal swabs and estimation of cumulative seroincidence rates; iii) end of the study (weeks 45-52) and samples S2 collection. HIA on paired sera (S1+S2) allowed estimating seroconversion rates. doi:10.1371/journal.pone.0025738.g001 Bp (baseline-proxy) seroprevalence rates were estimated on weeks 30-31 in each age group. b Cumulative incidence rates measured the raise between raw seroprevalence rates and age-specific baseline-proxy seroprevalence rate. In the group ''All ages'', cumulative incidence rates were standardized according to age structure of the community. doi:10.1371/journal.pone.0025738.t002 Data are numbers, percentages (95% confidence intervals) and ALR parameter test P value for comparison of seroconversion proportions according to time of first sample (S1) collection at inclusion, in each age group, after controlling for household selection. In the group ''All ages'', rates of seroconversion were standardized according to age structure of the community. NA: not assessed. Seroconversion was defined as a shift from seronegative at inclusion (i.e. HIA titer ,1/40) to seropositive on follow-up sample, or as a 4-fold increase of reciprocal HIA titer between first and second paired samples for sera tested seropositive on inclusion (i.e. HIA titer $1/40). for age ,20 years (vs age $60 years) and 11.35 (95%CI: 0.41-4.47, P = 0.62) for age 20-60 years (vs age $60 years). The observed and predicted seroconversion rates according to age and baseline HIA titer are displayed Figure 4 .
Finally, we considered the 46 subjects who had been infected by the pandemic virus over the course of the study, verified by a positive qRT-PCR nasal swab, and for whom paired sera were available. Initial HIA antibody titers in this group were ,1/40,
The CoPanFlu-RUN cohort was set up to conduct a prospective population-based study investigating the herd immunity induced by the 2009 pandemic influenza virus and identifying risk factors for pH1N1/2009v infection from paired sera collected in an entire community. Most works published to date have used either extensive cross-sectional serosurveys on pre-and post-epidemic independent serum samples, the baseline immunity being assessed from stored frozen samples [5, 7, 8] , or non representative adult cohorts (military, health care workers, long-stay patients). Antibody titers were measured by HIA using a cut-off value set at 1/40 as classically recommended. This HIA titer at 1/40 is considered protective, i.e. conferring 50% protection against a viral challenge [20] . Our assay has introduced some changes in the experimental protocol compared to the classic one. The use of a non-inactivated viral antigen, i.e. a native virus, with nondenatured epitopes probably allows detection of antibodies to epitopes of the hemagglutinin not detected in the classic HIA test. This can induce slight differences in the sensitivity of detection of cross-reacting antibodies, but this does not modify the kinetics of Ab and the epidemiological evolution of seroprevalence and does not jeopardize the global comparability of serological results. This is confirmed by the fact that our HI assay detected seroprotective antibody titers in 93.5% and gave evidence seroconversion in 73.9% of qRT-PCR confirmed pH1N1/2009 influenza, all figures close to those reported in the literature [5, 21] .
We considered that titers of .1/40, in sera collected from individuals enrolled during weeks 30 and 31 were cross reactive antibodies and not de novo antibodies triggered by the pandemic virus and hence used them as a proxy for baseline pre epidemic immunity. Several arguments support this assumption: i) the first case indicating autochthonous transmission in Reunion Island was reported by the epidemiological surveillance department of La Réunion on 21st July (week 30), i.e. the same day when inclusion started in our study cohort; ii) 7 to 15 days are required to develop an antibody response after viral infection; iii) On weeks 30 and 31, the epidemic activity due to the pandemic virus was very low in our study cohort and it became significant only after week 32. Hence, during weeks 30-31, 103 households were recruited and only 2 households reported ILI cases. Nasal swabs collected from these 2 individuals were tested qRT-PCR negative to the pandemic virus whereas one had evidence of coronavirus and rhinovirus using a multiplex RT-PCR to respiratory viruses (H. Pascalis, manuscript in preparation). In contrast, during weeks 32 to 39, 199 individuals belonging to 99 households reported ILI, among whom 60 individuals had documented infection by the pandemic virus.
Our study shows that a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus with the highest baseline-proxy seroprevalence rate observed among adults aged of 60 years or more. Other studies from all continents had also reported high pre-epidemic seropositivity rates among the elderly [5, 6, 8, [22] [23] [24] [25] [26] , though large variations do exist between countries [10, 11, 23, 27, 28] . These cross reactive antibodies have been interpreted as being the residual signature of the remote exposure of these individuals to H1N1 viruses circulating before 1957 [24, 25, 29, 30] . Baseline seropositivity rates that we report in children and in younger adults (i.e. 30%-35%) were notably higher than those reported from other parts of the world [6, 8, 22, 23, [31] [32] [33] . However one should note that these baseline antibodies were of low titer, just at the level of the HIA threshold (i.e. 1/40). Several factors could have contributed to this comparatively high baseline rates found in our study: i) It may reflect the fact that the HI test used in our study was marginally more sensitive than the classic one [17] ; ii) Some individuals may have already been infected with pH1N1/ 2009 virus at weeks 30 and 31 and may have triggered an antibody response to the virus. This hypothesis seems unlikely in view of the arguments presented above and of a similar high proportion of sera titering HIA = 1/40 among 122 sera from adult patients sent for diagnostic purposes to the Regional Hospital microbiology laboratory, during the first half of 2009 (i.e. before the 2009 pandemic) (data not shown). However we cannot formally exclude this hypothesis in view of a recently reported study from Taiwan [11] that showed evidence of subclinical community transmission with proved seroconversion several weeks before report of the first documented case in the island. A similar conclusion was also drawn from Australia [34] ; iii) our serological test might detect cross-reactive antibodies triggered by recent vaccination with trivalent seasonal influenza vaccine as reported [4, [35] [36] [37] [38] [39] . However, seasonal influenza vaccines were of rather limited use in Reunion Island, especially in children and young adults; iv) Finally the high baseline titers may reflect the infectious history of the individuals to seasonal influenza viruses cross antigenic with pH1N1/2009 virus as recently suggested for seasonal 2007 H1N1 infection [40] . This serosurvey indicates that a large fraction of the Reunion Island population was infected with the pandemic virus. Younger people, have paid the main tribute to the epidemic as almost two thirds show evidence of seroconversion, confirming earlier clinical reports from the island [12] and accumulating reports from other countries [17, 32, 41, 42] and suggesting that school children have likely played the central role in the epidemic diffusion of the pandemic virus. Lower infection rates were found in adults and the lowest rates were recorded in the elderly.
Based on clinical cases reported to the epidemiological surveillance services [12] , it was estimated that 66,915 persons in Reunion Island who consulted a physician were infected by the pH1N1/2009 virus during the 9 weeks of the epidemic, giving a cumulative attack rate of 8.26%. Taking into account those who did not consult a physician, the number of symptomatic infected persons was estimated to 104,067 (attack rate: 12.85%). In fact, the attack rate of pH1N1/2009 infection in our serosurvey was about 42%-44% at the peak of the antibody response (i.e., weeks 40-44), a figure which is at least 3 to 4 times higher than rates of infection based on clinical cases The wide gap between the two estimates indicates that a large fraction (almost two thirds) of those who got infected by pH1N1/2009 virus escaped medical detection, probably because they developed mild disease or asymptomatic infection, a further indication of the benign nature of the virus, at least at the community level. In England, Baguelin et al. [43] estimated that the cumulative incidence rates of infection by the pandemic virus in children were 20 to 40 times higher than that estimated from clinical surveillance.
Our study, as others [6] , indicates that pre-existing cross reactive antibodies to pH1N1/2009 at titers $1/40 prevented from seroconversion in response to the pandemic virus. This level of pre-existing cross reactive immunity likely confers true protection against infection as about two thirds and one third of documented infection (qRT-PCR positive) in our series have occurred in individuals with baseline HIA titers ,1/40 and = 1/ 40 respectively and less than 5% of documented infections occurred in individuals with base line titers .1/40. The protection was effective not only in older adults but also in younger persons. This indicates that protection was conferred not only by baseline cross reactive antibodies triggered by close pH1N1/2009 viruses that circulated before 1957 (as in the elderly), but also by antibodies likely resulting from recent exposure to seasonal influenza epidemics (as shown in younger persons) [40] . The observed seroconversion rates depend on age, after adjusting for baseline pH1N1/2009 titers. The protective role of increasing age might be explained by a stronger cross-immunity in adults and elderly or by a higher exposure of young subjects to the virus during the 2009 epidemic (due to social contacts and mixing patterns). It may also indicate that immune mechanisms other than cross reactive antibodies detected by HIA (i.e. immunity to neuraminidase and conserved T cells epitopes [44] might develop throughout life, providing additional protection from infection or severe disease, especially in the elderly. Interestingly, evidence is seen for a decline in antibody titers, which occurred soon after the passage of the epidemic wave. In paired sera, this decline was significant enough to bring, within a few weeks, almost 27% of sera that tested positive (i.e. HI titers $1/40) in the immediate post epidemic phase to levels under the cut-off value in the second serum sample. This decay accounts for the observation that older adults ($60 yrs) in the study cohort were apparently almost completely spared by the epidemic if one only considers cumulative incidence rates derived from IHA titration on samples 2 (weeks 45-52). In fact, the cumulative incidence rate in older adults measured just after the epidemic peak (i.e. weeks 40-44) was 20.4%. Similar results of early antibody decay were recently reported [10, 45] . More generally, these data show that serosurveys conducted months after passage of the epidemic, likely underestimate the real extent of pH1N1/2009 infection, compared to antibody titration performed earlier, when humoral responses are at their highest level. Whether the decline in antibody titers has functional immunologic consequence to individuals or within the communities warrants further investigation. However, one should note that there was no second epidemic wave in Reunion Island during the subsequent austral winter seasons in 2010 and 2011.
Influenza during the 2010 winter was at a level not higher than the usual passages of seasonal flu, though almost two thirds of documented cases in 2010 were also due to pH1N1/2009v [46] . In addition many fewer pandemic virus isolates were noted during the ongoing 2011 austral winter, strongly suggesting that the first epidemic wave had conferred a solid herd immunity, at the community level.
Our study has some limitations. The fact that the epidemic progression coincided with the implementation of the prospective study, we were not able to collect, strictly speaking, pre-epidemic sera from the cohort members. Therefore we used as proxy base line seroprevalence data from individuals recruited at the very beginning of the investigation when the epidemic activity in the cohort was very low. This may overestimate the base line immunity if subclinical community transmission had occurred before the first cases of pH1N1/2009 influenza were reported. Antibodies to the pandemic virus were detected by HIA, a test that has a good specificity but a rather low sensitivity [46] . Hence, the threshold of 1/40 may underestimate the number of infected individuals. However, rates of seroconversion, the serologic gold standard test based on paired sera, likely gave the most accurate picture of the pandemic in at the community level in Reunion Island.
|
What is the classical cutoff value for antibody titers?
| 5,257 |
1/40
| 22,699 |
1,601 |
Pandemic Influenza Due to pH1N1/2009 Virus: Estimation of Infection Burden in Reunion Island through a Prospective Serosurvey, Austral Winter 2009
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3183080/
SHA: ee6d70a53e3262cea6f85bd8b226f6b4c8b5f64b
Authors: Dellagi, Koussay; Rollot, Olivier; Temmam, Sarah; Salez, Nicolas; Guernier, Vanina; Pascalis, Hervé; Gérardin, Patrick; Fianu, Adrian; Lapidus, Nathanael; Naty, Nadège; Tortosa, Pablo; Boussaïd, Karim; Jaffar-Banjee, Marie-Christine; Filleul, Laurent; Flahault, Antoine; Carrat, Fabrice; Favier, Francois; de Lamballerie, Xavier
Date: 2011-09-29
DOI: 10.1371/journal.pone.0025738
License: cc-by
Abstract: BACKGROUND: To date, there is little information that reflects the true extent of spread of the pH1N1/2009v influenza pandemic at the community level as infection often results in mild or no clinical symptoms. This study aimed at assessing through a prospective study, the attack rate of pH1N1/2009 virus in Reunion Island and risk factors of infection, during the 2009 season. METHODOLOGY/PRINCIPAL FINDINGS: A serosurvey was conducted during the 2009 austral winter, in the frame of a prospective population study. Pairs of sera were collected from 1687 individuals belonging to 772 households, during and after passage of the pandemic wave. Antibodies to pH1N1/2009v were titered using the hemagglutination inhibition assay (HIA) with titers ≥1/40 being considered positive. Seroprevalence during the first two weeks of detection of pH1N1/2009v in Reunion Island was 29.8% in people under 20 years of age, 35.6% in adults (20–59 years) and 73.3% in the elderly (≥60 years) (P<0.0001). Baseline corrected cumulative incidence rates, were 42.9%, 13.9% and 0% in these age groups respectively (P<0.0001). A significant decline in antibody titers occurred soon after the passage of the epidemic wave. Seroconversion rates to pH1N1/2009 correlated negatively with age: 63.2%, 39.4% and 16.7%, in each age group respectively (P<0.0001). Seroconversion occurred in 65.2% of individuals who were seronegative at inclusion compared to 6.8% in those who were initially seropositive. CONCLUSIONS: Seroincidence of pH1N1/2009v infection was three times that estimated from clinical surveillance, indicating that almost two thirds of infections occurring at the community level have escaped medical detection. People under 20 years of age were the most affected group. Pre-epidemic titers ≥1/40 prevented seroconversion and are likely protective against infection. A concern was raised about the long term stability of the antibody responses.
Text: In April 2009, the first cases of acute respiratory infections caused by a novel triple-reassortant influenza virus, pH1N1/ 2009v, occurred in Mexico and the United States [1] . The rapid spread of infection to other continents led the World Health Organization (WHO) to declare on 11 June 2009 that a pandemic of pH1N1/2009v influenza was under way, which raised major international concern about the risk of high morbidity and lethality and the potential for severe socio-economic impact. Actually, the potential impact of this first third-millenium influenza pandemic has been revisited downwards as morbidity and case-fatality rates were less severe than initially anticipated [2] . Illness surveillance data do not allow to an accurate estimate of the true influenza infection rate, as a substantial proportion of infections are asymptomatic or mild [3] . Serological surveys can overcome this limitation, but must take into account that a significant proportion of the population that exhibited crossprotective antibody titers before circulation of the pH1N1/2009v [4] . This so-called ''baseline immunity'' has to be subtracted from the seroprevalence observed after the pandemic wave, to determine seroincidence in serosurveys [5] [6] [7] [8] . However, except for few studies [9] [10] [11] , most of these serosurveys did not use serial measurements in the same person, which allows for a better understanding of antibody kinetics and the dynamics of infection within individuals and communities.
Reunion Island (805,500 inhabitants) is a French overseas department located in the southwestern Indian Ocean, 700 km east of Madagascar and 200 km southwest of Mauritius. The first imported case of pH1N1/2009v was identified on 5 th July 2009 (week 29) in a traveller returning from Australia. The first case indicating community transmission was detected on 21 st July (week 30). pH1N1/2009v became the predominant circulating influenza virus within four weeks of its first detection, its activity peaked during week 35 (24) (25) (26) (27) (28) (29) (30) and ended at week 38 [12] . Contrary to initial fears, the health care system was not overwhelmed, as morbidity and mortality rates were lower than predicted [12] [13] [14] .
In order to assess at the community level, the actual magnitude of the pH1N1/2009v pandemic and the extent of the herd immunity acquired after passage of the epidemic wave, a prospective population serosurvey was conducted in Reunion Island during the passage of the epidemic wave in the 2009 austral winter season (July-December 2009): prevalence of infection was assessed on a weekly basis and seroconversion rates were measured using paired sera.
The CoPanFLu-RUN was part of the CoPanFLu international project, a consortium between the French National Institute of Health and Medical Research (INSERM), the Institute of Research for Development (IRD) and the Mérieux Fondation under the promotion of the School of Advanced Studies in Public Health (EHESP). To enable the rapid implementation of the study in anticipation of the imminent spread of the pandemic wave, we used a pre-existing sample of 2442 households established in October 2006 for the investigation of the Chikungunya outbreak (SEROCHIK) and updated in May 2008 throughout a follow-up telephone survey (TELECHIK) on a basis of 1148 households [15, 16] . We took special attention to select households representing a wide range of geographic locations in order to minimize the repartition bias.
The inclusion phase started on July 21 st (week 30) and was continued up to week 44, throughout the epidemic wave and beyond. A first serum sample (sample 1) was obtained from each household member. An active telephonic inquiry was then conducted twice a week to record symptoms compatible with influenza-like illness (ILI) occurring in households. Report of ILI (fever $37.8uC associated with any respiratory or systemic symptom) led to three consecutive visits of a nurse to the incident case-dwelling (on day 0, +3 and +8 post-report) to record symptoms and collect nasal swabs from all family members (for qRT-PCR detection of pH1N1/2009v. At week 45, the active inquiry was discontinued and a second (post-epidemic) serum sample (sample 2) was obtained (weeks 45-52) to determine seroconversion rates. Sera were aliquoted and stored at 280uC.
The protocol was conducted in accordance with the Declaration of Helsinki and French law for biomedical research (Nu ID RCB AFSSAPS: 2009-A00689-48) and was approved by the local Ethics Committee (Comité de Protection des Personnes of Bordeaux 2 University). Every eligible person for participation was asked for giving their written informed consent.
Viral genome detection by RT-PCR. Viral RNA was extracted from 140 mL of nasal swab eluate using the QIAamp Viral RNA kit (Qiagen) and processed for detection by TaqMan qRT-PCR targeting the heamagglutinin HA gene (SuperScript III Platinum one-step qRT-PCR system, Invitrogen) according to the recommendations of the Pasteur Institute (Van der Werf S. & Enouf V., SOP/FluA/130509). Confirmed pH1N1/2009v infection was defined as a positive qRT-PCR detection of the HA gene in at least one nasal swab.
Hemagglutination inhibition assay (HIA). A standard hemagglutination inhibition technique was adapted to detect and quantify pH1N1/2009v antibodies [17] . The antigen was prepared by diluting a non-inactivated cell culture supernatant producing a pdm H1N1v strain (strain OPYFLU-1 isolated from a young patient returning from Mexico in early May 2009) [18] . Briefly, the virus was propagated onto MDCK cells under standard conditions. The last passage (used for antigen preparation) was performed in the absence of trypsin and ht-FBS. The supernatant was collected at day seven p.i. clarified by centrifugation at 8006 g for 10 min at room temperature, aliquoted and conserved at 280uC. The hemagglutinating titer of the non inactivated viral antigen was immediately determined under the HIA format described below. The dilution providing 5.33 hemagglutinating units in a volume of 25 mL was used for subsequent HIA. Sera were heat-inactivated at 56uC for 30 min prior to use. Sequential twofold dilutions in PBS (1/10 to 1/1280) in volumes of 25 mL were performed and distributed in V-bottom 96 well microplates. Human red blood cells (RBC) were used for hemagglutination experiments. Detection and quantification of antibody to pH1N1/2009v was performed as follows: 25 mL of virus suspension was added to the serum dilution (25 mL) and incubated for 1 hour at room temperature. Each well was then filled with 25 mL of a 1% RBC suspension in PBS (v/v: 0.33%), followed by another 30 min incubation at room temperature. The HIA titer was determined as the last dilution providing clear inhibition of hemagglutination. All experiments were performed in the presence of the same negative and positive controls, the latter including sera with 1/40, 1/80, 1/160 and 1/320 antibody titers.
The results reported in this study were based only on serological analysis of paired sera. For the sake of analysis, four successive phases were identified throughout the pandemic wave: phase A (weeks 30-31) corresponded to early epidemic time, phase B (W32-39) to the epidemic unfolding, phase C (W40-44) to the immediate post-epidemic stage and phase D (W45-52) to the late post-epidemic stage. Seropositivity was defined as a HIA titer of 1/ 40 or more. The baseline-proxy seroprevalence rate was estimated on serum samples collected in phase A. The cumulative incidence rate of infection measured the raise between the raw seroprevalence rate at any given time during the epidemic phases (S2pi) and the age-specific baseline-proxy seroprevalence rate (S1pA) (s2 pi -s1 pA ). Seroconversion was defined as a shift from seronegative at inclusion (sample 1: HIA ,1/40) to seropositive on follow-up (sample 2: HIA $1/40), or for sera tested seropositive on inclusion as a four-fold increase of HIA titers between sample 1 and sample 2 paired sera. We also calculated the proportion of sera that tested seropositive in sample 1 for which the HIA titer decreased fourfold and passed under the cut-off value of 1/40 in sample 2. We considered this proportion as a ''seronegation'' rate.
The sample size was calculated for identifying risk factors in the prospective cohort study. Considering on average three individuals per household, an intra-household correlation of 0.3, a power greater than 80% could be obtained with a sample size of 840 comprising 2500 individuals, assuming exposure levels ranging from 10% to 90% and a relative risk greater than 1.3. With 2,500 subjects, the study allowed 1-2% absolute precision around the estimated values for seroconversion rates. Data entry used EpiData version 3.1 (The Epidata Association, Odense, Denmark). SAS version 9.1 (SAS Inc., Cary, NC, USA) was used for statistical analysis. The characteristics of the study cohort were compared to those of the population of Reunion Island and a Chi2 test (or Fisher's exact test when non applicable) was used to analyse differences in age, sex and geographic location. Cumulative incidence rates of infection (i.e. seroincidence) and seroconversion rates were standardized according to the age structure of the community (French National Institute for Statistics and Economical Studies (INSEE) source).
Baseline-proxy seroprevalence, cumulative incidence rates of infection, as well as seroconversion and seronegation rates, were expressed as percentages. Cumulative reverse distribution curves were used to show the distribution of antibody titers. In all tests, a P value,0.05 was considered significant.
We estimated 95% confidence intervals (CIs) of proportions by using a cluster bootstrap technique with 1000 re-samples [19] . After bootstraping, we used an ANOVA model to compare mean cumulative incidence proportions between pandemic phases, within each age group. We used an alternating logistic regression model (ALR) with an exchangeable log Odds Ratio (OR) to test the intra-household correlation-adjusted association between factors and the seroconversion outcome.
Data were analysed with respect to subject age. Initially, four age groups were considered: the children and adolescents (,20 yrs), young adults (20-39 yrs), middle-age adults (40-59 yrs), and elderly adults ($60 yrs). As the cumulative incidence of infection of the second and third groups were very close, both groups were merged into one adults group (20-59 yrs). Therefore we refer further in our study to three age groups: children and adolescents (,20 yrs), adults (20-59 yrs), elderly ($60 yrs).
A total of 2,164 individuals from 772 households were enrolled between weeks 30 and 44 in the CoPanFlu-RUN cohort, allowing the collection of 1,932 sera at inclusion (sample 1). During this period, 136 households (17.7% of households) containing 464 individuals (21.4% of individuals) reported at least one case of ILI. Sixty subjects among the 464 individuals (12.9%, belonging to 33 households [24.3%]) were qRT-PCR positive, which documented the pH1N1/2009v infection. No positive qRT-PCR could be detected after week 37 and no ILI was reported after week 40, the end of the epidemic wave. The second follow up serum sample (sample 2) was obtained for 1,759 subjects at least five weeks after the end of the epidemic wave (weeks 45-52) which allowed the constitution of a serobank of 1,687 paired-sera. The profile of the cohort and the major outcomes are displayed in Figure 1 . Details on inclusions and serum sample timing with respect to the circulation of pH1N1/2009v over the island are provided in figure 2 . The socio-demographic and space-time characteristics of the cohort are detailed in Table 1 . Compared to the community of Reunion Island, the sample of 1,687 individuals for whom pairedsera were available, was older (,20 yrs: 27% vs 35%, and $60 yrs: 17,9% vs 11,3%) and composed of a slight excess of females (54.1% vs 51.5%). The imbalance was due to a deficit in subjects aged under 40 years, reflecting men at work and the fact that parents declined the second serum for children younger than five.
Baseline-proxy (,pre-epidemic) HIA titers to the pH1N1/ 2009v were measured on sample 1 ( Table 2) , obtained from 249 subjects (103 households) recruited at the very beginning of the investigation during weeks 30 and 31 (phase A, Figure 2 ), when the epidemic activity in the cohort was still very low. Age distribution in this group was similar to that of the whole cohort (data not shown). The overall, the baseline-proxy seroprevalence rate (HIA $1/40), over all ages, was 43.4% (95%CI: 37.4%-49.6%). However the majority of positive sera had low antibody titers, at the cut off value for confirmation (i.e. = 1/40). The proportions of sera with HIA titer .1/40 were 0%, 3.0% and 24.6% in the young, middle-aged and older age groups respectively. These results indicate that pre-epidemic baseline antibody cross reactivity was stronger in the elderly ($60 yrs) and weaker in children and adolescents (,20 yrs) and adults (20-59 yrs), with highly significant differences between age groups (P,0.0001).
The reverse cumulative distribution curves of HIA titers are displayed for each age group and for the whole cohort on Figure 3 . The proportion of seropositive sera (HI $1/40) steadily increased during the epidemic unfolding (phase B, W32-39) and in immediate post epidemic period (phase C, W40-44) when it reached its maximum level, then declined in the late post epidemic period (phase D, W45-52). This decline was significant enough to return the reverse cumulative distribution curve to baseline levels in the elderly. The cumulative incidence rates, obtained after subtraction of the age-specific baseline-proxy seroprevalence from the raw seroprevalence at each phase of the epidemic are shown in Table 2 (note that the cumulative incidence rates of infection represented for the group ''all ages'' were standardized according to age structure of the community). The cumulative incidence rates were much higher in children and adolescents (,20 yrs), indicating very active transmission of infection within this age group. As mentioned earlier, cumulative incidence rates peaked in phase C (W40-44), and then declined indicating some lability of the humoral immune response against the pH1N1/2009v. The age-related difference observed in the incidence rates was highly statistically significant (P,0.0001).
To estimate more appropriately the decline of antibody titers occurring after the peak of the humoral response to the pH1N1/ 2009v, we considered paired-sera from the group of 264 subjects for whom the first serum sample (sample 1) was obtained just after the epidemic wave (phase C, W40-44), and the corresponding second sample was collected at the end of the survey (phase D, W45-52). Seronegation rates were 27.0% (61/226) for all age groups, 17.4% (12/69) in children and adolescents (,20 yrs), 32.3% (41/127) in adults (20-59 yrs) and 26.7% (8/30) in the elderly ($60 yrs). Differences between the seronegation rates according to age were statistically weakly significant (P = 0.0671).
We then considered the 1687 individuals for whom paired sera were available and we measured the seroconversion rates according to age and to the time of first serum sample collection (phase A, B or C). Criteria of seroconversion were defined in the method section. As shown in table 3, there was a sharp decline in seroconversion rates across all the age groups, depending on whether participants were enrolled during phase A, phase B, or phase C (P,0.0001). To interpret these data, one should remember that antibodies at seroprotective levels (HIA $1/40), in serum samples 1 collected during the per epidemic phase B or early post epidemic phase C could represent either base line cross reactive antibodies or rising pH1N1/2009 specific antibodies due to a recent or ongoing infection. This ambiguity could lead to underestimation of the seroconversion rate for subjects enrolled in phases B and C. In order to solve this ambiguity, we specifically considered the group of 249 subjects in whom cross reactive antibodies were detected at the time of phase A (W30-31). The seroconversion rate of this group is the most indicative of the exposure of individuals to the whole epidemic wave. It was the highest (63,2%, P,0.0001) in children and adolescents (,20 yrs), and still significantly high in adults (39.4%, P,0.0001).
We then tested in this particular group, the impact of (baseline) pre-epidemic cross reactive antibodies on the rate of seroconversion to pH1N1/2009 (Table 4) . No subject with HIA titer superior to 1/40 had evidence of seroconversion to pH1N1/2009. The seroconversion rate in individuals with a HIA titer equal to 1/40 was linked with age, being more important in children and adolescents (,20 yrs). The highest seroconversion rate (.56%) was registered in subjects with HIA titers inferior to 1/40, particularly for the under 20 years where it reached 85%. Hence, the risk of seroconversion decreased when pre-epidemic HIA titer was high after controlling for age (P,0.0001) (Figure 4) . The multivariate adjusted odds ratio for seroconversion were 0.15 (95%CI: 0.06-0.37, P,0.0001) per two-fold increase in baseline titer, 1.79 (95%CI: 1.23-2.59, P,0.003) per other household members who seroconverted, 5.33 (95%CI: 1.56-19.27, P,0.008) Figure 1 . The cohort profile and major outcomes. Figure 1 details the three phases of the protocol: i) inclusion (weeks 30-44) and serum samples S1 collection; ii) follow up for detection of ILI in households, qRT-PCR on nasal swabs and estimation of cumulative seroincidence rates; iii) end of the study (weeks 45-52) and samples S2 collection. HIA on paired sera (S1+S2) allowed estimating seroconversion rates. doi:10.1371/journal.pone.0025738.g001 Bp (baseline-proxy) seroprevalence rates were estimated on weeks 30-31 in each age group. b Cumulative incidence rates measured the raise between raw seroprevalence rates and age-specific baseline-proxy seroprevalence rate. In the group ''All ages'', cumulative incidence rates were standardized according to age structure of the community. doi:10.1371/journal.pone.0025738.t002 Data are numbers, percentages (95% confidence intervals) and ALR parameter test P value for comparison of seroconversion proportions according to time of first sample (S1) collection at inclusion, in each age group, after controlling for household selection. In the group ''All ages'', rates of seroconversion were standardized according to age structure of the community. NA: not assessed. Seroconversion was defined as a shift from seronegative at inclusion (i.e. HIA titer ,1/40) to seropositive on follow-up sample, or as a 4-fold increase of reciprocal HIA titer between first and second paired samples for sera tested seropositive on inclusion (i.e. HIA titer $1/40). for age ,20 years (vs age $60 years) and 11.35 (95%CI: 0.41-4.47, P = 0.62) for age 20-60 years (vs age $60 years). The observed and predicted seroconversion rates according to age and baseline HIA titer are displayed Figure 4 .
Finally, we considered the 46 subjects who had been infected by the pandemic virus over the course of the study, verified by a positive qRT-PCR nasal swab, and for whom paired sera were available. Initial HIA antibody titers in this group were ,1/40,
The CoPanFlu-RUN cohort was set up to conduct a prospective population-based study investigating the herd immunity induced by the 2009 pandemic influenza virus and identifying risk factors for pH1N1/2009v infection from paired sera collected in an entire community. Most works published to date have used either extensive cross-sectional serosurveys on pre-and post-epidemic independent serum samples, the baseline immunity being assessed from stored frozen samples [5, 7, 8] , or non representative adult cohorts (military, health care workers, long-stay patients). Antibody titers were measured by HIA using a cut-off value set at 1/40 as classically recommended. This HIA titer at 1/40 is considered protective, i.e. conferring 50% protection against a viral challenge [20] . Our assay has introduced some changes in the experimental protocol compared to the classic one. The use of a non-inactivated viral antigen, i.e. a native virus, with nondenatured epitopes probably allows detection of antibodies to epitopes of the hemagglutinin not detected in the classic HIA test. This can induce slight differences in the sensitivity of detection of cross-reacting antibodies, but this does not modify the kinetics of Ab and the epidemiological evolution of seroprevalence and does not jeopardize the global comparability of serological results. This is confirmed by the fact that our HI assay detected seroprotective antibody titers in 93.5% and gave evidence seroconversion in 73.9% of qRT-PCR confirmed pH1N1/2009 influenza, all figures close to those reported in the literature [5, 21] .
We considered that titers of .1/40, in sera collected from individuals enrolled during weeks 30 and 31 were cross reactive antibodies and not de novo antibodies triggered by the pandemic virus and hence used them as a proxy for baseline pre epidemic immunity. Several arguments support this assumption: i) the first case indicating autochthonous transmission in Reunion Island was reported by the epidemiological surveillance department of La Réunion on 21st July (week 30), i.e. the same day when inclusion started in our study cohort; ii) 7 to 15 days are required to develop an antibody response after viral infection; iii) On weeks 30 and 31, the epidemic activity due to the pandemic virus was very low in our study cohort and it became significant only after week 32. Hence, during weeks 30-31, 103 households were recruited and only 2 households reported ILI cases. Nasal swabs collected from these 2 individuals were tested qRT-PCR negative to the pandemic virus whereas one had evidence of coronavirus and rhinovirus using a multiplex RT-PCR to respiratory viruses (H. Pascalis, manuscript in preparation). In contrast, during weeks 32 to 39, 199 individuals belonging to 99 households reported ILI, among whom 60 individuals had documented infection by the pandemic virus.
Our study shows that a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus with the highest baseline-proxy seroprevalence rate observed among adults aged of 60 years or more. Other studies from all continents had also reported high pre-epidemic seropositivity rates among the elderly [5, 6, 8, [22] [23] [24] [25] [26] , though large variations do exist between countries [10, 11, 23, 27, 28] . These cross reactive antibodies have been interpreted as being the residual signature of the remote exposure of these individuals to H1N1 viruses circulating before 1957 [24, 25, 29, 30] . Baseline seropositivity rates that we report in children and in younger adults (i.e. 30%-35%) were notably higher than those reported from other parts of the world [6, 8, 22, 23, [31] [32] [33] . However one should note that these baseline antibodies were of low titer, just at the level of the HIA threshold (i.e. 1/40). Several factors could have contributed to this comparatively high baseline rates found in our study: i) It may reflect the fact that the HI test used in our study was marginally more sensitive than the classic one [17] ; ii) Some individuals may have already been infected with pH1N1/ 2009 virus at weeks 30 and 31 and may have triggered an antibody response to the virus. This hypothesis seems unlikely in view of the arguments presented above and of a similar high proportion of sera titering HIA = 1/40 among 122 sera from adult patients sent for diagnostic purposes to the Regional Hospital microbiology laboratory, during the first half of 2009 (i.e. before the 2009 pandemic) (data not shown). However we cannot formally exclude this hypothesis in view of a recently reported study from Taiwan [11] that showed evidence of subclinical community transmission with proved seroconversion several weeks before report of the first documented case in the island. A similar conclusion was also drawn from Australia [34] ; iii) our serological test might detect cross-reactive antibodies triggered by recent vaccination with trivalent seasonal influenza vaccine as reported [4, [35] [36] [37] [38] [39] . However, seasonal influenza vaccines were of rather limited use in Reunion Island, especially in children and young adults; iv) Finally the high baseline titers may reflect the infectious history of the individuals to seasonal influenza viruses cross antigenic with pH1N1/2009 virus as recently suggested for seasonal 2007 H1N1 infection [40] . This serosurvey indicates that a large fraction of the Reunion Island population was infected with the pandemic virus. Younger people, have paid the main tribute to the epidemic as almost two thirds show evidence of seroconversion, confirming earlier clinical reports from the island [12] and accumulating reports from other countries [17, 32, 41, 42] and suggesting that school children have likely played the central role in the epidemic diffusion of the pandemic virus. Lower infection rates were found in adults and the lowest rates were recorded in the elderly.
Based on clinical cases reported to the epidemiological surveillance services [12] , it was estimated that 66,915 persons in Reunion Island who consulted a physician were infected by the pH1N1/2009 virus during the 9 weeks of the epidemic, giving a cumulative attack rate of 8.26%. Taking into account those who did not consult a physician, the number of symptomatic infected persons was estimated to 104,067 (attack rate: 12.85%). In fact, the attack rate of pH1N1/2009 infection in our serosurvey was about 42%-44% at the peak of the antibody response (i.e., weeks 40-44), a figure which is at least 3 to 4 times higher than rates of infection based on clinical cases The wide gap between the two estimates indicates that a large fraction (almost two thirds) of those who got infected by pH1N1/2009 virus escaped medical detection, probably because they developed mild disease or asymptomatic infection, a further indication of the benign nature of the virus, at least at the community level. In England, Baguelin et al. [43] estimated that the cumulative incidence rates of infection by the pandemic virus in children were 20 to 40 times higher than that estimated from clinical surveillance.
Our study, as others [6] , indicates that pre-existing cross reactive antibodies to pH1N1/2009 at titers $1/40 prevented from seroconversion in response to the pandemic virus. This level of pre-existing cross reactive immunity likely confers true protection against infection as about two thirds and one third of documented infection (qRT-PCR positive) in our series have occurred in individuals with baseline HIA titers ,1/40 and = 1/ 40 respectively and less than 5% of documented infections occurred in individuals with base line titers .1/40. The protection was effective not only in older adults but also in younger persons. This indicates that protection was conferred not only by baseline cross reactive antibodies triggered by close pH1N1/2009 viruses that circulated before 1957 (as in the elderly), but also by antibodies likely resulting from recent exposure to seasonal influenza epidemics (as shown in younger persons) [40] . The observed seroconversion rates depend on age, after adjusting for baseline pH1N1/2009 titers. The protective role of increasing age might be explained by a stronger cross-immunity in adults and elderly or by a higher exposure of young subjects to the virus during the 2009 epidemic (due to social contacts and mixing patterns). It may also indicate that immune mechanisms other than cross reactive antibodies detected by HIA (i.e. immunity to neuraminidase and conserved T cells epitopes [44] might develop throughout life, providing additional protection from infection or severe disease, especially in the elderly. Interestingly, evidence is seen for a decline in antibody titers, which occurred soon after the passage of the epidemic wave. In paired sera, this decline was significant enough to bring, within a few weeks, almost 27% of sera that tested positive (i.e. HI titers $1/40) in the immediate post epidemic phase to levels under the cut-off value in the second serum sample. This decay accounts for the observation that older adults ($60 yrs) in the study cohort were apparently almost completely spared by the epidemic if one only considers cumulative incidence rates derived from IHA titration on samples 2 (weeks 45-52). In fact, the cumulative incidence rate in older adults measured just after the epidemic peak (i.e. weeks 40-44) was 20.4%. Similar results of early antibody decay were recently reported [10, 45] . More generally, these data show that serosurveys conducted months after passage of the epidemic, likely underestimate the real extent of pH1N1/2009 infection, compared to antibody titration performed earlier, when humoral responses are at their highest level. Whether the decline in antibody titers has functional immunologic consequence to individuals or within the communities warrants further investigation. However, one should note that there was no second epidemic wave in Reunion Island during the subsequent austral winter seasons in 2010 and 2011.
Influenza during the 2010 winter was at a level not higher than the usual passages of seasonal flu, though almost two thirds of documented cases in 2010 were also due to pH1N1/2009v [46] . In addition many fewer pandemic virus isolates were noted during the ongoing 2011 austral winter, strongly suggesting that the first epidemic wave had conferred a solid herd immunity, at the community level.
Our study has some limitations. The fact that the epidemic progression coincided with the implementation of the prospective study, we were not able to collect, strictly speaking, pre-epidemic sera from the cohort members. Therefore we used as proxy base line seroprevalence data from individuals recruited at the very beginning of the investigation when the epidemic activity in the cohort was very low. This may overestimate the base line immunity if subclinical community transmission had occurred before the first cases of pH1N1/2009 influenza were reported. Antibodies to the pandemic virus were detected by HIA, a test that has a good specificity but a rather low sensitivity [46] . Hence, the threshold of 1/40 may underestimate the number of infected individuals. However, rates of seroconversion, the serologic gold standard test based on paired sera, likely gave the most accurate picture of the pandemic in at the community level in Reunion Island.
|
What is meant by a protective HIA titer?
| 5,258 |
conferring 50% protection against a viral challenge
| 22,786 |
1,601 |
Pandemic Influenza Due to pH1N1/2009 Virus: Estimation of Infection Burden in Reunion Island through a Prospective Serosurvey, Austral Winter 2009
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3183080/
SHA: ee6d70a53e3262cea6f85bd8b226f6b4c8b5f64b
Authors: Dellagi, Koussay; Rollot, Olivier; Temmam, Sarah; Salez, Nicolas; Guernier, Vanina; Pascalis, Hervé; Gérardin, Patrick; Fianu, Adrian; Lapidus, Nathanael; Naty, Nadège; Tortosa, Pablo; Boussaïd, Karim; Jaffar-Banjee, Marie-Christine; Filleul, Laurent; Flahault, Antoine; Carrat, Fabrice; Favier, Francois; de Lamballerie, Xavier
Date: 2011-09-29
DOI: 10.1371/journal.pone.0025738
License: cc-by
Abstract: BACKGROUND: To date, there is little information that reflects the true extent of spread of the pH1N1/2009v influenza pandemic at the community level as infection often results in mild or no clinical symptoms. This study aimed at assessing through a prospective study, the attack rate of pH1N1/2009 virus in Reunion Island and risk factors of infection, during the 2009 season. METHODOLOGY/PRINCIPAL FINDINGS: A serosurvey was conducted during the 2009 austral winter, in the frame of a prospective population study. Pairs of sera were collected from 1687 individuals belonging to 772 households, during and after passage of the pandemic wave. Antibodies to pH1N1/2009v were titered using the hemagglutination inhibition assay (HIA) with titers ≥1/40 being considered positive. Seroprevalence during the first two weeks of detection of pH1N1/2009v in Reunion Island was 29.8% in people under 20 years of age, 35.6% in adults (20–59 years) and 73.3% in the elderly (≥60 years) (P<0.0001). Baseline corrected cumulative incidence rates, were 42.9%, 13.9% and 0% in these age groups respectively (P<0.0001). A significant decline in antibody titers occurred soon after the passage of the epidemic wave. Seroconversion rates to pH1N1/2009 correlated negatively with age: 63.2%, 39.4% and 16.7%, in each age group respectively (P<0.0001). Seroconversion occurred in 65.2% of individuals who were seronegative at inclusion compared to 6.8% in those who were initially seropositive. CONCLUSIONS: Seroincidence of pH1N1/2009v infection was three times that estimated from clinical surveillance, indicating that almost two thirds of infections occurring at the community level have escaped medical detection. People under 20 years of age were the most affected group. Pre-epidemic titers ≥1/40 prevented seroconversion and are likely protective against infection. A concern was raised about the long term stability of the antibody responses.
Text: In April 2009, the first cases of acute respiratory infections caused by a novel triple-reassortant influenza virus, pH1N1/ 2009v, occurred in Mexico and the United States [1] . The rapid spread of infection to other continents led the World Health Organization (WHO) to declare on 11 June 2009 that a pandemic of pH1N1/2009v influenza was under way, which raised major international concern about the risk of high morbidity and lethality and the potential for severe socio-economic impact. Actually, the potential impact of this first third-millenium influenza pandemic has been revisited downwards as morbidity and case-fatality rates were less severe than initially anticipated [2] . Illness surveillance data do not allow to an accurate estimate of the true influenza infection rate, as a substantial proportion of infections are asymptomatic or mild [3] . Serological surveys can overcome this limitation, but must take into account that a significant proportion of the population that exhibited crossprotective antibody titers before circulation of the pH1N1/2009v [4] . This so-called ''baseline immunity'' has to be subtracted from the seroprevalence observed after the pandemic wave, to determine seroincidence in serosurveys [5] [6] [7] [8] . However, except for few studies [9] [10] [11] , most of these serosurveys did not use serial measurements in the same person, which allows for a better understanding of antibody kinetics and the dynamics of infection within individuals and communities.
Reunion Island (805,500 inhabitants) is a French overseas department located in the southwestern Indian Ocean, 700 km east of Madagascar and 200 km southwest of Mauritius. The first imported case of pH1N1/2009v was identified on 5 th July 2009 (week 29) in a traveller returning from Australia. The first case indicating community transmission was detected on 21 st July (week 30). pH1N1/2009v became the predominant circulating influenza virus within four weeks of its first detection, its activity peaked during week 35 (24) (25) (26) (27) (28) (29) (30) and ended at week 38 [12] . Contrary to initial fears, the health care system was not overwhelmed, as morbidity and mortality rates were lower than predicted [12] [13] [14] .
In order to assess at the community level, the actual magnitude of the pH1N1/2009v pandemic and the extent of the herd immunity acquired after passage of the epidemic wave, a prospective population serosurvey was conducted in Reunion Island during the passage of the epidemic wave in the 2009 austral winter season (July-December 2009): prevalence of infection was assessed on a weekly basis and seroconversion rates were measured using paired sera.
The CoPanFLu-RUN was part of the CoPanFLu international project, a consortium between the French National Institute of Health and Medical Research (INSERM), the Institute of Research for Development (IRD) and the Mérieux Fondation under the promotion of the School of Advanced Studies in Public Health (EHESP). To enable the rapid implementation of the study in anticipation of the imminent spread of the pandemic wave, we used a pre-existing sample of 2442 households established in October 2006 for the investigation of the Chikungunya outbreak (SEROCHIK) and updated in May 2008 throughout a follow-up telephone survey (TELECHIK) on a basis of 1148 households [15, 16] . We took special attention to select households representing a wide range of geographic locations in order to minimize the repartition bias.
The inclusion phase started on July 21 st (week 30) and was continued up to week 44, throughout the epidemic wave and beyond. A first serum sample (sample 1) was obtained from each household member. An active telephonic inquiry was then conducted twice a week to record symptoms compatible with influenza-like illness (ILI) occurring in households. Report of ILI (fever $37.8uC associated with any respiratory or systemic symptom) led to three consecutive visits of a nurse to the incident case-dwelling (on day 0, +3 and +8 post-report) to record symptoms and collect nasal swabs from all family members (for qRT-PCR detection of pH1N1/2009v. At week 45, the active inquiry was discontinued and a second (post-epidemic) serum sample (sample 2) was obtained (weeks 45-52) to determine seroconversion rates. Sera were aliquoted and stored at 280uC.
The protocol was conducted in accordance with the Declaration of Helsinki and French law for biomedical research (Nu ID RCB AFSSAPS: 2009-A00689-48) and was approved by the local Ethics Committee (Comité de Protection des Personnes of Bordeaux 2 University). Every eligible person for participation was asked for giving their written informed consent.
Viral genome detection by RT-PCR. Viral RNA was extracted from 140 mL of nasal swab eluate using the QIAamp Viral RNA kit (Qiagen) and processed for detection by TaqMan qRT-PCR targeting the heamagglutinin HA gene (SuperScript III Platinum one-step qRT-PCR system, Invitrogen) according to the recommendations of the Pasteur Institute (Van der Werf S. & Enouf V., SOP/FluA/130509). Confirmed pH1N1/2009v infection was defined as a positive qRT-PCR detection of the HA gene in at least one nasal swab.
Hemagglutination inhibition assay (HIA). A standard hemagglutination inhibition technique was adapted to detect and quantify pH1N1/2009v antibodies [17] . The antigen was prepared by diluting a non-inactivated cell culture supernatant producing a pdm H1N1v strain (strain OPYFLU-1 isolated from a young patient returning from Mexico in early May 2009) [18] . Briefly, the virus was propagated onto MDCK cells under standard conditions. The last passage (used for antigen preparation) was performed in the absence of trypsin and ht-FBS. The supernatant was collected at day seven p.i. clarified by centrifugation at 8006 g for 10 min at room temperature, aliquoted and conserved at 280uC. The hemagglutinating titer of the non inactivated viral antigen was immediately determined under the HIA format described below. The dilution providing 5.33 hemagglutinating units in a volume of 25 mL was used for subsequent HIA. Sera were heat-inactivated at 56uC for 30 min prior to use. Sequential twofold dilutions in PBS (1/10 to 1/1280) in volumes of 25 mL were performed and distributed in V-bottom 96 well microplates. Human red blood cells (RBC) were used for hemagglutination experiments. Detection and quantification of antibody to pH1N1/2009v was performed as follows: 25 mL of virus suspension was added to the serum dilution (25 mL) and incubated for 1 hour at room temperature. Each well was then filled with 25 mL of a 1% RBC suspension in PBS (v/v: 0.33%), followed by another 30 min incubation at room temperature. The HIA titer was determined as the last dilution providing clear inhibition of hemagglutination. All experiments were performed in the presence of the same negative and positive controls, the latter including sera with 1/40, 1/80, 1/160 and 1/320 antibody titers.
The results reported in this study were based only on serological analysis of paired sera. For the sake of analysis, four successive phases were identified throughout the pandemic wave: phase A (weeks 30-31) corresponded to early epidemic time, phase B (W32-39) to the epidemic unfolding, phase C (W40-44) to the immediate post-epidemic stage and phase D (W45-52) to the late post-epidemic stage. Seropositivity was defined as a HIA titer of 1/ 40 or more. The baseline-proxy seroprevalence rate was estimated on serum samples collected in phase A. The cumulative incidence rate of infection measured the raise between the raw seroprevalence rate at any given time during the epidemic phases (S2pi) and the age-specific baseline-proxy seroprevalence rate (S1pA) (s2 pi -s1 pA ). Seroconversion was defined as a shift from seronegative at inclusion (sample 1: HIA ,1/40) to seropositive on follow-up (sample 2: HIA $1/40), or for sera tested seropositive on inclusion as a four-fold increase of HIA titers between sample 1 and sample 2 paired sera. We also calculated the proportion of sera that tested seropositive in sample 1 for which the HIA titer decreased fourfold and passed under the cut-off value of 1/40 in sample 2. We considered this proportion as a ''seronegation'' rate.
The sample size was calculated for identifying risk factors in the prospective cohort study. Considering on average three individuals per household, an intra-household correlation of 0.3, a power greater than 80% could be obtained with a sample size of 840 comprising 2500 individuals, assuming exposure levels ranging from 10% to 90% and a relative risk greater than 1.3. With 2,500 subjects, the study allowed 1-2% absolute precision around the estimated values for seroconversion rates. Data entry used EpiData version 3.1 (The Epidata Association, Odense, Denmark). SAS version 9.1 (SAS Inc., Cary, NC, USA) was used for statistical analysis. The characteristics of the study cohort were compared to those of the population of Reunion Island and a Chi2 test (or Fisher's exact test when non applicable) was used to analyse differences in age, sex and geographic location. Cumulative incidence rates of infection (i.e. seroincidence) and seroconversion rates were standardized according to the age structure of the community (French National Institute for Statistics and Economical Studies (INSEE) source).
Baseline-proxy seroprevalence, cumulative incidence rates of infection, as well as seroconversion and seronegation rates, were expressed as percentages. Cumulative reverse distribution curves were used to show the distribution of antibody titers. In all tests, a P value,0.05 was considered significant.
We estimated 95% confidence intervals (CIs) of proportions by using a cluster bootstrap technique with 1000 re-samples [19] . After bootstraping, we used an ANOVA model to compare mean cumulative incidence proportions between pandemic phases, within each age group. We used an alternating logistic regression model (ALR) with an exchangeable log Odds Ratio (OR) to test the intra-household correlation-adjusted association between factors and the seroconversion outcome.
Data were analysed with respect to subject age. Initially, four age groups were considered: the children and adolescents (,20 yrs), young adults (20-39 yrs), middle-age adults (40-59 yrs), and elderly adults ($60 yrs). As the cumulative incidence of infection of the second and third groups were very close, both groups were merged into one adults group (20-59 yrs). Therefore we refer further in our study to three age groups: children and adolescents (,20 yrs), adults (20-59 yrs), elderly ($60 yrs).
A total of 2,164 individuals from 772 households were enrolled between weeks 30 and 44 in the CoPanFlu-RUN cohort, allowing the collection of 1,932 sera at inclusion (sample 1). During this period, 136 households (17.7% of households) containing 464 individuals (21.4% of individuals) reported at least one case of ILI. Sixty subjects among the 464 individuals (12.9%, belonging to 33 households [24.3%]) were qRT-PCR positive, which documented the pH1N1/2009v infection. No positive qRT-PCR could be detected after week 37 and no ILI was reported after week 40, the end of the epidemic wave. The second follow up serum sample (sample 2) was obtained for 1,759 subjects at least five weeks after the end of the epidemic wave (weeks 45-52) which allowed the constitution of a serobank of 1,687 paired-sera. The profile of the cohort and the major outcomes are displayed in Figure 1 . Details on inclusions and serum sample timing with respect to the circulation of pH1N1/2009v over the island are provided in figure 2 . The socio-demographic and space-time characteristics of the cohort are detailed in Table 1 . Compared to the community of Reunion Island, the sample of 1,687 individuals for whom pairedsera were available, was older (,20 yrs: 27% vs 35%, and $60 yrs: 17,9% vs 11,3%) and composed of a slight excess of females (54.1% vs 51.5%). The imbalance was due to a deficit in subjects aged under 40 years, reflecting men at work and the fact that parents declined the second serum for children younger than five.
Baseline-proxy (,pre-epidemic) HIA titers to the pH1N1/ 2009v were measured on sample 1 ( Table 2) , obtained from 249 subjects (103 households) recruited at the very beginning of the investigation during weeks 30 and 31 (phase A, Figure 2 ), when the epidemic activity in the cohort was still very low. Age distribution in this group was similar to that of the whole cohort (data not shown). The overall, the baseline-proxy seroprevalence rate (HIA $1/40), over all ages, was 43.4% (95%CI: 37.4%-49.6%). However the majority of positive sera had low antibody titers, at the cut off value for confirmation (i.e. = 1/40). The proportions of sera with HIA titer .1/40 were 0%, 3.0% and 24.6% in the young, middle-aged and older age groups respectively. These results indicate that pre-epidemic baseline antibody cross reactivity was stronger in the elderly ($60 yrs) and weaker in children and adolescents (,20 yrs) and adults (20-59 yrs), with highly significant differences between age groups (P,0.0001).
The reverse cumulative distribution curves of HIA titers are displayed for each age group and for the whole cohort on Figure 3 . The proportion of seropositive sera (HI $1/40) steadily increased during the epidemic unfolding (phase B, W32-39) and in immediate post epidemic period (phase C, W40-44) when it reached its maximum level, then declined in the late post epidemic period (phase D, W45-52). This decline was significant enough to return the reverse cumulative distribution curve to baseline levels in the elderly. The cumulative incidence rates, obtained after subtraction of the age-specific baseline-proxy seroprevalence from the raw seroprevalence at each phase of the epidemic are shown in Table 2 (note that the cumulative incidence rates of infection represented for the group ''all ages'' were standardized according to age structure of the community). The cumulative incidence rates were much higher in children and adolescents (,20 yrs), indicating very active transmission of infection within this age group. As mentioned earlier, cumulative incidence rates peaked in phase C (W40-44), and then declined indicating some lability of the humoral immune response against the pH1N1/2009v. The age-related difference observed in the incidence rates was highly statistically significant (P,0.0001).
To estimate more appropriately the decline of antibody titers occurring after the peak of the humoral response to the pH1N1/ 2009v, we considered paired-sera from the group of 264 subjects for whom the first serum sample (sample 1) was obtained just after the epidemic wave (phase C, W40-44), and the corresponding second sample was collected at the end of the survey (phase D, W45-52). Seronegation rates were 27.0% (61/226) for all age groups, 17.4% (12/69) in children and adolescents (,20 yrs), 32.3% (41/127) in adults (20-59 yrs) and 26.7% (8/30) in the elderly ($60 yrs). Differences between the seronegation rates according to age were statistically weakly significant (P = 0.0671).
We then considered the 1687 individuals for whom paired sera were available and we measured the seroconversion rates according to age and to the time of first serum sample collection (phase A, B or C). Criteria of seroconversion were defined in the method section. As shown in table 3, there was a sharp decline in seroconversion rates across all the age groups, depending on whether participants were enrolled during phase A, phase B, or phase C (P,0.0001). To interpret these data, one should remember that antibodies at seroprotective levels (HIA $1/40), in serum samples 1 collected during the per epidemic phase B or early post epidemic phase C could represent either base line cross reactive antibodies or rising pH1N1/2009 specific antibodies due to a recent or ongoing infection. This ambiguity could lead to underestimation of the seroconversion rate for subjects enrolled in phases B and C. In order to solve this ambiguity, we specifically considered the group of 249 subjects in whom cross reactive antibodies were detected at the time of phase A (W30-31). The seroconversion rate of this group is the most indicative of the exposure of individuals to the whole epidemic wave. It was the highest (63,2%, P,0.0001) in children and adolescents (,20 yrs), and still significantly high in adults (39.4%, P,0.0001).
We then tested in this particular group, the impact of (baseline) pre-epidemic cross reactive antibodies on the rate of seroconversion to pH1N1/2009 (Table 4) . No subject with HIA titer superior to 1/40 had evidence of seroconversion to pH1N1/2009. The seroconversion rate in individuals with a HIA titer equal to 1/40 was linked with age, being more important in children and adolescents (,20 yrs). The highest seroconversion rate (.56%) was registered in subjects with HIA titers inferior to 1/40, particularly for the under 20 years where it reached 85%. Hence, the risk of seroconversion decreased when pre-epidemic HIA titer was high after controlling for age (P,0.0001) (Figure 4) . The multivariate adjusted odds ratio for seroconversion were 0.15 (95%CI: 0.06-0.37, P,0.0001) per two-fold increase in baseline titer, 1.79 (95%CI: 1.23-2.59, P,0.003) per other household members who seroconverted, 5.33 (95%CI: 1.56-19.27, P,0.008) Figure 1 . The cohort profile and major outcomes. Figure 1 details the three phases of the protocol: i) inclusion (weeks 30-44) and serum samples S1 collection; ii) follow up for detection of ILI in households, qRT-PCR on nasal swabs and estimation of cumulative seroincidence rates; iii) end of the study (weeks 45-52) and samples S2 collection. HIA on paired sera (S1+S2) allowed estimating seroconversion rates. doi:10.1371/journal.pone.0025738.g001 Bp (baseline-proxy) seroprevalence rates were estimated on weeks 30-31 in each age group. b Cumulative incidence rates measured the raise between raw seroprevalence rates and age-specific baseline-proxy seroprevalence rate. In the group ''All ages'', cumulative incidence rates were standardized according to age structure of the community. doi:10.1371/journal.pone.0025738.t002 Data are numbers, percentages (95% confidence intervals) and ALR parameter test P value for comparison of seroconversion proportions according to time of first sample (S1) collection at inclusion, in each age group, after controlling for household selection. In the group ''All ages'', rates of seroconversion were standardized according to age structure of the community. NA: not assessed. Seroconversion was defined as a shift from seronegative at inclusion (i.e. HIA titer ,1/40) to seropositive on follow-up sample, or as a 4-fold increase of reciprocal HIA titer between first and second paired samples for sera tested seropositive on inclusion (i.e. HIA titer $1/40). for age ,20 years (vs age $60 years) and 11.35 (95%CI: 0.41-4.47, P = 0.62) for age 20-60 years (vs age $60 years). The observed and predicted seroconversion rates according to age and baseline HIA titer are displayed Figure 4 .
Finally, we considered the 46 subjects who had been infected by the pandemic virus over the course of the study, verified by a positive qRT-PCR nasal swab, and for whom paired sera were available. Initial HIA antibody titers in this group were ,1/40,
The CoPanFlu-RUN cohort was set up to conduct a prospective population-based study investigating the herd immunity induced by the 2009 pandemic influenza virus and identifying risk factors for pH1N1/2009v infection from paired sera collected in an entire community. Most works published to date have used either extensive cross-sectional serosurveys on pre-and post-epidemic independent serum samples, the baseline immunity being assessed from stored frozen samples [5, 7, 8] , or non representative adult cohorts (military, health care workers, long-stay patients). Antibody titers were measured by HIA using a cut-off value set at 1/40 as classically recommended. This HIA titer at 1/40 is considered protective, i.e. conferring 50% protection against a viral challenge [20] . Our assay has introduced some changes in the experimental protocol compared to the classic one. The use of a non-inactivated viral antigen, i.e. a native virus, with nondenatured epitopes probably allows detection of antibodies to epitopes of the hemagglutinin not detected in the classic HIA test. This can induce slight differences in the sensitivity of detection of cross-reacting antibodies, but this does not modify the kinetics of Ab and the epidemiological evolution of seroprevalence and does not jeopardize the global comparability of serological results. This is confirmed by the fact that our HI assay detected seroprotective antibody titers in 93.5% and gave evidence seroconversion in 73.9% of qRT-PCR confirmed pH1N1/2009 influenza, all figures close to those reported in the literature [5, 21] .
We considered that titers of .1/40, in sera collected from individuals enrolled during weeks 30 and 31 were cross reactive antibodies and not de novo antibodies triggered by the pandemic virus and hence used them as a proxy for baseline pre epidemic immunity. Several arguments support this assumption: i) the first case indicating autochthonous transmission in Reunion Island was reported by the epidemiological surveillance department of La Réunion on 21st July (week 30), i.e. the same day when inclusion started in our study cohort; ii) 7 to 15 days are required to develop an antibody response after viral infection; iii) On weeks 30 and 31, the epidemic activity due to the pandemic virus was very low in our study cohort and it became significant only after week 32. Hence, during weeks 30-31, 103 households were recruited and only 2 households reported ILI cases. Nasal swabs collected from these 2 individuals were tested qRT-PCR negative to the pandemic virus whereas one had evidence of coronavirus and rhinovirus using a multiplex RT-PCR to respiratory viruses (H. Pascalis, manuscript in preparation). In contrast, during weeks 32 to 39, 199 individuals belonging to 99 households reported ILI, among whom 60 individuals had documented infection by the pandemic virus.
Our study shows that a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus with the highest baseline-proxy seroprevalence rate observed among adults aged of 60 years or more. Other studies from all continents had also reported high pre-epidemic seropositivity rates among the elderly [5, 6, 8, [22] [23] [24] [25] [26] , though large variations do exist between countries [10, 11, 23, 27, 28] . These cross reactive antibodies have been interpreted as being the residual signature of the remote exposure of these individuals to H1N1 viruses circulating before 1957 [24, 25, 29, 30] . Baseline seropositivity rates that we report in children and in younger adults (i.e. 30%-35%) were notably higher than those reported from other parts of the world [6, 8, 22, 23, [31] [32] [33] . However one should note that these baseline antibodies were of low titer, just at the level of the HIA threshold (i.e. 1/40). Several factors could have contributed to this comparatively high baseline rates found in our study: i) It may reflect the fact that the HI test used in our study was marginally more sensitive than the classic one [17] ; ii) Some individuals may have already been infected with pH1N1/ 2009 virus at weeks 30 and 31 and may have triggered an antibody response to the virus. This hypothesis seems unlikely in view of the arguments presented above and of a similar high proportion of sera titering HIA = 1/40 among 122 sera from adult patients sent for diagnostic purposes to the Regional Hospital microbiology laboratory, during the first half of 2009 (i.e. before the 2009 pandemic) (data not shown). However we cannot formally exclude this hypothesis in view of a recently reported study from Taiwan [11] that showed evidence of subclinical community transmission with proved seroconversion several weeks before report of the first documented case in the island. A similar conclusion was also drawn from Australia [34] ; iii) our serological test might detect cross-reactive antibodies triggered by recent vaccination with trivalent seasonal influenza vaccine as reported [4, [35] [36] [37] [38] [39] . However, seasonal influenza vaccines were of rather limited use in Reunion Island, especially in children and young adults; iv) Finally the high baseline titers may reflect the infectious history of the individuals to seasonal influenza viruses cross antigenic with pH1N1/2009 virus as recently suggested for seasonal 2007 H1N1 infection [40] . This serosurvey indicates that a large fraction of the Reunion Island population was infected with the pandemic virus. Younger people, have paid the main tribute to the epidemic as almost two thirds show evidence of seroconversion, confirming earlier clinical reports from the island [12] and accumulating reports from other countries [17, 32, 41, 42] and suggesting that school children have likely played the central role in the epidemic diffusion of the pandemic virus. Lower infection rates were found in adults and the lowest rates were recorded in the elderly.
Based on clinical cases reported to the epidemiological surveillance services [12] , it was estimated that 66,915 persons in Reunion Island who consulted a physician were infected by the pH1N1/2009 virus during the 9 weeks of the epidemic, giving a cumulative attack rate of 8.26%. Taking into account those who did not consult a physician, the number of symptomatic infected persons was estimated to 104,067 (attack rate: 12.85%). In fact, the attack rate of pH1N1/2009 infection in our serosurvey was about 42%-44% at the peak of the antibody response (i.e., weeks 40-44), a figure which is at least 3 to 4 times higher than rates of infection based on clinical cases The wide gap between the two estimates indicates that a large fraction (almost two thirds) of those who got infected by pH1N1/2009 virus escaped medical detection, probably because they developed mild disease or asymptomatic infection, a further indication of the benign nature of the virus, at least at the community level. In England, Baguelin et al. [43] estimated that the cumulative incidence rates of infection by the pandemic virus in children were 20 to 40 times higher than that estimated from clinical surveillance.
Our study, as others [6] , indicates that pre-existing cross reactive antibodies to pH1N1/2009 at titers $1/40 prevented from seroconversion in response to the pandemic virus. This level of pre-existing cross reactive immunity likely confers true protection against infection as about two thirds and one third of documented infection (qRT-PCR positive) in our series have occurred in individuals with baseline HIA titers ,1/40 and = 1/ 40 respectively and less than 5% of documented infections occurred in individuals with base line titers .1/40. The protection was effective not only in older adults but also in younger persons. This indicates that protection was conferred not only by baseline cross reactive antibodies triggered by close pH1N1/2009 viruses that circulated before 1957 (as in the elderly), but also by antibodies likely resulting from recent exposure to seasonal influenza epidemics (as shown in younger persons) [40] . The observed seroconversion rates depend on age, after adjusting for baseline pH1N1/2009 titers. The protective role of increasing age might be explained by a stronger cross-immunity in adults and elderly or by a higher exposure of young subjects to the virus during the 2009 epidemic (due to social contacts and mixing patterns). It may also indicate that immune mechanisms other than cross reactive antibodies detected by HIA (i.e. immunity to neuraminidase and conserved T cells epitopes [44] might develop throughout life, providing additional protection from infection or severe disease, especially in the elderly. Interestingly, evidence is seen for a decline in antibody titers, which occurred soon after the passage of the epidemic wave. In paired sera, this decline was significant enough to bring, within a few weeks, almost 27% of sera that tested positive (i.e. HI titers $1/40) in the immediate post epidemic phase to levels under the cut-off value in the second serum sample. This decay accounts for the observation that older adults ($60 yrs) in the study cohort were apparently almost completely spared by the epidemic if one only considers cumulative incidence rates derived from IHA titration on samples 2 (weeks 45-52). In fact, the cumulative incidence rate in older adults measured just after the epidemic peak (i.e. weeks 40-44) was 20.4%. Similar results of early antibody decay were recently reported [10, 45] . More generally, these data show that serosurveys conducted months after passage of the epidemic, likely underestimate the real extent of pH1N1/2009 infection, compared to antibody titration performed earlier, when humoral responses are at their highest level. Whether the decline in antibody titers has functional immunologic consequence to individuals or within the communities warrants further investigation. However, one should note that there was no second epidemic wave in Reunion Island during the subsequent austral winter seasons in 2010 and 2011.
Influenza during the 2010 winter was at a level not higher than the usual passages of seasonal flu, though almost two thirds of documented cases in 2010 were also due to pH1N1/2009v [46] . In addition many fewer pandemic virus isolates were noted during the ongoing 2011 austral winter, strongly suggesting that the first epidemic wave had conferred a solid herd immunity, at the community level.
Our study has some limitations. The fact that the epidemic progression coincided with the implementation of the prospective study, we were not able to collect, strictly speaking, pre-epidemic sera from the cohort members. Therefore we used as proxy base line seroprevalence data from individuals recruited at the very beginning of the investigation when the epidemic activity in the cohort was very low. This may overestimate the base line immunity if subclinical community transmission had occurred before the first cases of pH1N1/2009 influenza were reported. Antibodies to the pandemic virus were detected by HIA, a test that has a good specificity but a rather low sensitivity [46] . Hence, the threshold of 1/40 may underestimate the number of infected individuals. However, rates of seroconversion, the serologic gold standard test based on paired sera, likely gave the most accurate picture of the pandemic in at the community level in Reunion Island.
|
What are the results of the study?
| 5,259 |
a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus
| 24,962 |
1,601 |
Pandemic Influenza Due to pH1N1/2009 Virus: Estimation of Infection Burden in Reunion Island through a Prospective Serosurvey, Austral Winter 2009
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3183080/
SHA: ee6d70a53e3262cea6f85bd8b226f6b4c8b5f64b
Authors: Dellagi, Koussay; Rollot, Olivier; Temmam, Sarah; Salez, Nicolas; Guernier, Vanina; Pascalis, Hervé; Gérardin, Patrick; Fianu, Adrian; Lapidus, Nathanael; Naty, Nadège; Tortosa, Pablo; Boussaïd, Karim; Jaffar-Banjee, Marie-Christine; Filleul, Laurent; Flahault, Antoine; Carrat, Fabrice; Favier, Francois; de Lamballerie, Xavier
Date: 2011-09-29
DOI: 10.1371/journal.pone.0025738
License: cc-by
Abstract: BACKGROUND: To date, there is little information that reflects the true extent of spread of the pH1N1/2009v influenza pandemic at the community level as infection often results in mild or no clinical symptoms. This study aimed at assessing through a prospective study, the attack rate of pH1N1/2009 virus in Reunion Island and risk factors of infection, during the 2009 season. METHODOLOGY/PRINCIPAL FINDINGS: A serosurvey was conducted during the 2009 austral winter, in the frame of a prospective population study. Pairs of sera were collected from 1687 individuals belonging to 772 households, during and after passage of the pandemic wave. Antibodies to pH1N1/2009v were titered using the hemagglutination inhibition assay (HIA) with titers ≥1/40 being considered positive. Seroprevalence during the first two weeks of detection of pH1N1/2009v in Reunion Island was 29.8% in people under 20 years of age, 35.6% in adults (20–59 years) and 73.3% in the elderly (≥60 years) (P<0.0001). Baseline corrected cumulative incidence rates, were 42.9%, 13.9% and 0% in these age groups respectively (P<0.0001). A significant decline in antibody titers occurred soon after the passage of the epidemic wave. Seroconversion rates to pH1N1/2009 correlated negatively with age: 63.2%, 39.4% and 16.7%, in each age group respectively (P<0.0001). Seroconversion occurred in 65.2% of individuals who were seronegative at inclusion compared to 6.8% in those who were initially seropositive. CONCLUSIONS: Seroincidence of pH1N1/2009v infection was three times that estimated from clinical surveillance, indicating that almost two thirds of infections occurring at the community level have escaped medical detection. People under 20 years of age were the most affected group. Pre-epidemic titers ≥1/40 prevented seroconversion and are likely protective against infection. A concern was raised about the long term stability of the antibody responses.
Text: In April 2009, the first cases of acute respiratory infections caused by a novel triple-reassortant influenza virus, pH1N1/ 2009v, occurred in Mexico and the United States [1] . The rapid spread of infection to other continents led the World Health Organization (WHO) to declare on 11 June 2009 that a pandemic of pH1N1/2009v influenza was under way, which raised major international concern about the risk of high morbidity and lethality and the potential for severe socio-economic impact. Actually, the potential impact of this first third-millenium influenza pandemic has been revisited downwards as morbidity and case-fatality rates were less severe than initially anticipated [2] . Illness surveillance data do not allow to an accurate estimate of the true influenza infection rate, as a substantial proportion of infections are asymptomatic or mild [3] . Serological surveys can overcome this limitation, but must take into account that a significant proportion of the population that exhibited crossprotective antibody titers before circulation of the pH1N1/2009v [4] . This so-called ''baseline immunity'' has to be subtracted from the seroprevalence observed after the pandemic wave, to determine seroincidence in serosurveys [5] [6] [7] [8] . However, except for few studies [9] [10] [11] , most of these serosurveys did not use serial measurements in the same person, which allows for a better understanding of antibody kinetics and the dynamics of infection within individuals and communities.
Reunion Island (805,500 inhabitants) is a French overseas department located in the southwestern Indian Ocean, 700 km east of Madagascar and 200 km southwest of Mauritius. The first imported case of pH1N1/2009v was identified on 5 th July 2009 (week 29) in a traveller returning from Australia. The first case indicating community transmission was detected on 21 st July (week 30). pH1N1/2009v became the predominant circulating influenza virus within four weeks of its first detection, its activity peaked during week 35 (24) (25) (26) (27) (28) (29) (30) and ended at week 38 [12] . Contrary to initial fears, the health care system was not overwhelmed, as morbidity and mortality rates were lower than predicted [12] [13] [14] .
In order to assess at the community level, the actual magnitude of the pH1N1/2009v pandemic and the extent of the herd immunity acquired after passage of the epidemic wave, a prospective population serosurvey was conducted in Reunion Island during the passage of the epidemic wave in the 2009 austral winter season (July-December 2009): prevalence of infection was assessed on a weekly basis and seroconversion rates were measured using paired sera.
The CoPanFLu-RUN was part of the CoPanFLu international project, a consortium between the French National Institute of Health and Medical Research (INSERM), the Institute of Research for Development (IRD) and the Mérieux Fondation under the promotion of the School of Advanced Studies in Public Health (EHESP). To enable the rapid implementation of the study in anticipation of the imminent spread of the pandemic wave, we used a pre-existing sample of 2442 households established in October 2006 for the investigation of the Chikungunya outbreak (SEROCHIK) and updated in May 2008 throughout a follow-up telephone survey (TELECHIK) on a basis of 1148 households [15, 16] . We took special attention to select households representing a wide range of geographic locations in order to minimize the repartition bias.
The inclusion phase started on July 21 st (week 30) and was continued up to week 44, throughout the epidemic wave and beyond. A first serum sample (sample 1) was obtained from each household member. An active telephonic inquiry was then conducted twice a week to record symptoms compatible with influenza-like illness (ILI) occurring in households. Report of ILI (fever $37.8uC associated with any respiratory or systemic symptom) led to three consecutive visits of a nurse to the incident case-dwelling (on day 0, +3 and +8 post-report) to record symptoms and collect nasal swabs from all family members (for qRT-PCR detection of pH1N1/2009v. At week 45, the active inquiry was discontinued and a second (post-epidemic) serum sample (sample 2) was obtained (weeks 45-52) to determine seroconversion rates. Sera were aliquoted and stored at 280uC.
The protocol was conducted in accordance with the Declaration of Helsinki and French law for biomedical research (Nu ID RCB AFSSAPS: 2009-A00689-48) and was approved by the local Ethics Committee (Comité de Protection des Personnes of Bordeaux 2 University). Every eligible person for participation was asked for giving their written informed consent.
Viral genome detection by RT-PCR. Viral RNA was extracted from 140 mL of nasal swab eluate using the QIAamp Viral RNA kit (Qiagen) and processed for detection by TaqMan qRT-PCR targeting the heamagglutinin HA gene (SuperScript III Platinum one-step qRT-PCR system, Invitrogen) according to the recommendations of the Pasteur Institute (Van der Werf S. & Enouf V., SOP/FluA/130509). Confirmed pH1N1/2009v infection was defined as a positive qRT-PCR detection of the HA gene in at least one nasal swab.
Hemagglutination inhibition assay (HIA). A standard hemagglutination inhibition technique was adapted to detect and quantify pH1N1/2009v antibodies [17] . The antigen was prepared by diluting a non-inactivated cell culture supernatant producing a pdm H1N1v strain (strain OPYFLU-1 isolated from a young patient returning from Mexico in early May 2009) [18] . Briefly, the virus was propagated onto MDCK cells under standard conditions. The last passage (used for antigen preparation) was performed in the absence of trypsin and ht-FBS. The supernatant was collected at day seven p.i. clarified by centrifugation at 8006 g for 10 min at room temperature, aliquoted and conserved at 280uC. The hemagglutinating titer of the non inactivated viral antigen was immediately determined under the HIA format described below. The dilution providing 5.33 hemagglutinating units in a volume of 25 mL was used for subsequent HIA. Sera were heat-inactivated at 56uC for 30 min prior to use. Sequential twofold dilutions in PBS (1/10 to 1/1280) in volumes of 25 mL were performed and distributed in V-bottom 96 well microplates. Human red blood cells (RBC) were used for hemagglutination experiments. Detection and quantification of antibody to pH1N1/2009v was performed as follows: 25 mL of virus suspension was added to the serum dilution (25 mL) and incubated for 1 hour at room temperature. Each well was then filled with 25 mL of a 1% RBC suspension in PBS (v/v: 0.33%), followed by another 30 min incubation at room temperature. The HIA titer was determined as the last dilution providing clear inhibition of hemagglutination. All experiments were performed in the presence of the same negative and positive controls, the latter including sera with 1/40, 1/80, 1/160 and 1/320 antibody titers.
The results reported in this study were based only on serological analysis of paired sera. For the sake of analysis, four successive phases were identified throughout the pandemic wave: phase A (weeks 30-31) corresponded to early epidemic time, phase B (W32-39) to the epidemic unfolding, phase C (W40-44) to the immediate post-epidemic stage and phase D (W45-52) to the late post-epidemic stage. Seropositivity was defined as a HIA titer of 1/ 40 or more. The baseline-proxy seroprevalence rate was estimated on serum samples collected in phase A. The cumulative incidence rate of infection measured the raise between the raw seroprevalence rate at any given time during the epidemic phases (S2pi) and the age-specific baseline-proxy seroprevalence rate (S1pA) (s2 pi -s1 pA ). Seroconversion was defined as a shift from seronegative at inclusion (sample 1: HIA ,1/40) to seropositive on follow-up (sample 2: HIA $1/40), or for sera tested seropositive on inclusion as a four-fold increase of HIA titers between sample 1 and sample 2 paired sera. We also calculated the proportion of sera that tested seropositive in sample 1 for which the HIA titer decreased fourfold and passed under the cut-off value of 1/40 in sample 2. We considered this proportion as a ''seronegation'' rate.
The sample size was calculated for identifying risk factors in the prospective cohort study. Considering on average three individuals per household, an intra-household correlation of 0.3, a power greater than 80% could be obtained with a sample size of 840 comprising 2500 individuals, assuming exposure levels ranging from 10% to 90% and a relative risk greater than 1.3. With 2,500 subjects, the study allowed 1-2% absolute precision around the estimated values for seroconversion rates. Data entry used EpiData version 3.1 (The Epidata Association, Odense, Denmark). SAS version 9.1 (SAS Inc., Cary, NC, USA) was used for statistical analysis. The characteristics of the study cohort were compared to those of the population of Reunion Island and a Chi2 test (or Fisher's exact test when non applicable) was used to analyse differences in age, sex and geographic location. Cumulative incidence rates of infection (i.e. seroincidence) and seroconversion rates were standardized according to the age structure of the community (French National Institute for Statistics and Economical Studies (INSEE) source).
Baseline-proxy seroprevalence, cumulative incidence rates of infection, as well as seroconversion and seronegation rates, were expressed as percentages. Cumulative reverse distribution curves were used to show the distribution of antibody titers. In all tests, a P value,0.05 was considered significant.
We estimated 95% confidence intervals (CIs) of proportions by using a cluster bootstrap technique with 1000 re-samples [19] . After bootstraping, we used an ANOVA model to compare mean cumulative incidence proportions between pandemic phases, within each age group. We used an alternating logistic regression model (ALR) with an exchangeable log Odds Ratio (OR) to test the intra-household correlation-adjusted association between factors and the seroconversion outcome.
Data were analysed with respect to subject age. Initially, four age groups were considered: the children and adolescents (,20 yrs), young adults (20-39 yrs), middle-age adults (40-59 yrs), and elderly adults ($60 yrs). As the cumulative incidence of infection of the second and third groups were very close, both groups were merged into one adults group (20-59 yrs). Therefore we refer further in our study to three age groups: children and adolescents (,20 yrs), adults (20-59 yrs), elderly ($60 yrs).
A total of 2,164 individuals from 772 households were enrolled between weeks 30 and 44 in the CoPanFlu-RUN cohort, allowing the collection of 1,932 sera at inclusion (sample 1). During this period, 136 households (17.7% of households) containing 464 individuals (21.4% of individuals) reported at least one case of ILI. Sixty subjects among the 464 individuals (12.9%, belonging to 33 households [24.3%]) were qRT-PCR positive, which documented the pH1N1/2009v infection. No positive qRT-PCR could be detected after week 37 and no ILI was reported after week 40, the end of the epidemic wave. The second follow up serum sample (sample 2) was obtained for 1,759 subjects at least five weeks after the end of the epidemic wave (weeks 45-52) which allowed the constitution of a serobank of 1,687 paired-sera. The profile of the cohort and the major outcomes are displayed in Figure 1 . Details on inclusions and serum sample timing with respect to the circulation of pH1N1/2009v over the island are provided in figure 2 . The socio-demographic and space-time characteristics of the cohort are detailed in Table 1 . Compared to the community of Reunion Island, the sample of 1,687 individuals for whom pairedsera were available, was older (,20 yrs: 27% vs 35%, and $60 yrs: 17,9% vs 11,3%) and composed of a slight excess of females (54.1% vs 51.5%). The imbalance was due to a deficit in subjects aged under 40 years, reflecting men at work and the fact that parents declined the second serum for children younger than five.
Baseline-proxy (,pre-epidemic) HIA titers to the pH1N1/ 2009v were measured on sample 1 ( Table 2) , obtained from 249 subjects (103 households) recruited at the very beginning of the investigation during weeks 30 and 31 (phase A, Figure 2 ), when the epidemic activity in the cohort was still very low. Age distribution in this group was similar to that of the whole cohort (data not shown). The overall, the baseline-proxy seroprevalence rate (HIA $1/40), over all ages, was 43.4% (95%CI: 37.4%-49.6%). However the majority of positive sera had low antibody titers, at the cut off value for confirmation (i.e. = 1/40). The proportions of sera with HIA titer .1/40 were 0%, 3.0% and 24.6% in the young, middle-aged and older age groups respectively. These results indicate that pre-epidemic baseline antibody cross reactivity was stronger in the elderly ($60 yrs) and weaker in children and adolescents (,20 yrs) and adults (20-59 yrs), with highly significant differences between age groups (P,0.0001).
The reverse cumulative distribution curves of HIA titers are displayed for each age group and for the whole cohort on Figure 3 . The proportion of seropositive sera (HI $1/40) steadily increased during the epidemic unfolding (phase B, W32-39) and in immediate post epidemic period (phase C, W40-44) when it reached its maximum level, then declined in the late post epidemic period (phase D, W45-52). This decline was significant enough to return the reverse cumulative distribution curve to baseline levels in the elderly. The cumulative incidence rates, obtained after subtraction of the age-specific baseline-proxy seroprevalence from the raw seroprevalence at each phase of the epidemic are shown in Table 2 (note that the cumulative incidence rates of infection represented for the group ''all ages'' were standardized according to age structure of the community). The cumulative incidence rates were much higher in children and adolescents (,20 yrs), indicating very active transmission of infection within this age group. As mentioned earlier, cumulative incidence rates peaked in phase C (W40-44), and then declined indicating some lability of the humoral immune response against the pH1N1/2009v. The age-related difference observed in the incidence rates was highly statistically significant (P,0.0001).
To estimate more appropriately the decline of antibody titers occurring after the peak of the humoral response to the pH1N1/ 2009v, we considered paired-sera from the group of 264 subjects for whom the first serum sample (sample 1) was obtained just after the epidemic wave (phase C, W40-44), and the corresponding second sample was collected at the end of the survey (phase D, W45-52). Seronegation rates were 27.0% (61/226) for all age groups, 17.4% (12/69) in children and adolescents (,20 yrs), 32.3% (41/127) in adults (20-59 yrs) and 26.7% (8/30) in the elderly ($60 yrs). Differences between the seronegation rates according to age were statistically weakly significant (P = 0.0671).
We then considered the 1687 individuals for whom paired sera were available and we measured the seroconversion rates according to age and to the time of first serum sample collection (phase A, B or C). Criteria of seroconversion were defined in the method section. As shown in table 3, there was a sharp decline in seroconversion rates across all the age groups, depending on whether participants were enrolled during phase A, phase B, or phase C (P,0.0001). To interpret these data, one should remember that antibodies at seroprotective levels (HIA $1/40), in serum samples 1 collected during the per epidemic phase B or early post epidemic phase C could represent either base line cross reactive antibodies or rising pH1N1/2009 specific antibodies due to a recent or ongoing infection. This ambiguity could lead to underestimation of the seroconversion rate for subjects enrolled in phases B and C. In order to solve this ambiguity, we specifically considered the group of 249 subjects in whom cross reactive antibodies were detected at the time of phase A (W30-31). The seroconversion rate of this group is the most indicative of the exposure of individuals to the whole epidemic wave. It was the highest (63,2%, P,0.0001) in children and adolescents (,20 yrs), and still significantly high in adults (39.4%, P,0.0001).
We then tested in this particular group, the impact of (baseline) pre-epidemic cross reactive antibodies on the rate of seroconversion to pH1N1/2009 (Table 4) . No subject with HIA titer superior to 1/40 had evidence of seroconversion to pH1N1/2009. The seroconversion rate in individuals with a HIA titer equal to 1/40 was linked with age, being more important in children and adolescents (,20 yrs). The highest seroconversion rate (.56%) was registered in subjects with HIA titers inferior to 1/40, particularly for the under 20 years where it reached 85%. Hence, the risk of seroconversion decreased when pre-epidemic HIA titer was high after controlling for age (P,0.0001) (Figure 4) . The multivariate adjusted odds ratio for seroconversion were 0.15 (95%CI: 0.06-0.37, P,0.0001) per two-fold increase in baseline titer, 1.79 (95%CI: 1.23-2.59, P,0.003) per other household members who seroconverted, 5.33 (95%CI: 1.56-19.27, P,0.008) Figure 1 . The cohort profile and major outcomes. Figure 1 details the three phases of the protocol: i) inclusion (weeks 30-44) and serum samples S1 collection; ii) follow up for detection of ILI in households, qRT-PCR on nasal swabs and estimation of cumulative seroincidence rates; iii) end of the study (weeks 45-52) and samples S2 collection. HIA on paired sera (S1+S2) allowed estimating seroconversion rates. doi:10.1371/journal.pone.0025738.g001 Bp (baseline-proxy) seroprevalence rates were estimated on weeks 30-31 in each age group. b Cumulative incidence rates measured the raise between raw seroprevalence rates and age-specific baseline-proxy seroprevalence rate. In the group ''All ages'', cumulative incidence rates were standardized according to age structure of the community. doi:10.1371/journal.pone.0025738.t002 Data are numbers, percentages (95% confidence intervals) and ALR parameter test P value for comparison of seroconversion proportions according to time of first sample (S1) collection at inclusion, in each age group, after controlling for household selection. In the group ''All ages'', rates of seroconversion were standardized according to age structure of the community. NA: not assessed. Seroconversion was defined as a shift from seronegative at inclusion (i.e. HIA titer ,1/40) to seropositive on follow-up sample, or as a 4-fold increase of reciprocal HIA titer between first and second paired samples for sera tested seropositive on inclusion (i.e. HIA titer $1/40). for age ,20 years (vs age $60 years) and 11.35 (95%CI: 0.41-4.47, P = 0.62) for age 20-60 years (vs age $60 years). The observed and predicted seroconversion rates according to age and baseline HIA titer are displayed Figure 4 .
Finally, we considered the 46 subjects who had been infected by the pandemic virus over the course of the study, verified by a positive qRT-PCR nasal swab, and for whom paired sera were available. Initial HIA antibody titers in this group were ,1/40,
The CoPanFlu-RUN cohort was set up to conduct a prospective population-based study investigating the herd immunity induced by the 2009 pandemic influenza virus and identifying risk factors for pH1N1/2009v infection from paired sera collected in an entire community. Most works published to date have used either extensive cross-sectional serosurveys on pre-and post-epidemic independent serum samples, the baseline immunity being assessed from stored frozen samples [5, 7, 8] , or non representative adult cohorts (military, health care workers, long-stay patients). Antibody titers were measured by HIA using a cut-off value set at 1/40 as classically recommended. This HIA titer at 1/40 is considered protective, i.e. conferring 50% protection against a viral challenge [20] . Our assay has introduced some changes in the experimental protocol compared to the classic one. The use of a non-inactivated viral antigen, i.e. a native virus, with nondenatured epitopes probably allows detection of antibodies to epitopes of the hemagglutinin not detected in the classic HIA test. This can induce slight differences in the sensitivity of detection of cross-reacting antibodies, but this does not modify the kinetics of Ab and the epidemiological evolution of seroprevalence and does not jeopardize the global comparability of serological results. This is confirmed by the fact that our HI assay detected seroprotective antibody titers in 93.5% and gave evidence seroconversion in 73.9% of qRT-PCR confirmed pH1N1/2009 influenza, all figures close to those reported in the literature [5, 21] .
We considered that titers of .1/40, in sera collected from individuals enrolled during weeks 30 and 31 were cross reactive antibodies and not de novo antibodies triggered by the pandemic virus and hence used them as a proxy for baseline pre epidemic immunity. Several arguments support this assumption: i) the first case indicating autochthonous transmission in Reunion Island was reported by the epidemiological surveillance department of La Réunion on 21st July (week 30), i.e. the same day when inclusion started in our study cohort; ii) 7 to 15 days are required to develop an antibody response after viral infection; iii) On weeks 30 and 31, the epidemic activity due to the pandemic virus was very low in our study cohort and it became significant only after week 32. Hence, during weeks 30-31, 103 households were recruited and only 2 households reported ILI cases. Nasal swabs collected from these 2 individuals were tested qRT-PCR negative to the pandemic virus whereas one had evidence of coronavirus and rhinovirus using a multiplex RT-PCR to respiratory viruses (H. Pascalis, manuscript in preparation). In contrast, during weeks 32 to 39, 199 individuals belonging to 99 households reported ILI, among whom 60 individuals had documented infection by the pandemic virus.
Our study shows that a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus with the highest baseline-proxy seroprevalence rate observed among adults aged of 60 years or more. Other studies from all continents had also reported high pre-epidemic seropositivity rates among the elderly [5, 6, 8, [22] [23] [24] [25] [26] , though large variations do exist between countries [10, 11, 23, 27, 28] . These cross reactive antibodies have been interpreted as being the residual signature of the remote exposure of these individuals to H1N1 viruses circulating before 1957 [24, 25, 29, 30] . Baseline seropositivity rates that we report in children and in younger adults (i.e. 30%-35%) were notably higher than those reported from other parts of the world [6, 8, 22, 23, [31] [32] [33] . However one should note that these baseline antibodies were of low titer, just at the level of the HIA threshold (i.e. 1/40). Several factors could have contributed to this comparatively high baseline rates found in our study: i) It may reflect the fact that the HI test used in our study was marginally more sensitive than the classic one [17] ; ii) Some individuals may have already been infected with pH1N1/ 2009 virus at weeks 30 and 31 and may have triggered an antibody response to the virus. This hypothesis seems unlikely in view of the arguments presented above and of a similar high proportion of sera titering HIA = 1/40 among 122 sera from adult patients sent for diagnostic purposes to the Regional Hospital microbiology laboratory, during the first half of 2009 (i.e. before the 2009 pandemic) (data not shown). However we cannot formally exclude this hypothesis in view of a recently reported study from Taiwan [11] that showed evidence of subclinical community transmission with proved seroconversion several weeks before report of the first documented case in the island. A similar conclusion was also drawn from Australia [34] ; iii) our serological test might detect cross-reactive antibodies triggered by recent vaccination with trivalent seasonal influenza vaccine as reported [4, [35] [36] [37] [38] [39] . However, seasonal influenza vaccines were of rather limited use in Reunion Island, especially in children and young adults; iv) Finally the high baseline titers may reflect the infectious history of the individuals to seasonal influenza viruses cross antigenic with pH1N1/2009 virus as recently suggested for seasonal 2007 H1N1 infection [40] . This serosurvey indicates that a large fraction of the Reunion Island population was infected with the pandemic virus. Younger people, have paid the main tribute to the epidemic as almost two thirds show evidence of seroconversion, confirming earlier clinical reports from the island [12] and accumulating reports from other countries [17, 32, 41, 42] and suggesting that school children have likely played the central role in the epidemic diffusion of the pandemic virus. Lower infection rates were found in adults and the lowest rates were recorded in the elderly.
Based on clinical cases reported to the epidemiological surveillance services [12] , it was estimated that 66,915 persons in Reunion Island who consulted a physician were infected by the pH1N1/2009 virus during the 9 weeks of the epidemic, giving a cumulative attack rate of 8.26%. Taking into account those who did not consult a physician, the number of symptomatic infected persons was estimated to 104,067 (attack rate: 12.85%). In fact, the attack rate of pH1N1/2009 infection in our serosurvey was about 42%-44% at the peak of the antibody response (i.e., weeks 40-44), a figure which is at least 3 to 4 times higher than rates of infection based on clinical cases The wide gap between the two estimates indicates that a large fraction (almost two thirds) of those who got infected by pH1N1/2009 virus escaped medical detection, probably because they developed mild disease or asymptomatic infection, a further indication of the benign nature of the virus, at least at the community level. In England, Baguelin et al. [43] estimated that the cumulative incidence rates of infection by the pandemic virus in children were 20 to 40 times higher than that estimated from clinical surveillance.
Our study, as others [6] , indicates that pre-existing cross reactive antibodies to pH1N1/2009 at titers $1/40 prevented from seroconversion in response to the pandemic virus. This level of pre-existing cross reactive immunity likely confers true protection against infection as about two thirds and one third of documented infection (qRT-PCR positive) in our series have occurred in individuals with baseline HIA titers ,1/40 and = 1/ 40 respectively and less than 5% of documented infections occurred in individuals with base line titers .1/40. The protection was effective not only in older adults but also in younger persons. This indicates that protection was conferred not only by baseline cross reactive antibodies triggered by close pH1N1/2009 viruses that circulated before 1957 (as in the elderly), but also by antibodies likely resulting from recent exposure to seasonal influenza epidemics (as shown in younger persons) [40] . The observed seroconversion rates depend on age, after adjusting for baseline pH1N1/2009 titers. The protective role of increasing age might be explained by a stronger cross-immunity in adults and elderly or by a higher exposure of young subjects to the virus during the 2009 epidemic (due to social contacts and mixing patterns). It may also indicate that immune mechanisms other than cross reactive antibodies detected by HIA (i.e. immunity to neuraminidase and conserved T cells epitopes [44] might develop throughout life, providing additional protection from infection or severe disease, especially in the elderly. Interestingly, evidence is seen for a decline in antibody titers, which occurred soon after the passage of the epidemic wave. In paired sera, this decline was significant enough to bring, within a few weeks, almost 27% of sera that tested positive (i.e. HI titers $1/40) in the immediate post epidemic phase to levels under the cut-off value in the second serum sample. This decay accounts for the observation that older adults ($60 yrs) in the study cohort were apparently almost completely spared by the epidemic if one only considers cumulative incidence rates derived from IHA titration on samples 2 (weeks 45-52). In fact, the cumulative incidence rate in older adults measured just after the epidemic peak (i.e. weeks 40-44) was 20.4%. Similar results of early antibody decay were recently reported [10, 45] . More generally, these data show that serosurveys conducted months after passage of the epidemic, likely underestimate the real extent of pH1N1/2009 infection, compared to antibody titration performed earlier, when humoral responses are at their highest level. Whether the decline in antibody titers has functional immunologic consequence to individuals or within the communities warrants further investigation. However, one should note that there was no second epidemic wave in Reunion Island during the subsequent austral winter seasons in 2010 and 2011.
Influenza during the 2010 winter was at a level not higher than the usual passages of seasonal flu, though almost two thirds of documented cases in 2010 were also due to pH1N1/2009v [46] . In addition many fewer pandemic virus isolates were noted during the ongoing 2011 austral winter, strongly suggesting that the first epidemic wave had conferred a solid herd immunity, at the community level.
Our study has some limitations. The fact that the epidemic progression coincided with the implementation of the prospective study, we were not able to collect, strictly speaking, pre-epidemic sera from the cohort members. Therefore we used as proxy base line seroprevalence data from individuals recruited at the very beginning of the investigation when the epidemic activity in the cohort was very low. This may overestimate the base line immunity if subclinical community transmission had occurred before the first cases of pH1N1/2009 influenza were reported. Antibodies to the pandemic virus were detected by HIA, a test that has a good specificity but a rather low sensitivity [46] . Hence, the threshold of 1/40 may underestimate the number of infected individuals. However, rates of seroconversion, the serologic gold standard test based on paired sera, likely gave the most accurate picture of the pandemic in at the community level in Reunion Island.
|
What was the interpretation for the crossreactive antibodies?
| 5,260 |
the remote exposure of these individuals to H1N1 viruses circulating before 1957
| 25,486 |
1,601 |
Pandemic Influenza Due to pH1N1/2009 Virus: Estimation of Infection Burden in Reunion Island through a Prospective Serosurvey, Austral Winter 2009
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3183080/
SHA: ee6d70a53e3262cea6f85bd8b226f6b4c8b5f64b
Authors: Dellagi, Koussay; Rollot, Olivier; Temmam, Sarah; Salez, Nicolas; Guernier, Vanina; Pascalis, Hervé; Gérardin, Patrick; Fianu, Adrian; Lapidus, Nathanael; Naty, Nadège; Tortosa, Pablo; Boussaïd, Karim; Jaffar-Banjee, Marie-Christine; Filleul, Laurent; Flahault, Antoine; Carrat, Fabrice; Favier, Francois; de Lamballerie, Xavier
Date: 2011-09-29
DOI: 10.1371/journal.pone.0025738
License: cc-by
Abstract: BACKGROUND: To date, there is little information that reflects the true extent of spread of the pH1N1/2009v influenza pandemic at the community level as infection often results in mild or no clinical symptoms. This study aimed at assessing through a prospective study, the attack rate of pH1N1/2009 virus in Reunion Island and risk factors of infection, during the 2009 season. METHODOLOGY/PRINCIPAL FINDINGS: A serosurvey was conducted during the 2009 austral winter, in the frame of a prospective population study. Pairs of sera were collected from 1687 individuals belonging to 772 households, during and after passage of the pandemic wave. Antibodies to pH1N1/2009v were titered using the hemagglutination inhibition assay (HIA) with titers ≥1/40 being considered positive. Seroprevalence during the first two weeks of detection of pH1N1/2009v in Reunion Island was 29.8% in people under 20 years of age, 35.6% in adults (20–59 years) and 73.3% in the elderly (≥60 years) (P<0.0001). Baseline corrected cumulative incidence rates, were 42.9%, 13.9% and 0% in these age groups respectively (P<0.0001). A significant decline in antibody titers occurred soon after the passage of the epidemic wave. Seroconversion rates to pH1N1/2009 correlated negatively with age: 63.2%, 39.4% and 16.7%, in each age group respectively (P<0.0001). Seroconversion occurred in 65.2% of individuals who were seronegative at inclusion compared to 6.8% in those who were initially seropositive. CONCLUSIONS: Seroincidence of pH1N1/2009v infection was three times that estimated from clinical surveillance, indicating that almost two thirds of infections occurring at the community level have escaped medical detection. People under 20 years of age were the most affected group. Pre-epidemic titers ≥1/40 prevented seroconversion and are likely protective against infection. A concern was raised about the long term stability of the antibody responses.
Text: In April 2009, the first cases of acute respiratory infections caused by a novel triple-reassortant influenza virus, pH1N1/ 2009v, occurred in Mexico and the United States [1] . The rapid spread of infection to other continents led the World Health Organization (WHO) to declare on 11 June 2009 that a pandemic of pH1N1/2009v influenza was under way, which raised major international concern about the risk of high morbidity and lethality and the potential for severe socio-economic impact. Actually, the potential impact of this first third-millenium influenza pandemic has been revisited downwards as morbidity and case-fatality rates were less severe than initially anticipated [2] . Illness surveillance data do not allow to an accurate estimate of the true influenza infection rate, as a substantial proportion of infections are asymptomatic or mild [3] . Serological surveys can overcome this limitation, but must take into account that a significant proportion of the population that exhibited crossprotective antibody titers before circulation of the pH1N1/2009v [4] . This so-called ''baseline immunity'' has to be subtracted from the seroprevalence observed after the pandemic wave, to determine seroincidence in serosurveys [5] [6] [7] [8] . However, except for few studies [9] [10] [11] , most of these serosurveys did not use serial measurements in the same person, which allows for a better understanding of antibody kinetics and the dynamics of infection within individuals and communities.
Reunion Island (805,500 inhabitants) is a French overseas department located in the southwestern Indian Ocean, 700 km east of Madagascar and 200 km southwest of Mauritius. The first imported case of pH1N1/2009v was identified on 5 th July 2009 (week 29) in a traveller returning from Australia. The first case indicating community transmission was detected on 21 st July (week 30). pH1N1/2009v became the predominant circulating influenza virus within four weeks of its first detection, its activity peaked during week 35 (24) (25) (26) (27) (28) (29) (30) and ended at week 38 [12] . Contrary to initial fears, the health care system was not overwhelmed, as morbidity and mortality rates were lower than predicted [12] [13] [14] .
In order to assess at the community level, the actual magnitude of the pH1N1/2009v pandemic and the extent of the herd immunity acquired after passage of the epidemic wave, a prospective population serosurvey was conducted in Reunion Island during the passage of the epidemic wave in the 2009 austral winter season (July-December 2009): prevalence of infection was assessed on a weekly basis and seroconversion rates were measured using paired sera.
The CoPanFLu-RUN was part of the CoPanFLu international project, a consortium between the French National Institute of Health and Medical Research (INSERM), the Institute of Research for Development (IRD) and the Mérieux Fondation under the promotion of the School of Advanced Studies in Public Health (EHESP). To enable the rapid implementation of the study in anticipation of the imminent spread of the pandemic wave, we used a pre-existing sample of 2442 households established in October 2006 for the investigation of the Chikungunya outbreak (SEROCHIK) and updated in May 2008 throughout a follow-up telephone survey (TELECHIK) on a basis of 1148 households [15, 16] . We took special attention to select households representing a wide range of geographic locations in order to minimize the repartition bias.
The inclusion phase started on July 21 st (week 30) and was continued up to week 44, throughout the epidemic wave and beyond. A first serum sample (sample 1) was obtained from each household member. An active telephonic inquiry was then conducted twice a week to record symptoms compatible with influenza-like illness (ILI) occurring in households. Report of ILI (fever $37.8uC associated with any respiratory or systemic symptom) led to three consecutive visits of a nurse to the incident case-dwelling (on day 0, +3 and +8 post-report) to record symptoms and collect nasal swabs from all family members (for qRT-PCR detection of pH1N1/2009v. At week 45, the active inquiry was discontinued and a second (post-epidemic) serum sample (sample 2) was obtained (weeks 45-52) to determine seroconversion rates. Sera were aliquoted and stored at 280uC.
The protocol was conducted in accordance with the Declaration of Helsinki and French law for biomedical research (Nu ID RCB AFSSAPS: 2009-A00689-48) and was approved by the local Ethics Committee (Comité de Protection des Personnes of Bordeaux 2 University). Every eligible person for participation was asked for giving their written informed consent.
Viral genome detection by RT-PCR. Viral RNA was extracted from 140 mL of nasal swab eluate using the QIAamp Viral RNA kit (Qiagen) and processed for detection by TaqMan qRT-PCR targeting the heamagglutinin HA gene (SuperScript III Platinum one-step qRT-PCR system, Invitrogen) according to the recommendations of the Pasteur Institute (Van der Werf S. & Enouf V., SOP/FluA/130509). Confirmed pH1N1/2009v infection was defined as a positive qRT-PCR detection of the HA gene in at least one nasal swab.
Hemagglutination inhibition assay (HIA). A standard hemagglutination inhibition technique was adapted to detect and quantify pH1N1/2009v antibodies [17] . The antigen was prepared by diluting a non-inactivated cell culture supernatant producing a pdm H1N1v strain (strain OPYFLU-1 isolated from a young patient returning from Mexico in early May 2009) [18] . Briefly, the virus was propagated onto MDCK cells under standard conditions. The last passage (used for antigen preparation) was performed in the absence of trypsin and ht-FBS. The supernatant was collected at day seven p.i. clarified by centrifugation at 8006 g for 10 min at room temperature, aliquoted and conserved at 280uC. The hemagglutinating titer of the non inactivated viral antigen was immediately determined under the HIA format described below. The dilution providing 5.33 hemagglutinating units in a volume of 25 mL was used for subsequent HIA. Sera were heat-inactivated at 56uC for 30 min prior to use. Sequential twofold dilutions in PBS (1/10 to 1/1280) in volumes of 25 mL were performed and distributed in V-bottom 96 well microplates. Human red blood cells (RBC) were used for hemagglutination experiments. Detection and quantification of antibody to pH1N1/2009v was performed as follows: 25 mL of virus suspension was added to the serum dilution (25 mL) and incubated for 1 hour at room temperature. Each well was then filled with 25 mL of a 1% RBC suspension in PBS (v/v: 0.33%), followed by another 30 min incubation at room temperature. The HIA titer was determined as the last dilution providing clear inhibition of hemagglutination. All experiments were performed in the presence of the same negative and positive controls, the latter including sera with 1/40, 1/80, 1/160 and 1/320 antibody titers.
The results reported in this study were based only on serological analysis of paired sera. For the sake of analysis, four successive phases were identified throughout the pandemic wave: phase A (weeks 30-31) corresponded to early epidemic time, phase B (W32-39) to the epidemic unfolding, phase C (W40-44) to the immediate post-epidemic stage and phase D (W45-52) to the late post-epidemic stage. Seropositivity was defined as a HIA titer of 1/ 40 or more. The baseline-proxy seroprevalence rate was estimated on serum samples collected in phase A. The cumulative incidence rate of infection measured the raise between the raw seroprevalence rate at any given time during the epidemic phases (S2pi) and the age-specific baseline-proxy seroprevalence rate (S1pA) (s2 pi -s1 pA ). Seroconversion was defined as a shift from seronegative at inclusion (sample 1: HIA ,1/40) to seropositive on follow-up (sample 2: HIA $1/40), or for sera tested seropositive on inclusion as a four-fold increase of HIA titers between sample 1 and sample 2 paired sera. We also calculated the proportion of sera that tested seropositive in sample 1 for which the HIA titer decreased fourfold and passed under the cut-off value of 1/40 in sample 2. We considered this proportion as a ''seronegation'' rate.
The sample size was calculated for identifying risk factors in the prospective cohort study. Considering on average three individuals per household, an intra-household correlation of 0.3, a power greater than 80% could be obtained with a sample size of 840 comprising 2500 individuals, assuming exposure levels ranging from 10% to 90% and a relative risk greater than 1.3. With 2,500 subjects, the study allowed 1-2% absolute precision around the estimated values for seroconversion rates. Data entry used EpiData version 3.1 (The Epidata Association, Odense, Denmark). SAS version 9.1 (SAS Inc., Cary, NC, USA) was used for statistical analysis. The characteristics of the study cohort were compared to those of the population of Reunion Island and a Chi2 test (or Fisher's exact test when non applicable) was used to analyse differences in age, sex and geographic location. Cumulative incidence rates of infection (i.e. seroincidence) and seroconversion rates were standardized according to the age structure of the community (French National Institute for Statistics and Economical Studies (INSEE) source).
Baseline-proxy seroprevalence, cumulative incidence rates of infection, as well as seroconversion and seronegation rates, were expressed as percentages. Cumulative reverse distribution curves were used to show the distribution of antibody titers. In all tests, a P value,0.05 was considered significant.
We estimated 95% confidence intervals (CIs) of proportions by using a cluster bootstrap technique with 1000 re-samples [19] . After bootstraping, we used an ANOVA model to compare mean cumulative incidence proportions between pandemic phases, within each age group. We used an alternating logistic regression model (ALR) with an exchangeable log Odds Ratio (OR) to test the intra-household correlation-adjusted association between factors and the seroconversion outcome.
Data were analysed with respect to subject age. Initially, four age groups were considered: the children and adolescents (,20 yrs), young adults (20-39 yrs), middle-age adults (40-59 yrs), and elderly adults ($60 yrs). As the cumulative incidence of infection of the second and third groups were very close, both groups were merged into one adults group (20-59 yrs). Therefore we refer further in our study to three age groups: children and adolescents (,20 yrs), adults (20-59 yrs), elderly ($60 yrs).
A total of 2,164 individuals from 772 households were enrolled between weeks 30 and 44 in the CoPanFlu-RUN cohort, allowing the collection of 1,932 sera at inclusion (sample 1). During this period, 136 households (17.7% of households) containing 464 individuals (21.4% of individuals) reported at least one case of ILI. Sixty subjects among the 464 individuals (12.9%, belonging to 33 households [24.3%]) were qRT-PCR positive, which documented the pH1N1/2009v infection. No positive qRT-PCR could be detected after week 37 and no ILI was reported after week 40, the end of the epidemic wave. The second follow up serum sample (sample 2) was obtained for 1,759 subjects at least five weeks after the end of the epidemic wave (weeks 45-52) which allowed the constitution of a serobank of 1,687 paired-sera. The profile of the cohort and the major outcomes are displayed in Figure 1 . Details on inclusions and serum sample timing with respect to the circulation of pH1N1/2009v over the island are provided in figure 2 . The socio-demographic and space-time characteristics of the cohort are detailed in Table 1 . Compared to the community of Reunion Island, the sample of 1,687 individuals for whom pairedsera were available, was older (,20 yrs: 27% vs 35%, and $60 yrs: 17,9% vs 11,3%) and composed of a slight excess of females (54.1% vs 51.5%). The imbalance was due to a deficit in subjects aged under 40 years, reflecting men at work and the fact that parents declined the second serum for children younger than five.
Baseline-proxy (,pre-epidemic) HIA titers to the pH1N1/ 2009v were measured on sample 1 ( Table 2) , obtained from 249 subjects (103 households) recruited at the very beginning of the investigation during weeks 30 and 31 (phase A, Figure 2 ), when the epidemic activity in the cohort was still very low. Age distribution in this group was similar to that of the whole cohort (data not shown). The overall, the baseline-proxy seroprevalence rate (HIA $1/40), over all ages, was 43.4% (95%CI: 37.4%-49.6%). However the majority of positive sera had low antibody titers, at the cut off value for confirmation (i.e. = 1/40). The proportions of sera with HIA titer .1/40 were 0%, 3.0% and 24.6% in the young, middle-aged and older age groups respectively. These results indicate that pre-epidemic baseline antibody cross reactivity was stronger in the elderly ($60 yrs) and weaker in children and adolescents (,20 yrs) and adults (20-59 yrs), with highly significant differences between age groups (P,0.0001).
The reverse cumulative distribution curves of HIA titers are displayed for each age group and for the whole cohort on Figure 3 . The proportion of seropositive sera (HI $1/40) steadily increased during the epidemic unfolding (phase B, W32-39) and in immediate post epidemic period (phase C, W40-44) when it reached its maximum level, then declined in the late post epidemic period (phase D, W45-52). This decline was significant enough to return the reverse cumulative distribution curve to baseline levels in the elderly. The cumulative incidence rates, obtained after subtraction of the age-specific baseline-proxy seroprevalence from the raw seroprevalence at each phase of the epidemic are shown in Table 2 (note that the cumulative incidence rates of infection represented for the group ''all ages'' were standardized according to age structure of the community). The cumulative incidence rates were much higher in children and adolescents (,20 yrs), indicating very active transmission of infection within this age group. As mentioned earlier, cumulative incidence rates peaked in phase C (W40-44), and then declined indicating some lability of the humoral immune response against the pH1N1/2009v. The age-related difference observed in the incidence rates was highly statistically significant (P,0.0001).
To estimate more appropriately the decline of antibody titers occurring after the peak of the humoral response to the pH1N1/ 2009v, we considered paired-sera from the group of 264 subjects for whom the first serum sample (sample 1) was obtained just after the epidemic wave (phase C, W40-44), and the corresponding second sample was collected at the end of the survey (phase D, W45-52). Seronegation rates were 27.0% (61/226) for all age groups, 17.4% (12/69) in children and adolescents (,20 yrs), 32.3% (41/127) in adults (20-59 yrs) and 26.7% (8/30) in the elderly ($60 yrs). Differences between the seronegation rates according to age were statistically weakly significant (P = 0.0671).
We then considered the 1687 individuals for whom paired sera were available and we measured the seroconversion rates according to age and to the time of first serum sample collection (phase A, B or C). Criteria of seroconversion were defined in the method section. As shown in table 3, there was a sharp decline in seroconversion rates across all the age groups, depending on whether participants were enrolled during phase A, phase B, or phase C (P,0.0001). To interpret these data, one should remember that antibodies at seroprotective levels (HIA $1/40), in serum samples 1 collected during the per epidemic phase B or early post epidemic phase C could represent either base line cross reactive antibodies or rising pH1N1/2009 specific antibodies due to a recent or ongoing infection. This ambiguity could lead to underestimation of the seroconversion rate for subjects enrolled in phases B and C. In order to solve this ambiguity, we specifically considered the group of 249 subjects in whom cross reactive antibodies were detected at the time of phase A (W30-31). The seroconversion rate of this group is the most indicative of the exposure of individuals to the whole epidemic wave. It was the highest (63,2%, P,0.0001) in children and adolescents (,20 yrs), and still significantly high in adults (39.4%, P,0.0001).
We then tested in this particular group, the impact of (baseline) pre-epidemic cross reactive antibodies on the rate of seroconversion to pH1N1/2009 (Table 4) . No subject with HIA titer superior to 1/40 had evidence of seroconversion to pH1N1/2009. The seroconversion rate in individuals with a HIA titer equal to 1/40 was linked with age, being more important in children and adolescents (,20 yrs). The highest seroconversion rate (.56%) was registered in subjects with HIA titers inferior to 1/40, particularly for the under 20 years where it reached 85%. Hence, the risk of seroconversion decreased when pre-epidemic HIA titer was high after controlling for age (P,0.0001) (Figure 4) . The multivariate adjusted odds ratio for seroconversion were 0.15 (95%CI: 0.06-0.37, P,0.0001) per two-fold increase in baseline titer, 1.79 (95%CI: 1.23-2.59, P,0.003) per other household members who seroconverted, 5.33 (95%CI: 1.56-19.27, P,0.008) Figure 1 . The cohort profile and major outcomes. Figure 1 details the three phases of the protocol: i) inclusion (weeks 30-44) and serum samples S1 collection; ii) follow up for detection of ILI in households, qRT-PCR on nasal swabs and estimation of cumulative seroincidence rates; iii) end of the study (weeks 45-52) and samples S2 collection. HIA on paired sera (S1+S2) allowed estimating seroconversion rates. doi:10.1371/journal.pone.0025738.g001 Bp (baseline-proxy) seroprevalence rates were estimated on weeks 30-31 in each age group. b Cumulative incidence rates measured the raise between raw seroprevalence rates and age-specific baseline-proxy seroprevalence rate. In the group ''All ages'', cumulative incidence rates were standardized according to age structure of the community. doi:10.1371/journal.pone.0025738.t002 Data are numbers, percentages (95% confidence intervals) and ALR parameter test P value for comparison of seroconversion proportions according to time of first sample (S1) collection at inclusion, in each age group, after controlling for household selection. In the group ''All ages'', rates of seroconversion were standardized according to age structure of the community. NA: not assessed. Seroconversion was defined as a shift from seronegative at inclusion (i.e. HIA titer ,1/40) to seropositive on follow-up sample, or as a 4-fold increase of reciprocal HIA titer between first and second paired samples for sera tested seropositive on inclusion (i.e. HIA titer $1/40). for age ,20 years (vs age $60 years) and 11.35 (95%CI: 0.41-4.47, P = 0.62) for age 20-60 years (vs age $60 years). The observed and predicted seroconversion rates according to age and baseline HIA titer are displayed Figure 4 .
Finally, we considered the 46 subjects who had been infected by the pandemic virus over the course of the study, verified by a positive qRT-PCR nasal swab, and for whom paired sera were available. Initial HIA antibody titers in this group were ,1/40,
The CoPanFlu-RUN cohort was set up to conduct a prospective population-based study investigating the herd immunity induced by the 2009 pandemic influenza virus and identifying risk factors for pH1N1/2009v infection from paired sera collected in an entire community. Most works published to date have used either extensive cross-sectional serosurveys on pre-and post-epidemic independent serum samples, the baseline immunity being assessed from stored frozen samples [5, 7, 8] , or non representative adult cohorts (military, health care workers, long-stay patients). Antibody titers were measured by HIA using a cut-off value set at 1/40 as classically recommended. This HIA titer at 1/40 is considered protective, i.e. conferring 50% protection against a viral challenge [20] . Our assay has introduced some changes in the experimental protocol compared to the classic one. The use of a non-inactivated viral antigen, i.e. a native virus, with nondenatured epitopes probably allows detection of antibodies to epitopes of the hemagglutinin not detected in the classic HIA test. This can induce slight differences in the sensitivity of detection of cross-reacting antibodies, but this does not modify the kinetics of Ab and the epidemiological evolution of seroprevalence and does not jeopardize the global comparability of serological results. This is confirmed by the fact that our HI assay detected seroprotective antibody titers in 93.5% and gave evidence seroconversion in 73.9% of qRT-PCR confirmed pH1N1/2009 influenza, all figures close to those reported in the literature [5, 21] .
We considered that titers of .1/40, in sera collected from individuals enrolled during weeks 30 and 31 were cross reactive antibodies and not de novo antibodies triggered by the pandemic virus and hence used them as a proxy for baseline pre epidemic immunity. Several arguments support this assumption: i) the first case indicating autochthonous transmission in Reunion Island was reported by the epidemiological surveillance department of La Réunion on 21st July (week 30), i.e. the same day when inclusion started in our study cohort; ii) 7 to 15 days are required to develop an antibody response after viral infection; iii) On weeks 30 and 31, the epidemic activity due to the pandemic virus was very low in our study cohort and it became significant only after week 32. Hence, during weeks 30-31, 103 households were recruited and only 2 households reported ILI cases. Nasal swabs collected from these 2 individuals were tested qRT-PCR negative to the pandemic virus whereas one had evidence of coronavirus and rhinovirus using a multiplex RT-PCR to respiratory viruses (H. Pascalis, manuscript in preparation). In contrast, during weeks 32 to 39, 199 individuals belonging to 99 households reported ILI, among whom 60 individuals had documented infection by the pandemic virus.
Our study shows that a substantial proportion of Reunion Island's population had pre-existing immunity to 2009 pandemic influenza virus with the highest baseline-proxy seroprevalence rate observed among adults aged of 60 years or more. Other studies from all continents had also reported high pre-epidemic seropositivity rates among the elderly [5, 6, 8, [22] [23] [24] [25] [26] , though large variations do exist between countries [10, 11, 23, 27, 28] . These cross reactive antibodies have been interpreted as being the residual signature of the remote exposure of these individuals to H1N1 viruses circulating before 1957 [24, 25, 29, 30] . Baseline seropositivity rates that we report in children and in younger adults (i.e. 30%-35%) were notably higher than those reported from other parts of the world [6, 8, 22, 23, [31] [32] [33] . However one should note that these baseline antibodies were of low titer, just at the level of the HIA threshold (i.e. 1/40). Several factors could have contributed to this comparatively high baseline rates found in our study: i) It may reflect the fact that the HI test used in our study was marginally more sensitive than the classic one [17] ; ii) Some individuals may have already been infected with pH1N1/ 2009 virus at weeks 30 and 31 and may have triggered an antibody response to the virus. This hypothesis seems unlikely in view of the arguments presented above and of a similar high proportion of sera titering HIA = 1/40 among 122 sera from adult patients sent for diagnostic purposes to the Regional Hospital microbiology laboratory, during the first half of 2009 (i.e. before the 2009 pandemic) (data not shown). However we cannot formally exclude this hypothesis in view of a recently reported study from Taiwan [11] that showed evidence of subclinical community transmission with proved seroconversion several weeks before report of the first documented case in the island. A similar conclusion was also drawn from Australia [34] ; iii) our serological test might detect cross-reactive antibodies triggered by recent vaccination with trivalent seasonal influenza vaccine as reported [4, [35] [36] [37] [38] [39] . However, seasonal influenza vaccines were of rather limited use in Reunion Island, especially in children and young adults; iv) Finally the high baseline titers may reflect the infectious history of the individuals to seasonal influenza viruses cross antigenic with pH1N1/2009 virus as recently suggested for seasonal 2007 H1N1 infection [40] . This serosurvey indicates that a large fraction of the Reunion Island population was infected with the pandemic virus. Younger people, have paid the main tribute to the epidemic as almost two thirds show evidence of seroconversion, confirming earlier clinical reports from the island [12] and accumulating reports from other countries [17, 32, 41, 42] and suggesting that school children have likely played the central role in the epidemic diffusion of the pandemic virus. Lower infection rates were found in adults and the lowest rates were recorded in the elderly.
Based on clinical cases reported to the epidemiological surveillance services [12] , it was estimated that 66,915 persons in Reunion Island who consulted a physician were infected by the pH1N1/2009 virus during the 9 weeks of the epidemic, giving a cumulative attack rate of 8.26%. Taking into account those who did not consult a physician, the number of symptomatic infected persons was estimated to 104,067 (attack rate: 12.85%). In fact, the attack rate of pH1N1/2009 infection in our serosurvey was about 42%-44% at the peak of the antibody response (i.e., weeks 40-44), a figure which is at least 3 to 4 times higher than rates of infection based on clinical cases The wide gap between the two estimates indicates that a large fraction (almost two thirds) of those who got infected by pH1N1/2009 virus escaped medical detection, probably because they developed mild disease or asymptomatic infection, a further indication of the benign nature of the virus, at least at the community level. In England, Baguelin et al. [43] estimated that the cumulative incidence rates of infection by the pandemic virus in children were 20 to 40 times higher than that estimated from clinical surveillance.
Our study, as others [6] , indicates that pre-existing cross reactive antibodies to pH1N1/2009 at titers $1/40 prevented from seroconversion in response to the pandemic virus. This level of pre-existing cross reactive immunity likely confers true protection against infection as about two thirds and one third of documented infection (qRT-PCR positive) in our series have occurred in individuals with baseline HIA titers ,1/40 and = 1/ 40 respectively and less than 5% of documented infections occurred in individuals with base line titers .1/40. The protection was effective not only in older adults but also in younger persons. This indicates that protection was conferred not only by baseline cross reactive antibodies triggered by close pH1N1/2009 viruses that circulated before 1957 (as in the elderly), but also by antibodies likely resulting from recent exposure to seasonal influenza epidemics (as shown in younger persons) [40] . The observed seroconversion rates depend on age, after adjusting for baseline pH1N1/2009 titers. The protective role of increasing age might be explained by a stronger cross-immunity in adults and elderly or by a higher exposure of young subjects to the virus during the 2009 epidemic (due to social contacts and mixing patterns). It may also indicate that immune mechanisms other than cross reactive antibodies detected by HIA (i.e. immunity to neuraminidase and conserved T cells epitopes [44] might develop throughout life, providing additional protection from infection or severe disease, especially in the elderly. Interestingly, evidence is seen for a decline in antibody titers, which occurred soon after the passage of the epidemic wave. In paired sera, this decline was significant enough to bring, within a few weeks, almost 27% of sera that tested positive (i.e. HI titers $1/40) in the immediate post epidemic phase to levels under the cut-off value in the second serum sample. This decay accounts for the observation that older adults ($60 yrs) in the study cohort were apparently almost completely spared by the epidemic if one only considers cumulative incidence rates derived from IHA titration on samples 2 (weeks 45-52). In fact, the cumulative incidence rate in older adults measured just after the epidemic peak (i.e. weeks 40-44) was 20.4%. Similar results of early antibody decay were recently reported [10, 45] . More generally, these data show that serosurveys conducted months after passage of the epidemic, likely underestimate the real extent of pH1N1/2009 infection, compared to antibody titration performed earlier, when humoral responses are at their highest level. Whether the decline in antibody titers has functional immunologic consequence to individuals or within the communities warrants further investigation. However, one should note that there was no second epidemic wave in Reunion Island during the subsequent austral winter seasons in 2010 and 2011.
Influenza during the 2010 winter was at a level not higher than the usual passages of seasonal flu, though almost two thirds of documented cases in 2010 were also due to pH1N1/2009v [46] . In addition many fewer pandemic virus isolates were noted during the ongoing 2011 austral winter, strongly suggesting that the first epidemic wave had conferred a solid herd immunity, at the community level.
Our study has some limitations. The fact that the epidemic progression coincided with the implementation of the prospective study, we were not able to collect, strictly speaking, pre-epidemic sera from the cohort members. Therefore we used as proxy base line seroprevalence data from individuals recruited at the very beginning of the investigation when the epidemic activity in the cohort was very low. This may overestimate the base line immunity if subclinical community transmission had occurred before the first cases of pH1N1/2009 influenza were reported. Antibodies to the pandemic virus were detected by HIA, a test that has a good specificity but a rather low sensitivity [46] . Hence, the threshold of 1/40 may underestimate the number of infected individuals. However, rates of seroconversion, the serologic gold standard test based on paired sera, likely gave the most accurate picture of the pandemic in at the community level in Reunion Island.
|
How long did the pH1N1/2009 viral outbreak last?
| 5,261 |
9 weeks
| 28,239 |
1,602 |
High Burden of Non-Influenza Viruses in Influenza-Like Illness in the Early Weeks of H1N1v Epidemic in France
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3157400/
SHA: f4c1afe385e9e31eb5678e15a3c280ba97326554
Authors: Schnepf, Nathalie; Resche-Rigon, Matthieu; Chaillon, Antoine; Scemla, Anne; Gras, Guillaume; Semoun, Oren; Taboulet, Pierre; Molina, Jean-Michel; Simon, François; Goudeau, Alain; LeGoff, Jérôme
Date: 2011-08-17
DOI: 10.1371/journal.pone.0023514
License: cc-by
Abstract: BACKGROUND: Influenza-like illness (ILI) may be caused by a variety of pathogens. Clinical observations are of little help to recognise myxovirus infection and implement appropriate prevention measures. The limited use of molecular tools underestimates the role of other common pathogens. OBJECTIVES: During the early weeks of the 2009–2010 flu pandemic, a clinical and virological survey was conducted in adult and paediatric patients with ILI referred to two French University hospitals in Paris and Tours. Aims were to investigate the different pathogens involved in ILI and describe the associated symptoms. METHODS: H1N1v pandemic influenza diagnosis was performed with real time RT-PCR assay. Other viral aetiologies were investigated by the molecular multiplex assay RespiFinder19®. Clinical data were collected prospectively by physicians using a standard questionnaire. RESULTS: From week 35 to 44, endonasal swabs were collected in 413 patients. Overall, 68 samples (16.5%) were positive for H1N1v. In 13 of them, other respiratory pathogens were also detected. Among H1N1v negative samples, 213 (61.9%) were positive for various respiratory agents, 190 in single infections and 23 in mixed infections. The most prevalent viruses in H1N1v negative single infections were rhinovirus (62.6%), followed by parainfluenza viruses (24.2%) and adenovirus (5.3%). 70.6% of H1N1v cases were identified in patients under 40 years and none after 65 years. There was no difference between clinical symptoms observed in patients infected with H1N1v or with other pathogens. CONCLUSION: Our results highlight the high frequency of non-influenza viruses involved in ILI during the pre-epidemic period of a flu alert and the lack of specific clinical signs associated with influenza infections. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
Text: In order to monitor the spread of influenza and alert health handlers, several epidemiological tools have been developed. In France, a network of 1300 general practitioners, ''Réseau Sentinelles'', working throughout the country, provides real-time clinical data used to evaluate regional and national influenza spreading [1, 2] . The criteria used by this network to define clinical influenza-like illness (ILI) are the occurrence of a sudden fever above 39uC with myalgia and respiratory signs. In general no formal viral diagnosis is carried out. The Groupes Régionaux d'Observation de la Grippe (GROG) is a second French network that surveys the emergence and the spread of the influenza viruses [3, 4] . This network is based on clinical surveillance of acute respiratory infections and laboratory analysis of nasal specimens collected from adults and children by volunteer general practitioners and pediatricians.
According to the sentinel network's criteria, French health authorities proclaimed that flu epidemic level was reached during the second week of September 2009 (week 37) [5, 6] . On the contrary, data provided by the GROG showed only sporadic H1N1v activity until the last week of October (week 44) [6, 7] . Thus, it became rapidly obvious that a variety of viruses were circulating in the community and that an overestimation of myxovirus infection was at stake [8, 9, 10, 11] .
As a better knowledge of the epidemic status was a key feature for national healthcare organization, hospital preparedness, patient management and disease control, unambiguous viral diagnosis appeared critical. In France, data on viral aetiologies associated with ILI were at best sporadic and correlations with clinical symptoms were often lacking. Extensive molecular assays to screening for respiratory viruses were not available countrywide for routine diagnosis. Therefore the epidemiological pattern of respiratory pathogens with overlapping seasonality was poorly known.
The aim of the present study was to investigate respiratory pathogens involved in ILI during the early weeks of the 2009-2010 H1N1v diffusion in France (weeks 35 through 44) and describe the associated symptoms in paediatric and adult populations.
This study was a non-interventional study with no addition to usual proceedures. Biological material and clinical data were obtained only for standard viral diagnostic following physicians' prescriptions (no specific sampling, no modification of the sampling protocol, no supplementary question in the national standardized questionnaire). Data analyses were carried out using an anonymized database. According to the French Health Public Law (CSP Art L 1121-1.1), such protocol does not require approval of an ethics committee and is exempted from informed consent application.
In the two academic hospitals, Saint-Louis hospital (SLS) in Paris and Tours hospital (TRS), influenza-like illness (ILI) was defined as a patient suffering from at least one general symptom (fever above 38uC, asthenia, myalgia, shivers or headache) and one respiratory symptom (cough, dyspnoea, rhinitis or pharyngitis), in agreement with the guidelines from the French Institut de Veille Sanitaire (InVS), a governmental institution responsible for surveillance and alert in all domains of public health [12] . Criteria for severe clinical presentation were temperature below 35uC or above 39uC despite antipyretic, cardiac frequency above 120/min, respiratory frequency above 30/min, respiratory distress, systolic arterial pressure below 90 mmHg or altered consciousness. Predisposing factors of critical illness were children younger than one year old, pregnant women, diabetes, chronic pre-existing disease (such as respiratory, cardiovascular, neurologic, renal, hepatic or hematologic diseases) and immunosuppression (associated with HIV infection, organ or hematopoietic stem cells transplantation, receipt of chemotherapy or corticosteroids) [13, 14] . A cluster of suspected influenza infections was defined as at least three possible cases in a week in a closed community (household, school,…) [15] .
In the two institutions, the prescription of H1N1v molecular testing was recommended for patients with ILI and with either a severe clinical presentation, an underlying risk factor of complications or a condition which was not improving under antiviral treatment. Investigation of grouped suspected cases was also recommended. From week 35 (last week of August) to 44 (last week of October), 413 endonasal swabs were collected in 3 ml of Universal Transport Medium (Copan Diagnostics Inc, Murrieta, CA) from adults and children seen in emergency rooms for suspected ILI (Table 1 ) and sent to SLS and TRS laboratories for H1N1v detection. The two microbiology laboratories participated in the reference laboratories network for the detection of pandemic influenza H1N1v.
Clinical data were collected at the time of medical attention and reported by clinicians on a national standardized questionnaire provided by InVS [1, 12] . This questionnaire included the presence or absence of the main general and respiratory symptoms associated with ILI (fever, asthenia, myalgia, shivers, headache, cough, rhinitis, pharyngitis, sudden onset) [12] .
Total nucleic acid was extracted from 400 mL of Universal Transport Medium using the EasyMag System (Biomérieux, Marcy l'Etoile, France) in SLS or the EZ1 Advanced XL (Qiagen, Courtaboeuf, France) in TRS, according to the manufacturers' instructions (elution volume: 100 mL in SLS or 90 mL in TRS). Before extraction, 5 ml of an Internal Amplification Control (IAC) which contained an encephalomyocarditis virus (EMC) RNA transcript was added into the sample.
Pandemic H1N1v infection was diagnosed by real-time reverse transcription-PCR (RT-PCR) assay on a 7500 Real Time PCR System (Applied Biosystems, Foster City, CA) according to the protocol of the Centers for Disease Control (CDC) [16] . Other respiratory infections were investigated by a multiplex molecular assay based on the Multiplex Ligation-dependent Probe-Amplification (MLPA) technology (RespiFinder19H, Pathofinder, Maastricht, The Netherlands) that allows the detection and differentiation of 14 respiratory viruses, including influenza virus A (InfA), influenza virus B (InfB), rhinovirus (RHV), parainfluenza viruses 1 to 4 (PIV-1 to PIV-4), human metapneumovirus (hMPV), adenovirus (ADV), respiratory syncytial virus A (RSVA), respiratory syncytial virus B (RSVB) and human coronaviruses 229E, OC43 and NL63 (Cor-229E, Cor-OC43, Cor-NL63) [17] . The test allows also the detection of H5N1 influenza A virus and of four bacteria: Chlamydophila pneumoniae (CP), Mycoplasma pneumoniae (MP), Legionella pneumophila (LP) and Bordetella pertussis (BP). The amplified MLPA products were analyzed on an ABI 3100 genetic analyzer (Applied Biosystems, Foster City, CA). Fragment sizing analysis was performed with the GeneMarker software (SoftGenetics, LLC, State College, PA).
Further testing for H1N1v was carried out with Simplexa TM Influenza A H1N1 (2009) (Focus Diagnostics, Cypress, California) when the CDC real time RT-PCR assay was negative for H1N1 and the RespiFinder19H assay was positive for Influenza A. If this latter assay was negative, H3N2 typing was performed as previously described [18] .
Data from our study are summarized as frequencies and percentages for categorical variables. Quantitative variables are presented as medians, 25th and 75th percentiles. To compare those variables according to the viral infection status, Fisher tests
By using CDC reference assay, H1N1v was detected in 66 samples out of 413 (16.6%), more frequently in SLS (38 samples) than in TRS (28 samples) (p,10 24 ). Overall, weekly percentage of H1N1v positive endonasal swabs remained under 10% until week 41 and increase significantly after (P Trend ,0.0001) ( Figure 1 ). Rate of H1N1v detection reached 30% in SLS at week 42 and in TRS at week 44. Overall, this rate was in agreement with results provided by the GROG network, showing an earlier start of H1N1v epidemic in Paris area [7, 19] .
All 413 nucleic acid extracts were analyzed using the RespiFinder19H assay ( Figure 2 ). Sixty six patients tested H1N1v positive with CDC real time RT-PCR assay were confirmed with the multiplex assay. Thirteen were also co-infected by one or two other respiratory pathogens (multiple infections) ( Figure 2 ). Three of the 347 H1N1v negative samples could not be studied with the multiplex assay because they contained RT-PCR inhibitors (no amplification of the internal control). Two hundred and fifteen (62.5%) of the remaining 344 H1N1v negative samples were found positive for at least one respiratory pathogen ( Figure 2 ). Two hundred and twelve were positive for non influenza pathogens (189 single infections and 23 mixed infections with two, three or four viruses) and three additional single infections by influenza A were identified in SLS, including two by pandemic H1N1v and one by seasonal H3N2, as determined after molecular typing (data not shown).
Overall, 68 patients (16.5%) were then positive for H1N1v, one for H3N2 and 212 for non influenza pathogens. There were 245 single infections (55 with H1N1v and 190 with other respiratory pathogens) and 36 mixed infections (13 with H1N1v and 23 without H1N1v) ( Figure 2 ).
Among H1N1v negative single infections, the most prevalent viruses were rhinovirus (62.6%, 119 patients), followed by parainfluenza viruses 1 to 4 (24.2%, 46 patients), adenovirus (5.3%, 10 patients), human coronavirus 229E, OC43 and NL63 (3.2%, 6 patients) and respiratory syncytial virus A and B (2.6%, 5 patients) (Figure 2 ). In addition, RespiFinder19H assay identified three patients with bacterial infection, two with Mycoplasma pneumoniae (one 25 years old female in SLS and one 39 years old female in TRS) and one with Bordetella pertussis (one 60 years old male in SLS). No single infection by influenza B, hMPV, Chlamydophila pneumoniae or Legionella pneumophila was identified ( Figure 2 To analyze if viral co-infections occurred more frequently for some viruses, we carried out a two by two comparisons, that showed a higher proportion of co-infection only for ADV (p = 0.05).
Non-influenza respiratory viruses presented a different epidemic profile compared to H1N1v. Overall, in both hospitals, weekly rate of non-H1N1v respiratory viruses whether alone or involved in co-infection increased between week 37 and 39 (from 51.4% to 81.3%) and then consistently decreased ( Figure 3 ). RHV infections that represented nearly half of non-H1N1v viral infections (141 out of 213, 66.2%) were a significant contributing factor. In both hospitals, emergence of H1N1v cases was associated with a rapid decline of RHV rate of infection from 50-60% down to less than 20% with a one to two weeks gap between SLS and TRS.
Data on age ( In both institutions, 85.5% (106/124) children younger than 15 years of age were infected by at least one respiratory pathogen ( Table 2 ). H1N1v infected patients were not significantly younger than H1N1v non infected patients (27 years old vs. 25 years old, p = 0.80) (Figure 4) . However, 70.6% (48/68) of H1N1v cases were identified in patients under 40 years old (22 in SLS and 26 in TRS) and no case was observed in patients older than 65 years ( Table 2) . PIV infection occurred in very young patients (median (Figure 4) . Consequently, PIV and ADV were more frequently detected in the younger population of TRS versus SLS (p,10 24 and p,10 23 respectively). In contrast, although individuals with RHV infection were slightly younger than individuals without (median age = 24 vs. 29 for patients without RHV, p = 0.05) (Figure 4) , influenza-like illness associated with RHV was more frequent in SLS than in TRS (p = 0.012). Finally, patients with viral multiple infection were significantly younger than those with single infection (median, IDR: 4, 2-18.5 vs. 25, 6-43) and rates of mixed infection At the time of medical attention, 383 (92.7%) standardized clinical questionnaires were collected out of 413 patients. Four of them could not be exploited because they were too incomplete. A review of the 379 workable questionnaires showed that 90.8% (344/379) of the patients included in this study fulfilled the criteria of ILI as defined above, and 52.5% had either a severe clinical presentation or an underlying risk factor of complications (45.9%, 174/379), or were in a suspected cluster of grouped cases (6.6%, 25/379).
Overall, most patients have fever (93.9%) and cough (86.1%) ( Table 3) . Other classical clinical signs associated with ILI such as asthenia, myalgia, shivers, headache, rhinitis or pharyngitis were less frequent. A sudden onset was also described in 59.2% of cases. Only 32.5% of the patients had a temperature above 39uC; the age of these patients ranged from zero to 86 years, with a median age of 32 years and a mean age of 34 years (data not shown).
In H1N1v infected patients (including single and multiple infections), the main symptoms were also fever (98.2%) and cough (89.5%) ( We then compared clinical characteristics between patients positive for H1N1v, patients positive for other respiratory pathogens and negative for H1N1v and patients without any detection of respiratory pathogens (as detected with RespiFin-der19H) ( Table 3 ). There was no difference between the three groups except for fever, cough, pharyngitis. However for these latter symptoms, the comparison between patients positive for H1N1v and those positive for other respiratory pathogens or between patients positive for H1N1v and those without any detection of respiratory pathogens, showed no difference except for pharyngitis, which was less frequent in patients positive for H1N1v than in patients positive for other respiratory pathogens ( Table 3) .
As RHV was the most frequent aetiology in ILI, we also compared clinical symptoms observed in patients with a single infection by RHV or by H1N1v (data not shown). There was no difference except that rhinitis and pharyngitis were significantly more frequent in RHV infection (62.7% vs. 34.1% [p = 0.006] and 39.0% vs. 10.0% [p = 0.001], respectively). Viral multiple infection (including samples with H1N1v) was not associated with a different clinical presentation. Fever and cough were observed in over 90% of the patients (90.6% and 90.3%, respectively), but only 33.3% of these patients had a temperature above 39uC, which was not different from patients with single viral infection (28.6%).
Our results highlight the high frequency of non-influenza viruses involved in acute respiratory infections during the epidemic period of a flu alert as defined by the Réseau Sentinelles according to ILI definition (a sudden fever above 39uC accompanied by myalgia and respiratory signs). These data extent previous observations in Europe reporting high prevalence of RHV infections before seasonal influenza [4, 20] or in 2009, before H1N1v pandemic influenza [1, 8, 9, 11, 21] . We confirm that RHV represent the most frequent aetiology of acute respiratory Table 2 . Age of patients with respiratory samples positive for H1N1v, positive for other respiratory pathogens or negative. infections both in adult and paediatric populations and may represent more than 50% of cases. We show that other viral infections than influenza and RHV may represent up to 30% of aetiologies. We observed differences between the two hospitals, with a higher frequency of parainfluenza and ADV infections in Tours in contrast with a higher frequency of RHV in Paris, likely explained by the higher proportion of paediatric samples collected in Tours. However, despite the distance between the two institutions (about 250 km) and differences between the two populations, both presented similar patterns of high frequency of non-influenza viruses in acute respiratory infections before the flu epidemic wave and a decline when influenza reached epidemic levels.
In the two cities, high frequencies of RHV were seen at the same level with a likely different evolution speed, with sudden increase and decrease in SLS and more progressive variation in TRS. In both institutions, there was a decrease in the proportion and number of RHV diagnoses roughly in parallel with the increase of influenza diagnoses. Indeed, H1N1v exceeds 20% of positive detection's rate only when RHV dropped under 40%. These data are thus consistent with negative interaction of the two epidemics at the population level. It was previously hypothesised that RHV epidemic could interfere with the spread of pandemic influenza [20, 21, 22] . Few in vitro data support this hypothesis. It has been reported that interferon and other cytokines production by RHV infected cells induced a refractory state to virus infection These data include the three patients whose respiratory samples could not be studied with the multiplex assay because of RT-PCR inhibitors. of neighbouring cells [23] . Further work is needed to confirm in vitro and in vivo such negative interactions and if viral interference are really translated to a population level. Analysis of rhinovirus and influenza epidemics in previous years should also help to determine if similar interferences were observed with seasonal influenza and to elaborate modelling and prediction of the spread of influenza according to respiratory viruses' circulation. Systematic extensive screening of respiratory viruses at a national level should be implemented for this purpose.
Very few RSV infections were observed in contrast to usual epidemiology which was characterized the last four past years by a start of epidemics in weeks 44-45 [1] . It has been confirmed by other laboratories and the French InVS that the 2009-10 RSV epidemic was delayed and had a lower impact compared with the previous winter season [1, 24] . Delayed and reduced RSV spread may be due to viral interference between RSV and influenza. Another possible explanation is better prevention behaviour about respiratory infections as recommended by a national campaign including recommendations for hands washing after sneezing and the use of mask [1] .
Influenza infections were mainly detected in patient under 40 years old and no case was found in patients older than 65. These results corroborate previous data suggesting that past seasonal H1N1 infections or vaccination may give partial crossed protection [10, 13, 25] . We have previously shown that the neutralizing titers against pandemic H1N1v virus correlate significantly with neutralizing titers against a seasonal H1N1 virus, and that the H1N1v pandemic influenza virus neutralizing titer was significantly higher in subjects who had recently been inoculated by a seasonal trivalent influenza vaccine [26] .
Viral co-infections were predominantly seen in paediatric patients, as previously described [4, 27, 28, 29] , both in influenza and non-influenza cases at a similar rate. No evidence of more pronounced respiratory impact was seen in these patients.
Our results showed the lack of specific clinical signs associated with proven H1N1v infections. Clinical characteristics did not differ between influenza infections or other viral infections. In particular, the proportion of patients with fever above 39uC was not higher in H1N1v positive patients. In addition, the patients without any evidence of respiratory viral infections did not have different symptoms. These patients may have been infected with other virus not included in the multiplex assay (human Bocavirus, coronavirus HKU1) [9, 10, 11] or were seen too late at the time of viral shedding was cleared [30] . However, to determine how specific the symptoms are for influenza would require to assess also the distribution of respiratory pathogens (H1N1v and other respiratory viruses) and related symptoms in patients presented at the emergency departments in SLS and TRS with respiratory syndromes, but not tested for H1N1v. In addition, despite some underlying conditions that were associated with complications not previously observed in seasonal influenza, most illnesses caused by the H1N1v virus were acute and self-limited [13, 31] . The higher proportion of non influenza viruses reported in ILI in 2009 was thus most likely a consequence of more frequent visits to a doctor for respiratory tract infections than usually observed for fear of the flu pandemic. The general lack of difference in symptoms in the particular context of H1N1v pandemic has therefore to be considered with caution and does not rule out that more significant differences may arise in future influenza epidemics with other influenza viruses. Our data confirm that it may be virtually impossible to recognize symptoms heralding H1N1v infections and virological data should be helpful along with clinical reports to monitor influenza epidemic [10] .
Molecular multiplex detection has recently emerged as a potent diagnostic tool to determine acute respiratory infections' aetiologies [11, 32, 33] . These data show that sensitive molecular multiplex detection of respiratory viruses is feasible and efficient for the detection of virus involved in acute respiratory infections and provides insights into their epidemic profile. Our results confirm the performance of RespiFinder19H assay to detecting respiratory viruses in the general population as recently shown in transplant patients with ILI [34] . RespiFinder19H confirmed all H1N1 infections detected by the CDC reference assay and was able to identify two additional H1N1 cases suggesting a high sensitivity of this multiplex assay to detect influenza A infections.
In conclusion, our results highlight that successive and mixed outbreaks of respiratory viral infections may affect influenza epidemiology and can lead to misinterpret the early development of a flu epidemic. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
|
What were the aims of this study?
| 5,262 |
to investigate the different pathogens involved in ILI and describe the associated symptoms
| 1,012 |
1,602 |
High Burden of Non-Influenza Viruses in Influenza-Like Illness in the Early Weeks of H1N1v Epidemic in France
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3157400/
SHA: f4c1afe385e9e31eb5678e15a3c280ba97326554
Authors: Schnepf, Nathalie; Resche-Rigon, Matthieu; Chaillon, Antoine; Scemla, Anne; Gras, Guillaume; Semoun, Oren; Taboulet, Pierre; Molina, Jean-Michel; Simon, François; Goudeau, Alain; LeGoff, Jérôme
Date: 2011-08-17
DOI: 10.1371/journal.pone.0023514
License: cc-by
Abstract: BACKGROUND: Influenza-like illness (ILI) may be caused by a variety of pathogens. Clinical observations are of little help to recognise myxovirus infection and implement appropriate prevention measures. The limited use of molecular tools underestimates the role of other common pathogens. OBJECTIVES: During the early weeks of the 2009–2010 flu pandemic, a clinical and virological survey was conducted in adult and paediatric patients with ILI referred to two French University hospitals in Paris and Tours. Aims were to investigate the different pathogens involved in ILI and describe the associated symptoms. METHODS: H1N1v pandemic influenza diagnosis was performed with real time RT-PCR assay. Other viral aetiologies were investigated by the molecular multiplex assay RespiFinder19®. Clinical data were collected prospectively by physicians using a standard questionnaire. RESULTS: From week 35 to 44, endonasal swabs were collected in 413 patients. Overall, 68 samples (16.5%) were positive for H1N1v. In 13 of them, other respiratory pathogens were also detected. Among H1N1v negative samples, 213 (61.9%) were positive for various respiratory agents, 190 in single infections and 23 in mixed infections. The most prevalent viruses in H1N1v negative single infections were rhinovirus (62.6%), followed by parainfluenza viruses (24.2%) and adenovirus (5.3%). 70.6% of H1N1v cases were identified in patients under 40 years and none after 65 years. There was no difference between clinical symptoms observed in patients infected with H1N1v or with other pathogens. CONCLUSION: Our results highlight the high frequency of non-influenza viruses involved in ILI during the pre-epidemic period of a flu alert and the lack of specific clinical signs associated with influenza infections. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
Text: In order to monitor the spread of influenza and alert health handlers, several epidemiological tools have been developed. In France, a network of 1300 general practitioners, ''Réseau Sentinelles'', working throughout the country, provides real-time clinical data used to evaluate regional and national influenza spreading [1, 2] . The criteria used by this network to define clinical influenza-like illness (ILI) are the occurrence of a sudden fever above 39uC with myalgia and respiratory signs. In general no formal viral diagnosis is carried out. The Groupes Régionaux d'Observation de la Grippe (GROG) is a second French network that surveys the emergence and the spread of the influenza viruses [3, 4] . This network is based on clinical surveillance of acute respiratory infections and laboratory analysis of nasal specimens collected from adults and children by volunteer general practitioners and pediatricians.
According to the sentinel network's criteria, French health authorities proclaimed that flu epidemic level was reached during the second week of September 2009 (week 37) [5, 6] . On the contrary, data provided by the GROG showed only sporadic H1N1v activity until the last week of October (week 44) [6, 7] . Thus, it became rapidly obvious that a variety of viruses were circulating in the community and that an overestimation of myxovirus infection was at stake [8, 9, 10, 11] .
As a better knowledge of the epidemic status was a key feature for national healthcare organization, hospital preparedness, patient management and disease control, unambiguous viral diagnosis appeared critical. In France, data on viral aetiologies associated with ILI were at best sporadic and correlations with clinical symptoms were often lacking. Extensive molecular assays to screening for respiratory viruses were not available countrywide for routine diagnosis. Therefore the epidemiological pattern of respiratory pathogens with overlapping seasonality was poorly known.
The aim of the present study was to investigate respiratory pathogens involved in ILI during the early weeks of the 2009-2010 H1N1v diffusion in France (weeks 35 through 44) and describe the associated symptoms in paediatric and adult populations.
This study was a non-interventional study with no addition to usual proceedures. Biological material and clinical data were obtained only for standard viral diagnostic following physicians' prescriptions (no specific sampling, no modification of the sampling protocol, no supplementary question in the national standardized questionnaire). Data analyses were carried out using an anonymized database. According to the French Health Public Law (CSP Art L 1121-1.1), such protocol does not require approval of an ethics committee and is exempted from informed consent application.
In the two academic hospitals, Saint-Louis hospital (SLS) in Paris and Tours hospital (TRS), influenza-like illness (ILI) was defined as a patient suffering from at least one general symptom (fever above 38uC, asthenia, myalgia, shivers or headache) and one respiratory symptom (cough, dyspnoea, rhinitis or pharyngitis), in agreement with the guidelines from the French Institut de Veille Sanitaire (InVS), a governmental institution responsible for surveillance and alert in all domains of public health [12] . Criteria for severe clinical presentation were temperature below 35uC or above 39uC despite antipyretic, cardiac frequency above 120/min, respiratory frequency above 30/min, respiratory distress, systolic arterial pressure below 90 mmHg or altered consciousness. Predisposing factors of critical illness were children younger than one year old, pregnant women, diabetes, chronic pre-existing disease (such as respiratory, cardiovascular, neurologic, renal, hepatic or hematologic diseases) and immunosuppression (associated with HIV infection, organ or hematopoietic stem cells transplantation, receipt of chemotherapy or corticosteroids) [13, 14] . A cluster of suspected influenza infections was defined as at least three possible cases in a week in a closed community (household, school,…) [15] .
In the two institutions, the prescription of H1N1v molecular testing was recommended for patients with ILI and with either a severe clinical presentation, an underlying risk factor of complications or a condition which was not improving under antiviral treatment. Investigation of grouped suspected cases was also recommended. From week 35 (last week of August) to 44 (last week of October), 413 endonasal swabs were collected in 3 ml of Universal Transport Medium (Copan Diagnostics Inc, Murrieta, CA) from adults and children seen in emergency rooms for suspected ILI (Table 1 ) and sent to SLS and TRS laboratories for H1N1v detection. The two microbiology laboratories participated in the reference laboratories network for the detection of pandemic influenza H1N1v.
Clinical data were collected at the time of medical attention and reported by clinicians on a national standardized questionnaire provided by InVS [1, 12] . This questionnaire included the presence or absence of the main general and respiratory symptoms associated with ILI (fever, asthenia, myalgia, shivers, headache, cough, rhinitis, pharyngitis, sudden onset) [12] .
Total nucleic acid was extracted from 400 mL of Universal Transport Medium using the EasyMag System (Biomérieux, Marcy l'Etoile, France) in SLS or the EZ1 Advanced XL (Qiagen, Courtaboeuf, France) in TRS, according to the manufacturers' instructions (elution volume: 100 mL in SLS or 90 mL in TRS). Before extraction, 5 ml of an Internal Amplification Control (IAC) which contained an encephalomyocarditis virus (EMC) RNA transcript was added into the sample.
Pandemic H1N1v infection was diagnosed by real-time reverse transcription-PCR (RT-PCR) assay on a 7500 Real Time PCR System (Applied Biosystems, Foster City, CA) according to the protocol of the Centers for Disease Control (CDC) [16] . Other respiratory infections were investigated by a multiplex molecular assay based on the Multiplex Ligation-dependent Probe-Amplification (MLPA) technology (RespiFinder19H, Pathofinder, Maastricht, The Netherlands) that allows the detection and differentiation of 14 respiratory viruses, including influenza virus A (InfA), influenza virus B (InfB), rhinovirus (RHV), parainfluenza viruses 1 to 4 (PIV-1 to PIV-4), human metapneumovirus (hMPV), adenovirus (ADV), respiratory syncytial virus A (RSVA), respiratory syncytial virus B (RSVB) and human coronaviruses 229E, OC43 and NL63 (Cor-229E, Cor-OC43, Cor-NL63) [17] . The test allows also the detection of H5N1 influenza A virus and of four bacteria: Chlamydophila pneumoniae (CP), Mycoplasma pneumoniae (MP), Legionella pneumophila (LP) and Bordetella pertussis (BP). The amplified MLPA products were analyzed on an ABI 3100 genetic analyzer (Applied Biosystems, Foster City, CA). Fragment sizing analysis was performed with the GeneMarker software (SoftGenetics, LLC, State College, PA).
Further testing for H1N1v was carried out with Simplexa TM Influenza A H1N1 (2009) (Focus Diagnostics, Cypress, California) when the CDC real time RT-PCR assay was negative for H1N1 and the RespiFinder19H assay was positive for Influenza A. If this latter assay was negative, H3N2 typing was performed as previously described [18] .
Data from our study are summarized as frequencies and percentages for categorical variables. Quantitative variables are presented as medians, 25th and 75th percentiles. To compare those variables according to the viral infection status, Fisher tests
By using CDC reference assay, H1N1v was detected in 66 samples out of 413 (16.6%), more frequently in SLS (38 samples) than in TRS (28 samples) (p,10 24 ). Overall, weekly percentage of H1N1v positive endonasal swabs remained under 10% until week 41 and increase significantly after (P Trend ,0.0001) ( Figure 1 ). Rate of H1N1v detection reached 30% in SLS at week 42 and in TRS at week 44. Overall, this rate was in agreement with results provided by the GROG network, showing an earlier start of H1N1v epidemic in Paris area [7, 19] .
All 413 nucleic acid extracts were analyzed using the RespiFinder19H assay ( Figure 2 ). Sixty six patients tested H1N1v positive with CDC real time RT-PCR assay were confirmed with the multiplex assay. Thirteen were also co-infected by one or two other respiratory pathogens (multiple infections) ( Figure 2 ). Three of the 347 H1N1v negative samples could not be studied with the multiplex assay because they contained RT-PCR inhibitors (no amplification of the internal control). Two hundred and fifteen (62.5%) of the remaining 344 H1N1v negative samples were found positive for at least one respiratory pathogen ( Figure 2 ). Two hundred and twelve were positive for non influenza pathogens (189 single infections and 23 mixed infections with two, three or four viruses) and three additional single infections by influenza A were identified in SLS, including two by pandemic H1N1v and one by seasonal H3N2, as determined after molecular typing (data not shown).
Overall, 68 patients (16.5%) were then positive for H1N1v, one for H3N2 and 212 for non influenza pathogens. There were 245 single infections (55 with H1N1v and 190 with other respiratory pathogens) and 36 mixed infections (13 with H1N1v and 23 without H1N1v) ( Figure 2 ).
Among H1N1v negative single infections, the most prevalent viruses were rhinovirus (62.6%, 119 patients), followed by parainfluenza viruses 1 to 4 (24.2%, 46 patients), adenovirus (5.3%, 10 patients), human coronavirus 229E, OC43 and NL63 (3.2%, 6 patients) and respiratory syncytial virus A and B (2.6%, 5 patients) (Figure 2 ). In addition, RespiFinder19H assay identified three patients with bacterial infection, two with Mycoplasma pneumoniae (one 25 years old female in SLS and one 39 years old female in TRS) and one with Bordetella pertussis (one 60 years old male in SLS). No single infection by influenza B, hMPV, Chlamydophila pneumoniae or Legionella pneumophila was identified ( Figure 2 To analyze if viral co-infections occurred more frequently for some viruses, we carried out a two by two comparisons, that showed a higher proportion of co-infection only for ADV (p = 0.05).
Non-influenza respiratory viruses presented a different epidemic profile compared to H1N1v. Overall, in both hospitals, weekly rate of non-H1N1v respiratory viruses whether alone or involved in co-infection increased between week 37 and 39 (from 51.4% to 81.3%) and then consistently decreased ( Figure 3 ). RHV infections that represented nearly half of non-H1N1v viral infections (141 out of 213, 66.2%) were a significant contributing factor. In both hospitals, emergence of H1N1v cases was associated with a rapid decline of RHV rate of infection from 50-60% down to less than 20% with a one to two weeks gap between SLS and TRS.
Data on age ( In both institutions, 85.5% (106/124) children younger than 15 years of age were infected by at least one respiratory pathogen ( Table 2 ). H1N1v infected patients were not significantly younger than H1N1v non infected patients (27 years old vs. 25 years old, p = 0.80) (Figure 4) . However, 70.6% (48/68) of H1N1v cases were identified in patients under 40 years old (22 in SLS and 26 in TRS) and no case was observed in patients older than 65 years ( Table 2) . PIV infection occurred in very young patients (median (Figure 4) . Consequently, PIV and ADV were more frequently detected in the younger population of TRS versus SLS (p,10 24 and p,10 23 respectively). In contrast, although individuals with RHV infection were slightly younger than individuals without (median age = 24 vs. 29 for patients without RHV, p = 0.05) (Figure 4) , influenza-like illness associated with RHV was more frequent in SLS than in TRS (p = 0.012). Finally, patients with viral multiple infection were significantly younger than those with single infection (median, IDR: 4, 2-18.5 vs. 25, 6-43) and rates of mixed infection At the time of medical attention, 383 (92.7%) standardized clinical questionnaires were collected out of 413 patients. Four of them could not be exploited because they were too incomplete. A review of the 379 workable questionnaires showed that 90.8% (344/379) of the patients included in this study fulfilled the criteria of ILI as defined above, and 52.5% had either a severe clinical presentation or an underlying risk factor of complications (45.9%, 174/379), or were in a suspected cluster of grouped cases (6.6%, 25/379).
Overall, most patients have fever (93.9%) and cough (86.1%) ( Table 3) . Other classical clinical signs associated with ILI such as asthenia, myalgia, shivers, headache, rhinitis or pharyngitis were less frequent. A sudden onset was also described in 59.2% of cases. Only 32.5% of the patients had a temperature above 39uC; the age of these patients ranged from zero to 86 years, with a median age of 32 years and a mean age of 34 years (data not shown).
In H1N1v infected patients (including single and multiple infections), the main symptoms were also fever (98.2%) and cough (89.5%) ( We then compared clinical characteristics between patients positive for H1N1v, patients positive for other respiratory pathogens and negative for H1N1v and patients without any detection of respiratory pathogens (as detected with RespiFin-der19H) ( Table 3 ). There was no difference between the three groups except for fever, cough, pharyngitis. However for these latter symptoms, the comparison between patients positive for H1N1v and those positive for other respiratory pathogens or between patients positive for H1N1v and those without any detection of respiratory pathogens, showed no difference except for pharyngitis, which was less frequent in patients positive for H1N1v than in patients positive for other respiratory pathogens ( Table 3) .
As RHV was the most frequent aetiology in ILI, we also compared clinical symptoms observed in patients with a single infection by RHV or by H1N1v (data not shown). There was no difference except that rhinitis and pharyngitis were significantly more frequent in RHV infection (62.7% vs. 34.1% [p = 0.006] and 39.0% vs. 10.0% [p = 0.001], respectively). Viral multiple infection (including samples with H1N1v) was not associated with a different clinical presentation. Fever and cough were observed in over 90% of the patients (90.6% and 90.3%, respectively), but only 33.3% of these patients had a temperature above 39uC, which was not different from patients with single viral infection (28.6%).
Our results highlight the high frequency of non-influenza viruses involved in acute respiratory infections during the epidemic period of a flu alert as defined by the Réseau Sentinelles according to ILI definition (a sudden fever above 39uC accompanied by myalgia and respiratory signs). These data extent previous observations in Europe reporting high prevalence of RHV infections before seasonal influenza [4, 20] or in 2009, before H1N1v pandemic influenza [1, 8, 9, 11, 21] . We confirm that RHV represent the most frequent aetiology of acute respiratory Table 2 . Age of patients with respiratory samples positive for H1N1v, positive for other respiratory pathogens or negative. infections both in adult and paediatric populations and may represent more than 50% of cases. We show that other viral infections than influenza and RHV may represent up to 30% of aetiologies. We observed differences between the two hospitals, with a higher frequency of parainfluenza and ADV infections in Tours in contrast with a higher frequency of RHV in Paris, likely explained by the higher proportion of paediatric samples collected in Tours. However, despite the distance between the two institutions (about 250 km) and differences between the two populations, both presented similar patterns of high frequency of non-influenza viruses in acute respiratory infections before the flu epidemic wave and a decline when influenza reached epidemic levels.
In the two cities, high frequencies of RHV were seen at the same level with a likely different evolution speed, with sudden increase and decrease in SLS and more progressive variation in TRS. In both institutions, there was a decrease in the proportion and number of RHV diagnoses roughly in parallel with the increase of influenza diagnoses. Indeed, H1N1v exceeds 20% of positive detection's rate only when RHV dropped under 40%. These data are thus consistent with negative interaction of the two epidemics at the population level. It was previously hypothesised that RHV epidemic could interfere with the spread of pandemic influenza [20, 21, 22] . Few in vitro data support this hypothesis. It has been reported that interferon and other cytokines production by RHV infected cells induced a refractory state to virus infection These data include the three patients whose respiratory samples could not be studied with the multiplex assay because of RT-PCR inhibitors. of neighbouring cells [23] . Further work is needed to confirm in vitro and in vivo such negative interactions and if viral interference are really translated to a population level. Analysis of rhinovirus and influenza epidemics in previous years should also help to determine if similar interferences were observed with seasonal influenza and to elaborate modelling and prediction of the spread of influenza according to respiratory viruses' circulation. Systematic extensive screening of respiratory viruses at a national level should be implemented for this purpose.
Very few RSV infections were observed in contrast to usual epidemiology which was characterized the last four past years by a start of epidemics in weeks 44-45 [1] . It has been confirmed by other laboratories and the French InVS that the 2009-10 RSV epidemic was delayed and had a lower impact compared with the previous winter season [1, 24] . Delayed and reduced RSV spread may be due to viral interference between RSV and influenza. Another possible explanation is better prevention behaviour about respiratory infections as recommended by a national campaign including recommendations for hands washing after sneezing and the use of mask [1] .
Influenza infections were mainly detected in patient under 40 years old and no case was found in patients older than 65. These results corroborate previous data suggesting that past seasonal H1N1 infections or vaccination may give partial crossed protection [10, 13, 25] . We have previously shown that the neutralizing titers against pandemic H1N1v virus correlate significantly with neutralizing titers against a seasonal H1N1 virus, and that the H1N1v pandemic influenza virus neutralizing titer was significantly higher in subjects who had recently been inoculated by a seasonal trivalent influenza vaccine [26] .
Viral co-infections were predominantly seen in paediatric patients, as previously described [4, 27, 28, 29] , both in influenza and non-influenza cases at a similar rate. No evidence of more pronounced respiratory impact was seen in these patients.
Our results showed the lack of specific clinical signs associated with proven H1N1v infections. Clinical characteristics did not differ between influenza infections or other viral infections. In particular, the proportion of patients with fever above 39uC was not higher in H1N1v positive patients. In addition, the patients without any evidence of respiratory viral infections did not have different symptoms. These patients may have been infected with other virus not included in the multiplex assay (human Bocavirus, coronavirus HKU1) [9, 10, 11] or were seen too late at the time of viral shedding was cleared [30] . However, to determine how specific the symptoms are for influenza would require to assess also the distribution of respiratory pathogens (H1N1v and other respiratory viruses) and related symptoms in patients presented at the emergency departments in SLS and TRS with respiratory syndromes, but not tested for H1N1v. In addition, despite some underlying conditions that were associated with complications not previously observed in seasonal influenza, most illnesses caused by the H1N1v virus were acute and self-limited [13, 31] . The higher proportion of non influenza viruses reported in ILI in 2009 was thus most likely a consequence of more frequent visits to a doctor for respiratory tract infections than usually observed for fear of the flu pandemic. The general lack of difference in symptoms in the particular context of H1N1v pandemic has therefore to be considered with caution and does not rule out that more significant differences may arise in future influenza epidemics with other influenza viruses. Our data confirm that it may be virtually impossible to recognize symptoms heralding H1N1v infections and virological data should be helpful along with clinical reports to monitor influenza epidemic [10] .
Molecular multiplex detection has recently emerged as a potent diagnostic tool to determine acute respiratory infections' aetiologies [11, 32, 33] . These data show that sensitive molecular multiplex detection of respiratory viruses is feasible and efficient for the detection of virus involved in acute respiratory infections and provides insights into their epidemic profile. Our results confirm the performance of RespiFinder19H assay to detecting respiratory viruses in the general population as recently shown in transplant patients with ILI [34] . RespiFinder19H confirmed all H1N1 infections detected by the CDC reference assay and was able to identify two additional H1N1 cases suggesting a high sensitivity of this multiplex assay to detect influenza A infections.
In conclusion, our results highlight that successive and mixed outbreaks of respiratory viral infections may affect influenza epidemiology and can lead to misinterpret the early development of a flu epidemic. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
|
What network of physicians provides real-time clinical data on the spread of influenza in France?
| 5,263 |
Réseau Sentinelles
| 2,654 |
1,602 |
High Burden of Non-Influenza Viruses in Influenza-Like Illness in the Early Weeks of H1N1v Epidemic in France
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3157400/
SHA: f4c1afe385e9e31eb5678e15a3c280ba97326554
Authors: Schnepf, Nathalie; Resche-Rigon, Matthieu; Chaillon, Antoine; Scemla, Anne; Gras, Guillaume; Semoun, Oren; Taboulet, Pierre; Molina, Jean-Michel; Simon, François; Goudeau, Alain; LeGoff, Jérôme
Date: 2011-08-17
DOI: 10.1371/journal.pone.0023514
License: cc-by
Abstract: BACKGROUND: Influenza-like illness (ILI) may be caused by a variety of pathogens. Clinical observations are of little help to recognise myxovirus infection and implement appropriate prevention measures. The limited use of molecular tools underestimates the role of other common pathogens. OBJECTIVES: During the early weeks of the 2009–2010 flu pandemic, a clinical and virological survey was conducted in adult and paediatric patients with ILI referred to two French University hospitals in Paris and Tours. Aims were to investigate the different pathogens involved in ILI and describe the associated symptoms. METHODS: H1N1v pandemic influenza diagnosis was performed with real time RT-PCR assay. Other viral aetiologies were investigated by the molecular multiplex assay RespiFinder19®. Clinical data were collected prospectively by physicians using a standard questionnaire. RESULTS: From week 35 to 44, endonasal swabs were collected in 413 patients. Overall, 68 samples (16.5%) were positive for H1N1v. In 13 of them, other respiratory pathogens were also detected. Among H1N1v negative samples, 213 (61.9%) were positive for various respiratory agents, 190 in single infections and 23 in mixed infections. The most prevalent viruses in H1N1v negative single infections were rhinovirus (62.6%), followed by parainfluenza viruses (24.2%) and adenovirus (5.3%). 70.6% of H1N1v cases were identified in patients under 40 years and none after 65 years. There was no difference between clinical symptoms observed in patients infected with H1N1v or with other pathogens. CONCLUSION: Our results highlight the high frequency of non-influenza viruses involved in ILI during the pre-epidemic period of a flu alert and the lack of specific clinical signs associated with influenza infections. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
Text: In order to monitor the spread of influenza and alert health handlers, several epidemiological tools have been developed. In France, a network of 1300 general practitioners, ''Réseau Sentinelles'', working throughout the country, provides real-time clinical data used to evaluate regional and national influenza spreading [1, 2] . The criteria used by this network to define clinical influenza-like illness (ILI) are the occurrence of a sudden fever above 39uC with myalgia and respiratory signs. In general no formal viral diagnosis is carried out. The Groupes Régionaux d'Observation de la Grippe (GROG) is a second French network that surveys the emergence and the spread of the influenza viruses [3, 4] . This network is based on clinical surveillance of acute respiratory infections and laboratory analysis of nasal specimens collected from adults and children by volunteer general practitioners and pediatricians.
According to the sentinel network's criteria, French health authorities proclaimed that flu epidemic level was reached during the second week of September 2009 (week 37) [5, 6] . On the contrary, data provided by the GROG showed only sporadic H1N1v activity until the last week of October (week 44) [6, 7] . Thus, it became rapidly obvious that a variety of viruses were circulating in the community and that an overestimation of myxovirus infection was at stake [8, 9, 10, 11] .
As a better knowledge of the epidemic status was a key feature for national healthcare organization, hospital preparedness, patient management and disease control, unambiguous viral diagnosis appeared critical. In France, data on viral aetiologies associated with ILI were at best sporadic and correlations with clinical symptoms were often lacking. Extensive molecular assays to screening for respiratory viruses were not available countrywide for routine diagnosis. Therefore the epidemiological pattern of respiratory pathogens with overlapping seasonality was poorly known.
The aim of the present study was to investigate respiratory pathogens involved in ILI during the early weeks of the 2009-2010 H1N1v diffusion in France (weeks 35 through 44) and describe the associated symptoms in paediatric and adult populations.
This study was a non-interventional study with no addition to usual proceedures. Biological material and clinical data were obtained only for standard viral diagnostic following physicians' prescriptions (no specific sampling, no modification of the sampling protocol, no supplementary question in the national standardized questionnaire). Data analyses were carried out using an anonymized database. According to the French Health Public Law (CSP Art L 1121-1.1), such protocol does not require approval of an ethics committee and is exempted from informed consent application.
In the two academic hospitals, Saint-Louis hospital (SLS) in Paris and Tours hospital (TRS), influenza-like illness (ILI) was defined as a patient suffering from at least one general symptom (fever above 38uC, asthenia, myalgia, shivers or headache) and one respiratory symptom (cough, dyspnoea, rhinitis or pharyngitis), in agreement with the guidelines from the French Institut de Veille Sanitaire (InVS), a governmental institution responsible for surveillance and alert in all domains of public health [12] . Criteria for severe clinical presentation were temperature below 35uC or above 39uC despite antipyretic, cardiac frequency above 120/min, respiratory frequency above 30/min, respiratory distress, systolic arterial pressure below 90 mmHg or altered consciousness. Predisposing factors of critical illness were children younger than one year old, pregnant women, diabetes, chronic pre-existing disease (such as respiratory, cardiovascular, neurologic, renal, hepatic or hematologic diseases) and immunosuppression (associated with HIV infection, organ or hematopoietic stem cells transplantation, receipt of chemotherapy or corticosteroids) [13, 14] . A cluster of suspected influenza infections was defined as at least three possible cases in a week in a closed community (household, school,…) [15] .
In the two institutions, the prescription of H1N1v molecular testing was recommended for patients with ILI and with either a severe clinical presentation, an underlying risk factor of complications or a condition which was not improving under antiviral treatment. Investigation of grouped suspected cases was also recommended. From week 35 (last week of August) to 44 (last week of October), 413 endonasal swabs were collected in 3 ml of Universal Transport Medium (Copan Diagnostics Inc, Murrieta, CA) from adults and children seen in emergency rooms for suspected ILI (Table 1 ) and sent to SLS and TRS laboratories for H1N1v detection. The two microbiology laboratories participated in the reference laboratories network for the detection of pandemic influenza H1N1v.
Clinical data were collected at the time of medical attention and reported by clinicians on a national standardized questionnaire provided by InVS [1, 12] . This questionnaire included the presence or absence of the main general and respiratory symptoms associated with ILI (fever, asthenia, myalgia, shivers, headache, cough, rhinitis, pharyngitis, sudden onset) [12] .
Total nucleic acid was extracted from 400 mL of Universal Transport Medium using the EasyMag System (Biomérieux, Marcy l'Etoile, France) in SLS or the EZ1 Advanced XL (Qiagen, Courtaboeuf, France) in TRS, according to the manufacturers' instructions (elution volume: 100 mL in SLS or 90 mL in TRS). Before extraction, 5 ml of an Internal Amplification Control (IAC) which contained an encephalomyocarditis virus (EMC) RNA transcript was added into the sample.
Pandemic H1N1v infection was diagnosed by real-time reverse transcription-PCR (RT-PCR) assay on a 7500 Real Time PCR System (Applied Biosystems, Foster City, CA) according to the protocol of the Centers for Disease Control (CDC) [16] . Other respiratory infections were investigated by a multiplex molecular assay based on the Multiplex Ligation-dependent Probe-Amplification (MLPA) technology (RespiFinder19H, Pathofinder, Maastricht, The Netherlands) that allows the detection and differentiation of 14 respiratory viruses, including influenza virus A (InfA), influenza virus B (InfB), rhinovirus (RHV), parainfluenza viruses 1 to 4 (PIV-1 to PIV-4), human metapneumovirus (hMPV), adenovirus (ADV), respiratory syncytial virus A (RSVA), respiratory syncytial virus B (RSVB) and human coronaviruses 229E, OC43 and NL63 (Cor-229E, Cor-OC43, Cor-NL63) [17] . The test allows also the detection of H5N1 influenza A virus and of four bacteria: Chlamydophila pneumoniae (CP), Mycoplasma pneumoniae (MP), Legionella pneumophila (LP) and Bordetella pertussis (BP). The amplified MLPA products were analyzed on an ABI 3100 genetic analyzer (Applied Biosystems, Foster City, CA). Fragment sizing analysis was performed with the GeneMarker software (SoftGenetics, LLC, State College, PA).
Further testing for H1N1v was carried out with Simplexa TM Influenza A H1N1 (2009) (Focus Diagnostics, Cypress, California) when the CDC real time RT-PCR assay was negative for H1N1 and the RespiFinder19H assay was positive for Influenza A. If this latter assay was negative, H3N2 typing was performed as previously described [18] .
Data from our study are summarized as frequencies and percentages for categorical variables. Quantitative variables are presented as medians, 25th and 75th percentiles. To compare those variables according to the viral infection status, Fisher tests
By using CDC reference assay, H1N1v was detected in 66 samples out of 413 (16.6%), more frequently in SLS (38 samples) than in TRS (28 samples) (p,10 24 ). Overall, weekly percentage of H1N1v positive endonasal swabs remained under 10% until week 41 and increase significantly after (P Trend ,0.0001) ( Figure 1 ). Rate of H1N1v detection reached 30% in SLS at week 42 and in TRS at week 44. Overall, this rate was in agreement with results provided by the GROG network, showing an earlier start of H1N1v epidemic in Paris area [7, 19] .
All 413 nucleic acid extracts were analyzed using the RespiFinder19H assay ( Figure 2 ). Sixty six patients tested H1N1v positive with CDC real time RT-PCR assay were confirmed with the multiplex assay. Thirteen were also co-infected by one or two other respiratory pathogens (multiple infections) ( Figure 2 ). Three of the 347 H1N1v negative samples could not be studied with the multiplex assay because they contained RT-PCR inhibitors (no amplification of the internal control). Two hundred and fifteen (62.5%) of the remaining 344 H1N1v negative samples were found positive for at least one respiratory pathogen ( Figure 2 ). Two hundred and twelve were positive for non influenza pathogens (189 single infections and 23 mixed infections with two, three or four viruses) and three additional single infections by influenza A were identified in SLS, including two by pandemic H1N1v and one by seasonal H3N2, as determined after molecular typing (data not shown).
Overall, 68 patients (16.5%) were then positive for H1N1v, one for H3N2 and 212 for non influenza pathogens. There were 245 single infections (55 with H1N1v and 190 with other respiratory pathogens) and 36 mixed infections (13 with H1N1v and 23 without H1N1v) ( Figure 2 ).
Among H1N1v negative single infections, the most prevalent viruses were rhinovirus (62.6%, 119 patients), followed by parainfluenza viruses 1 to 4 (24.2%, 46 patients), adenovirus (5.3%, 10 patients), human coronavirus 229E, OC43 and NL63 (3.2%, 6 patients) and respiratory syncytial virus A and B (2.6%, 5 patients) (Figure 2 ). In addition, RespiFinder19H assay identified three patients with bacterial infection, two with Mycoplasma pneumoniae (one 25 years old female in SLS and one 39 years old female in TRS) and one with Bordetella pertussis (one 60 years old male in SLS). No single infection by influenza B, hMPV, Chlamydophila pneumoniae or Legionella pneumophila was identified ( Figure 2 To analyze if viral co-infections occurred more frequently for some viruses, we carried out a two by two comparisons, that showed a higher proportion of co-infection only for ADV (p = 0.05).
Non-influenza respiratory viruses presented a different epidemic profile compared to H1N1v. Overall, in both hospitals, weekly rate of non-H1N1v respiratory viruses whether alone or involved in co-infection increased between week 37 and 39 (from 51.4% to 81.3%) and then consistently decreased ( Figure 3 ). RHV infections that represented nearly half of non-H1N1v viral infections (141 out of 213, 66.2%) were a significant contributing factor. In both hospitals, emergence of H1N1v cases was associated with a rapid decline of RHV rate of infection from 50-60% down to less than 20% with a one to two weeks gap between SLS and TRS.
Data on age ( In both institutions, 85.5% (106/124) children younger than 15 years of age were infected by at least one respiratory pathogen ( Table 2 ). H1N1v infected patients were not significantly younger than H1N1v non infected patients (27 years old vs. 25 years old, p = 0.80) (Figure 4) . However, 70.6% (48/68) of H1N1v cases were identified in patients under 40 years old (22 in SLS and 26 in TRS) and no case was observed in patients older than 65 years ( Table 2) . PIV infection occurred in very young patients (median (Figure 4) . Consequently, PIV and ADV were more frequently detected in the younger population of TRS versus SLS (p,10 24 and p,10 23 respectively). In contrast, although individuals with RHV infection were slightly younger than individuals without (median age = 24 vs. 29 for patients without RHV, p = 0.05) (Figure 4) , influenza-like illness associated with RHV was more frequent in SLS than in TRS (p = 0.012). Finally, patients with viral multiple infection were significantly younger than those with single infection (median, IDR: 4, 2-18.5 vs. 25, 6-43) and rates of mixed infection At the time of medical attention, 383 (92.7%) standardized clinical questionnaires were collected out of 413 patients. Four of them could not be exploited because they were too incomplete. A review of the 379 workable questionnaires showed that 90.8% (344/379) of the patients included in this study fulfilled the criteria of ILI as defined above, and 52.5% had either a severe clinical presentation or an underlying risk factor of complications (45.9%, 174/379), or were in a suspected cluster of grouped cases (6.6%, 25/379).
Overall, most patients have fever (93.9%) and cough (86.1%) ( Table 3) . Other classical clinical signs associated with ILI such as asthenia, myalgia, shivers, headache, rhinitis or pharyngitis were less frequent. A sudden onset was also described in 59.2% of cases. Only 32.5% of the patients had a temperature above 39uC; the age of these patients ranged from zero to 86 years, with a median age of 32 years and a mean age of 34 years (data not shown).
In H1N1v infected patients (including single and multiple infections), the main symptoms were also fever (98.2%) and cough (89.5%) ( We then compared clinical characteristics between patients positive for H1N1v, patients positive for other respiratory pathogens and negative for H1N1v and patients without any detection of respiratory pathogens (as detected with RespiFin-der19H) ( Table 3 ). There was no difference between the three groups except for fever, cough, pharyngitis. However for these latter symptoms, the comparison between patients positive for H1N1v and those positive for other respiratory pathogens or between patients positive for H1N1v and those without any detection of respiratory pathogens, showed no difference except for pharyngitis, which was less frequent in patients positive for H1N1v than in patients positive for other respiratory pathogens ( Table 3) .
As RHV was the most frequent aetiology in ILI, we also compared clinical symptoms observed in patients with a single infection by RHV or by H1N1v (data not shown). There was no difference except that rhinitis and pharyngitis were significantly more frequent in RHV infection (62.7% vs. 34.1% [p = 0.006] and 39.0% vs. 10.0% [p = 0.001], respectively). Viral multiple infection (including samples with H1N1v) was not associated with a different clinical presentation. Fever and cough were observed in over 90% of the patients (90.6% and 90.3%, respectively), but only 33.3% of these patients had a temperature above 39uC, which was not different from patients with single viral infection (28.6%).
Our results highlight the high frequency of non-influenza viruses involved in acute respiratory infections during the epidemic period of a flu alert as defined by the Réseau Sentinelles according to ILI definition (a sudden fever above 39uC accompanied by myalgia and respiratory signs). These data extent previous observations in Europe reporting high prevalence of RHV infections before seasonal influenza [4, 20] or in 2009, before H1N1v pandemic influenza [1, 8, 9, 11, 21] . We confirm that RHV represent the most frequent aetiology of acute respiratory Table 2 . Age of patients with respiratory samples positive for H1N1v, positive for other respiratory pathogens or negative. infections both in adult and paediatric populations and may represent more than 50% of cases. We show that other viral infections than influenza and RHV may represent up to 30% of aetiologies. We observed differences between the two hospitals, with a higher frequency of parainfluenza and ADV infections in Tours in contrast with a higher frequency of RHV in Paris, likely explained by the higher proportion of paediatric samples collected in Tours. However, despite the distance between the two institutions (about 250 km) and differences between the two populations, both presented similar patterns of high frequency of non-influenza viruses in acute respiratory infections before the flu epidemic wave and a decline when influenza reached epidemic levels.
In the two cities, high frequencies of RHV were seen at the same level with a likely different evolution speed, with sudden increase and decrease in SLS and more progressive variation in TRS. In both institutions, there was a decrease in the proportion and number of RHV diagnoses roughly in parallel with the increase of influenza diagnoses. Indeed, H1N1v exceeds 20% of positive detection's rate only when RHV dropped under 40%. These data are thus consistent with negative interaction of the two epidemics at the population level. It was previously hypothesised that RHV epidemic could interfere with the spread of pandemic influenza [20, 21, 22] . Few in vitro data support this hypothesis. It has been reported that interferon and other cytokines production by RHV infected cells induced a refractory state to virus infection These data include the three patients whose respiratory samples could not be studied with the multiplex assay because of RT-PCR inhibitors. of neighbouring cells [23] . Further work is needed to confirm in vitro and in vivo such negative interactions and if viral interference are really translated to a population level. Analysis of rhinovirus and influenza epidemics in previous years should also help to determine if similar interferences were observed with seasonal influenza and to elaborate modelling and prediction of the spread of influenza according to respiratory viruses' circulation. Systematic extensive screening of respiratory viruses at a national level should be implemented for this purpose.
Very few RSV infections were observed in contrast to usual epidemiology which was characterized the last four past years by a start of epidemics in weeks 44-45 [1] . It has been confirmed by other laboratories and the French InVS that the 2009-10 RSV epidemic was delayed and had a lower impact compared with the previous winter season [1, 24] . Delayed and reduced RSV spread may be due to viral interference between RSV and influenza. Another possible explanation is better prevention behaviour about respiratory infections as recommended by a national campaign including recommendations for hands washing after sneezing and the use of mask [1] .
Influenza infections were mainly detected in patient under 40 years old and no case was found in patients older than 65. These results corroborate previous data suggesting that past seasonal H1N1 infections or vaccination may give partial crossed protection [10, 13, 25] . We have previously shown that the neutralizing titers against pandemic H1N1v virus correlate significantly with neutralizing titers against a seasonal H1N1 virus, and that the H1N1v pandemic influenza virus neutralizing titer was significantly higher in subjects who had recently been inoculated by a seasonal trivalent influenza vaccine [26] .
Viral co-infections were predominantly seen in paediatric patients, as previously described [4, 27, 28, 29] , both in influenza and non-influenza cases at a similar rate. No evidence of more pronounced respiratory impact was seen in these patients.
Our results showed the lack of specific clinical signs associated with proven H1N1v infections. Clinical characteristics did not differ between influenza infections or other viral infections. In particular, the proportion of patients with fever above 39uC was not higher in H1N1v positive patients. In addition, the patients without any evidence of respiratory viral infections did not have different symptoms. These patients may have been infected with other virus not included in the multiplex assay (human Bocavirus, coronavirus HKU1) [9, 10, 11] or were seen too late at the time of viral shedding was cleared [30] . However, to determine how specific the symptoms are for influenza would require to assess also the distribution of respiratory pathogens (H1N1v and other respiratory viruses) and related symptoms in patients presented at the emergency departments in SLS and TRS with respiratory syndromes, but not tested for H1N1v. In addition, despite some underlying conditions that were associated with complications not previously observed in seasonal influenza, most illnesses caused by the H1N1v virus were acute and self-limited [13, 31] . The higher proportion of non influenza viruses reported in ILI in 2009 was thus most likely a consequence of more frequent visits to a doctor for respiratory tract infections than usually observed for fear of the flu pandemic. The general lack of difference in symptoms in the particular context of H1N1v pandemic has therefore to be considered with caution and does not rule out that more significant differences may arise in future influenza epidemics with other influenza viruses. Our data confirm that it may be virtually impossible to recognize symptoms heralding H1N1v infections and virological data should be helpful along with clinical reports to monitor influenza epidemic [10] .
Molecular multiplex detection has recently emerged as a potent diagnostic tool to determine acute respiratory infections' aetiologies [11, 32, 33] . These data show that sensitive molecular multiplex detection of respiratory viruses is feasible and efficient for the detection of virus involved in acute respiratory infections and provides insights into their epidemic profile. Our results confirm the performance of RespiFinder19H assay to detecting respiratory viruses in the general population as recently shown in transplant patients with ILI [34] . RespiFinder19H confirmed all H1N1 infections detected by the CDC reference assay and was able to identify two additional H1N1 cases suggesting a high sensitivity of this multiplex assay to detect influenza A infections.
In conclusion, our results highlight that successive and mixed outbreaks of respiratory viral infections may affect influenza epidemiology and can lead to misinterpret the early development of a flu epidemic. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
|
What are the criteria used to define an influenza-like illness in France?
| 5,264 |
a sudden fever above 39uC with myalgia and respiratory signs
| 2,913 |
1,602 |
High Burden of Non-Influenza Viruses in Influenza-Like Illness in the Early Weeks of H1N1v Epidemic in France
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3157400/
SHA: f4c1afe385e9e31eb5678e15a3c280ba97326554
Authors: Schnepf, Nathalie; Resche-Rigon, Matthieu; Chaillon, Antoine; Scemla, Anne; Gras, Guillaume; Semoun, Oren; Taboulet, Pierre; Molina, Jean-Michel; Simon, François; Goudeau, Alain; LeGoff, Jérôme
Date: 2011-08-17
DOI: 10.1371/journal.pone.0023514
License: cc-by
Abstract: BACKGROUND: Influenza-like illness (ILI) may be caused by a variety of pathogens. Clinical observations are of little help to recognise myxovirus infection and implement appropriate prevention measures. The limited use of molecular tools underestimates the role of other common pathogens. OBJECTIVES: During the early weeks of the 2009–2010 flu pandemic, a clinical and virological survey was conducted in adult and paediatric patients with ILI referred to two French University hospitals in Paris and Tours. Aims were to investigate the different pathogens involved in ILI and describe the associated symptoms. METHODS: H1N1v pandemic influenza diagnosis was performed with real time RT-PCR assay. Other viral aetiologies were investigated by the molecular multiplex assay RespiFinder19®. Clinical data were collected prospectively by physicians using a standard questionnaire. RESULTS: From week 35 to 44, endonasal swabs were collected in 413 patients. Overall, 68 samples (16.5%) were positive for H1N1v. In 13 of them, other respiratory pathogens were also detected. Among H1N1v negative samples, 213 (61.9%) were positive for various respiratory agents, 190 in single infections and 23 in mixed infections. The most prevalent viruses in H1N1v negative single infections were rhinovirus (62.6%), followed by parainfluenza viruses (24.2%) and adenovirus (5.3%). 70.6% of H1N1v cases were identified in patients under 40 years and none after 65 years. There was no difference between clinical symptoms observed in patients infected with H1N1v or with other pathogens. CONCLUSION: Our results highlight the high frequency of non-influenza viruses involved in ILI during the pre-epidemic period of a flu alert and the lack of specific clinical signs associated with influenza infections. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
Text: In order to monitor the spread of influenza and alert health handlers, several epidemiological tools have been developed. In France, a network of 1300 general practitioners, ''Réseau Sentinelles'', working throughout the country, provides real-time clinical data used to evaluate regional and national influenza spreading [1, 2] . The criteria used by this network to define clinical influenza-like illness (ILI) are the occurrence of a sudden fever above 39uC with myalgia and respiratory signs. In general no formal viral diagnosis is carried out. The Groupes Régionaux d'Observation de la Grippe (GROG) is a second French network that surveys the emergence and the spread of the influenza viruses [3, 4] . This network is based on clinical surveillance of acute respiratory infections and laboratory analysis of nasal specimens collected from adults and children by volunteer general practitioners and pediatricians.
According to the sentinel network's criteria, French health authorities proclaimed that flu epidemic level was reached during the second week of September 2009 (week 37) [5, 6] . On the contrary, data provided by the GROG showed only sporadic H1N1v activity until the last week of October (week 44) [6, 7] . Thus, it became rapidly obvious that a variety of viruses were circulating in the community and that an overestimation of myxovirus infection was at stake [8, 9, 10, 11] .
As a better knowledge of the epidemic status was a key feature for national healthcare organization, hospital preparedness, patient management and disease control, unambiguous viral diagnosis appeared critical. In France, data on viral aetiologies associated with ILI were at best sporadic and correlations with clinical symptoms were often lacking. Extensive molecular assays to screening for respiratory viruses were not available countrywide for routine diagnosis. Therefore the epidemiological pattern of respiratory pathogens with overlapping seasonality was poorly known.
The aim of the present study was to investigate respiratory pathogens involved in ILI during the early weeks of the 2009-2010 H1N1v diffusion in France (weeks 35 through 44) and describe the associated symptoms in paediatric and adult populations.
This study was a non-interventional study with no addition to usual proceedures. Biological material and clinical data were obtained only for standard viral diagnostic following physicians' prescriptions (no specific sampling, no modification of the sampling protocol, no supplementary question in the national standardized questionnaire). Data analyses were carried out using an anonymized database. According to the French Health Public Law (CSP Art L 1121-1.1), such protocol does not require approval of an ethics committee and is exempted from informed consent application.
In the two academic hospitals, Saint-Louis hospital (SLS) in Paris and Tours hospital (TRS), influenza-like illness (ILI) was defined as a patient suffering from at least one general symptom (fever above 38uC, asthenia, myalgia, shivers or headache) and one respiratory symptom (cough, dyspnoea, rhinitis or pharyngitis), in agreement with the guidelines from the French Institut de Veille Sanitaire (InVS), a governmental institution responsible for surveillance and alert in all domains of public health [12] . Criteria for severe clinical presentation were temperature below 35uC or above 39uC despite antipyretic, cardiac frequency above 120/min, respiratory frequency above 30/min, respiratory distress, systolic arterial pressure below 90 mmHg or altered consciousness. Predisposing factors of critical illness were children younger than one year old, pregnant women, diabetes, chronic pre-existing disease (such as respiratory, cardiovascular, neurologic, renal, hepatic or hematologic diseases) and immunosuppression (associated with HIV infection, organ or hematopoietic stem cells transplantation, receipt of chemotherapy or corticosteroids) [13, 14] . A cluster of suspected influenza infections was defined as at least three possible cases in a week in a closed community (household, school,…) [15] .
In the two institutions, the prescription of H1N1v molecular testing was recommended for patients with ILI and with either a severe clinical presentation, an underlying risk factor of complications or a condition which was not improving under antiviral treatment. Investigation of grouped suspected cases was also recommended. From week 35 (last week of August) to 44 (last week of October), 413 endonasal swabs were collected in 3 ml of Universal Transport Medium (Copan Diagnostics Inc, Murrieta, CA) from adults and children seen in emergency rooms for suspected ILI (Table 1 ) and sent to SLS and TRS laboratories for H1N1v detection. The two microbiology laboratories participated in the reference laboratories network for the detection of pandemic influenza H1N1v.
Clinical data were collected at the time of medical attention and reported by clinicians on a national standardized questionnaire provided by InVS [1, 12] . This questionnaire included the presence or absence of the main general and respiratory symptoms associated with ILI (fever, asthenia, myalgia, shivers, headache, cough, rhinitis, pharyngitis, sudden onset) [12] .
Total nucleic acid was extracted from 400 mL of Universal Transport Medium using the EasyMag System (Biomérieux, Marcy l'Etoile, France) in SLS or the EZ1 Advanced XL (Qiagen, Courtaboeuf, France) in TRS, according to the manufacturers' instructions (elution volume: 100 mL in SLS or 90 mL in TRS). Before extraction, 5 ml of an Internal Amplification Control (IAC) which contained an encephalomyocarditis virus (EMC) RNA transcript was added into the sample.
Pandemic H1N1v infection was diagnosed by real-time reverse transcription-PCR (RT-PCR) assay on a 7500 Real Time PCR System (Applied Biosystems, Foster City, CA) according to the protocol of the Centers for Disease Control (CDC) [16] . Other respiratory infections were investigated by a multiplex molecular assay based on the Multiplex Ligation-dependent Probe-Amplification (MLPA) technology (RespiFinder19H, Pathofinder, Maastricht, The Netherlands) that allows the detection and differentiation of 14 respiratory viruses, including influenza virus A (InfA), influenza virus B (InfB), rhinovirus (RHV), parainfluenza viruses 1 to 4 (PIV-1 to PIV-4), human metapneumovirus (hMPV), adenovirus (ADV), respiratory syncytial virus A (RSVA), respiratory syncytial virus B (RSVB) and human coronaviruses 229E, OC43 and NL63 (Cor-229E, Cor-OC43, Cor-NL63) [17] . The test allows also the detection of H5N1 influenza A virus and of four bacteria: Chlamydophila pneumoniae (CP), Mycoplasma pneumoniae (MP), Legionella pneumophila (LP) and Bordetella pertussis (BP). The amplified MLPA products were analyzed on an ABI 3100 genetic analyzer (Applied Biosystems, Foster City, CA). Fragment sizing analysis was performed with the GeneMarker software (SoftGenetics, LLC, State College, PA).
Further testing for H1N1v was carried out with Simplexa TM Influenza A H1N1 (2009) (Focus Diagnostics, Cypress, California) when the CDC real time RT-PCR assay was negative for H1N1 and the RespiFinder19H assay was positive for Influenza A. If this latter assay was negative, H3N2 typing was performed as previously described [18] .
Data from our study are summarized as frequencies and percentages for categorical variables. Quantitative variables are presented as medians, 25th and 75th percentiles. To compare those variables according to the viral infection status, Fisher tests
By using CDC reference assay, H1N1v was detected in 66 samples out of 413 (16.6%), more frequently in SLS (38 samples) than in TRS (28 samples) (p,10 24 ). Overall, weekly percentage of H1N1v positive endonasal swabs remained under 10% until week 41 and increase significantly after (P Trend ,0.0001) ( Figure 1 ). Rate of H1N1v detection reached 30% in SLS at week 42 and in TRS at week 44. Overall, this rate was in agreement with results provided by the GROG network, showing an earlier start of H1N1v epidemic in Paris area [7, 19] .
All 413 nucleic acid extracts were analyzed using the RespiFinder19H assay ( Figure 2 ). Sixty six patients tested H1N1v positive with CDC real time RT-PCR assay were confirmed with the multiplex assay. Thirteen were also co-infected by one or two other respiratory pathogens (multiple infections) ( Figure 2 ). Three of the 347 H1N1v negative samples could not be studied with the multiplex assay because they contained RT-PCR inhibitors (no amplification of the internal control). Two hundred and fifteen (62.5%) of the remaining 344 H1N1v negative samples were found positive for at least one respiratory pathogen ( Figure 2 ). Two hundred and twelve were positive for non influenza pathogens (189 single infections and 23 mixed infections with two, three or four viruses) and three additional single infections by influenza A were identified in SLS, including two by pandemic H1N1v and one by seasonal H3N2, as determined after molecular typing (data not shown).
Overall, 68 patients (16.5%) were then positive for H1N1v, one for H3N2 and 212 for non influenza pathogens. There were 245 single infections (55 with H1N1v and 190 with other respiratory pathogens) and 36 mixed infections (13 with H1N1v and 23 without H1N1v) ( Figure 2 ).
Among H1N1v negative single infections, the most prevalent viruses were rhinovirus (62.6%, 119 patients), followed by parainfluenza viruses 1 to 4 (24.2%, 46 patients), adenovirus (5.3%, 10 patients), human coronavirus 229E, OC43 and NL63 (3.2%, 6 patients) and respiratory syncytial virus A and B (2.6%, 5 patients) (Figure 2 ). In addition, RespiFinder19H assay identified three patients with bacterial infection, two with Mycoplasma pneumoniae (one 25 years old female in SLS and one 39 years old female in TRS) and one with Bordetella pertussis (one 60 years old male in SLS). No single infection by influenza B, hMPV, Chlamydophila pneumoniae or Legionella pneumophila was identified ( Figure 2 To analyze if viral co-infections occurred more frequently for some viruses, we carried out a two by two comparisons, that showed a higher proportion of co-infection only for ADV (p = 0.05).
Non-influenza respiratory viruses presented a different epidemic profile compared to H1N1v. Overall, in both hospitals, weekly rate of non-H1N1v respiratory viruses whether alone or involved in co-infection increased between week 37 and 39 (from 51.4% to 81.3%) and then consistently decreased ( Figure 3 ). RHV infections that represented nearly half of non-H1N1v viral infections (141 out of 213, 66.2%) were a significant contributing factor. In both hospitals, emergence of H1N1v cases was associated with a rapid decline of RHV rate of infection from 50-60% down to less than 20% with a one to two weeks gap between SLS and TRS.
Data on age ( In both institutions, 85.5% (106/124) children younger than 15 years of age were infected by at least one respiratory pathogen ( Table 2 ). H1N1v infected patients were not significantly younger than H1N1v non infected patients (27 years old vs. 25 years old, p = 0.80) (Figure 4) . However, 70.6% (48/68) of H1N1v cases were identified in patients under 40 years old (22 in SLS and 26 in TRS) and no case was observed in patients older than 65 years ( Table 2) . PIV infection occurred in very young patients (median (Figure 4) . Consequently, PIV and ADV were more frequently detected in the younger population of TRS versus SLS (p,10 24 and p,10 23 respectively). In contrast, although individuals with RHV infection were slightly younger than individuals without (median age = 24 vs. 29 for patients without RHV, p = 0.05) (Figure 4) , influenza-like illness associated with RHV was more frequent in SLS than in TRS (p = 0.012). Finally, patients with viral multiple infection were significantly younger than those with single infection (median, IDR: 4, 2-18.5 vs. 25, 6-43) and rates of mixed infection At the time of medical attention, 383 (92.7%) standardized clinical questionnaires were collected out of 413 patients. Four of them could not be exploited because they were too incomplete. A review of the 379 workable questionnaires showed that 90.8% (344/379) of the patients included in this study fulfilled the criteria of ILI as defined above, and 52.5% had either a severe clinical presentation or an underlying risk factor of complications (45.9%, 174/379), or were in a suspected cluster of grouped cases (6.6%, 25/379).
Overall, most patients have fever (93.9%) and cough (86.1%) ( Table 3) . Other classical clinical signs associated with ILI such as asthenia, myalgia, shivers, headache, rhinitis or pharyngitis were less frequent. A sudden onset was also described in 59.2% of cases. Only 32.5% of the patients had a temperature above 39uC; the age of these patients ranged from zero to 86 years, with a median age of 32 years and a mean age of 34 years (data not shown).
In H1N1v infected patients (including single and multiple infections), the main symptoms were also fever (98.2%) and cough (89.5%) ( We then compared clinical characteristics between patients positive for H1N1v, patients positive for other respiratory pathogens and negative for H1N1v and patients without any detection of respiratory pathogens (as detected with RespiFin-der19H) ( Table 3 ). There was no difference between the three groups except for fever, cough, pharyngitis. However for these latter symptoms, the comparison between patients positive for H1N1v and those positive for other respiratory pathogens or between patients positive for H1N1v and those without any detection of respiratory pathogens, showed no difference except for pharyngitis, which was less frequent in patients positive for H1N1v than in patients positive for other respiratory pathogens ( Table 3) .
As RHV was the most frequent aetiology in ILI, we also compared clinical symptoms observed in patients with a single infection by RHV or by H1N1v (data not shown). There was no difference except that rhinitis and pharyngitis were significantly more frequent in RHV infection (62.7% vs. 34.1% [p = 0.006] and 39.0% vs. 10.0% [p = 0.001], respectively). Viral multiple infection (including samples with H1N1v) was not associated with a different clinical presentation. Fever and cough were observed in over 90% of the patients (90.6% and 90.3%, respectively), but only 33.3% of these patients had a temperature above 39uC, which was not different from patients with single viral infection (28.6%).
Our results highlight the high frequency of non-influenza viruses involved in acute respiratory infections during the epidemic period of a flu alert as defined by the Réseau Sentinelles according to ILI definition (a sudden fever above 39uC accompanied by myalgia and respiratory signs). These data extent previous observations in Europe reporting high prevalence of RHV infections before seasonal influenza [4, 20] or in 2009, before H1N1v pandemic influenza [1, 8, 9, 11, 21] . We confirm that RHV represent the most frequent aetiology of acute respiratory Table 2 . Age of patients with respiratory samples positive for H1N1v, positive for other respiratory pathogens or negative. infections both in adult and paediatric populations and may represent more than 50% of cases. We show that other viral infections than influenza and RHV may represent up to 30% of aetiologies. We observed differences between the two hospitals, with a higher frequency of parainfluenza and ADV infections in Tours in contrast with a higher frequency of RHV in Paris, likely explained by the higher proportion of paediatric samples collected in Tours. However, despite the distance between the two institutions (about 250 km) and differences between the two populations, both presented similar patterns of high frequency of non-influenza viruses in acute respiratory infections before the flu epidemic wave and a decline when influenza reached epidemic levels.
In the two cities, high frequencies of RHV were seen at the same level with a likely different evolution speed, with sudden increase and decrease in SLS and more progressive variation in TRS. In both institutions, there was a decrease in the proportion and number of RHV diagnoses roughly in parallel with the increase of influenza diagnoses. Indeed, H1N1v exceeds 20% of positive detection's rate only when RHV dropped under 40%. These data are thus consistent with negative interaction of the two epidemics at the population level. It was previously hypothesised that RHV epidemic could interfere with the spread of pandemic influenza [20, 21, 22] . Few in vitro data support this hypothesis. It has been reported that interferon and other cytokines production by RHV infected cells induced a refractory state to virus infection These data include the three patients whose respiratory samples could not be studied with the multiplex assay because of RT-PCR inhibitors. of neighbouring cells [23] . Further work is needed to confirm in vitro and in vivo such negative interactions and if viral interference are really translated to a population level. Analysis of rhinovirus and influenza epidemics in previous years should also help to determine if similar interferences were observed with seasonal influenza and to elaborate modelling and prediction of the spread of influenza according to respiratory viruses' circulation. Systematic extensive screening of respiratory viruses at a national level should be implemented for this purpose.
Very few RSV infections were observed in contrast to usual epidemiology which was characterized the last four past years by a start of epidemics in weeks 44-45 [1] . It has been confirmed by other laboratories and the French InVS that the 2009-10 RSV epidemic was delayed and had a lower impact compared with the previous winter season [1, 24] . Delayed and reduced RSV spread may be due to viral interference between RSV and influenza. Another possible explanation is better prevention behaviour about respiratory infections as recommended by a national campaign including recommendations for hands washing after sneezing and the use of mask [1] .
Influenza infections were mainly detected in patient under 40 years old and no case was found in patients older than 65. These results corroborate previous data suggesting that past seasonal H1N1 infections or vaccination may give partial crossed protection [10, 13, 25] . We have previously shown that the neutralizing titers against pandemic H1N1v virus correlate significantly with neutralizing titers against a seasonal H1N1 virus, and that the H1N1v pandemic influenza virus neutralizing titer was significantly higher in subjects who had recently been inoculated by a seasonal trivalent influenza vaccine [26] .
Viral co-infections were predominantly seen in paediatric patients, as previously described [4, 27, 28, 29] , both in influenza and non-influenza cases at a similar rate. No evidence of more pronounced respiratory impact was seen in these patients.
Our results showed the lack of specific clinical signs associated with proven H1N1v infections. Clinical characteristics did not differ between influenza infections or other viral infections. In particular, the proportion of patients with fever above 39uC was not higher in H1N1v positive patients. In addition, the patients without any evidence of respiratory viral infections did not have different symptoms. These patients may have been infected with other virus not included in the multiplex assay (human Bocavirus, coronavirus HKU1) [9, 10, 11] or were seen too late at the time of viral shedding was cleared [30] . However, to determine how specific the symptoms are for influenza would require to assess also the distribution of respiratory pathogens (H1N1v and other respiratory viruses) and related symptoms in patients presented at the emergency departments in SLS and TRS with respiratory syndromes, but not tested for H1N1v. In addition, despite some underlying conditions that were associated with complications not previously observed in seasonal influenza, most illnesses caused by the H1N1v virus were acute and self-limited [13, 31] . The higher proportion of non influenza viruses reported in ILI in 2009 was thus most likely a consequence of more frequent visits to a doctor for respiratory tract infections than usually observed for fear of the flu pandemic. The general lack of difference in symptoms in the particular context of H1N1v pandemic has therefore to be considered with caution and does not rule out that more significant differences may arise in future influenza epidemics with other influenza viruses. Our data confirm that it may be virtually impossible to recognize symptoms heralding H1N1v infections and virological data should be helpful along with clinical reports to monitor influenza epidemic [10] .
Molecular multiplex detection has recently emerged as a potent diagnostic tool to determine acute respiratory infections' aetiologies [11, 32, 33] . These data show that sensitive molecular multiplex detection of respiratory viruses is feasible and efficient for the detection of virus involved in acute respiratory infections and provides insights into their epidemic profile. Our results confirm the performance of RespiFinder19H assay to detecting respiratory viruses in the general population as recently shown in transplant patients with ILI [34] . RespiFinder19H confirmed all H1N1 infections detected by the CDC reference assay and was able to identify two additional H1N1 cases suggesting a high sensitivity of this multiplex assay to detect influenza A infections.
In conclusion, our results highlight that successive and mixed outbreaks of respiratory viral infections may affect influenza epidemiology and can lead to misinterpret the early development of a flu epidemic. Rapid diagnostic screening of a large panel of respiratory pathogens may be critical to define and survey the epidemic situation and to provide critical information for patient management.
|
What virus was the most common among the H1N1v negative patients?
| 5,265 |
rhinovirus
| 11,928 |
1,605 |
Livestock Drugs and Disease: The Fatal Combination behind Breeding Failure in Endangered Bearded Vultures
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2994777/
SHA: f4f804bac7b32c84ad6572776df684df2a2e5fda
Authors: Blanco, Guillermo; Lemus, Jesús A.
Date: 2010-11-30
DOI: 10.1371/journal.pone.0014163
License: cc-by
Abstract: There is increasing concern about the impact of veterinary drugs and livestock pathogens as factors damaging wildlife health, especially of threatened avian scavengers feeding upon medicated livestock carcasses. We conducted a comprehensive study of failed eggs and dead nestlings in bearded vultures (Gypaetus barbatus) to attempt to elucidate the proximate causes of breeding failure behind the recent decline in productivity in the Spanish Pyrenees. We found high concentrations of multiple veterinary drugs, primarily fluoroquinolones, in most failed eggs and nestlings, associated with multiple internal organ damage and livestock pathogens causing disease, especially septicaemia by swine pathogens and infectious bursal disease. The combined impact of drugs and disease as stochastic factors may result in potentially devastating effects exacerbating an already high risk of extinction and should be considered in current conservation programs for bearded vultures and other scavenger species, especially in regards to dangerous veterinary drugs and highly pathogenic poultry viruses.
Text: Environmental pollutants are increasingly documented as a driver of wildlife endangerment due to their roles in organ damage, hormonal disruption and alteration of the immune system [1, 2] . Disease may also facilitate endangerment and extinction at global and local scales, especially when pathogens interact with other drivers such as pollutants [3] . There is increasing concern about the impact of veterinary drugs and livestock pathogens as factors damaging wildlife health [4] [5] [6] , and even causing declines approaching extinction [7] . These threats may be especially detrimental to wildlife as they increasingly concur and interact as a consequence of the elimination of livestock residues containing veterinary pharmaceuticals and resistant pathogens due to growing intensive livestock operations worldwide [6, 8, 9] . In particular, the ingestion of antimicrobials, primarily fluoroquinolones, has been recently related to immunodepression-mediated acquisition of opportunistic pathogens and disease, as well as to organ damage in nestling vultures [6, 10, 11] . Fluoroquinolone residues have also been found in avian scavenger eggs and are associated with severe alterations in the development of embryo cartilage and bones that could preclude embryo movement and subsequently normal development, pre-hatch position and successful hatching [12] . Therefore, antimicrobials and other drugs may negatively affect embryo and nestling health with potentially devastating consequences on breeding success and conservation of vultures and other threatened avian scavengers.
The bearded vulture (Gypaetus barbatus) is one of the most endangered birds in Europe, with a main stronghold in the Pyrenees. Increasing declines in productivity (average number of fledglings raised per territorial pair) have recently been reported in the Spanish Pyrenees associated with habitat saturation processes [13, 14] . Given that bearded vultures may raise only one fledgling per breeding attempt, this productivity decline should be linked to increasing breeding failure when the proportion of territorial pairs that are breeding does not greatly vary with time [15] . The proximate mechanisms by which density can affect productivity have been investigated, including habitat heterogeneity, with progressively poorer territories being used, territory shrinkage and interference with breeders and floaters [13] . However, the proximate causes of breeding failure are poorly known despite the long-term interests in the conservation of this species [16] . To evaluate these causes, the examination of failed eggs and dead nestlings is imperative, including the study of the presence and impact of injury, developmental problems, poor nutritional condition, pollutants, organ damage, pathogens causing disease, etc. in order to determine the most likely cause of breeding failure.
Here, we conducted a comprehensive study of failed eggs and dead nestling bearded vultures collected during recent years in the Pyrenees. Both the productivity and survival rates of adults and young birds have reached the lowest values since the bovine spongiform encephalopathy (BSE) crisis [13, 14, 17] . This temporal decline could be related to illegal poisoning [17] and recent changes in the abundance, distribution and quality of carrion available to avian scavengers as a consequence of EU regulations derived from the BSE crisis [6, [18] [19] [20] . In particular, the BSE crisis caused the lack or scarcity of unstabled livestock available to scavengers and their subsequent increase in the consumption of carrion from stabled livestock, which is intensively medicated [21] . Therefore, we specifically focused on determining whether breeding failure in bearded vultures is related to the ingestion of veterinary drugs from stabled livestock carrion, as documented in other avian scavenger species [12] . We also assessed the potential effects of veterinary drugs on embryo damage and immunodepression increasing the probability of acquisition and proliferation of pathogens causing fatal disease [6, [10] [11] [12] 21] . Because veterinary drugs should be exclusively acquired from the ingestion of carrion from livestock medicated to combat disease, we predict that their presence should be associated with that of pathogens acquired from the same livestock, especially poultry pathogens more likely transmitted between avian species [22] . Alternatively, if the temporal decline in productivity was primarily associated with breeding failure due to the effects of habitat saturation processes [13, 17] , we should expect egg and nestling mortality to be directly related to developmental and nutritional problems indicating progressively lower quality territories (e.g. embryo emaciation, nestling starvation) and interference by both conspecifics and heterospecifics (e.g. incubation failure, injury due to predation attempts or disturbance).
Failed eggs (n = 5) and dead nestlings (n = 4) were collected from bearded vulture nests located in the Spanish Pyrenees between 2005 and 2008. The study of this material did not require of the approval of an ethics committee because it was collected after breeding failure (egg or nestling death) was confirmed in the field. Three of the specimens (two nestlings and one egg) were collected in 2005, 2007 and 2008 from a particular territory. Eggs and nestlings were collected after breeding failure and frozen. Necropsies were performed on all specimens according to standard protocols [12] . The age of embryos and nestlings were estimated according to size and development. Samples of liver, kidney, spleen, large and small intestines, lungs, brain, lymphoid organs (thymus, bursa of Fabricius, Peyer's patches) and knee joints were fixed in 10% buffered formalin, sectioned at 4 mm and stained for histopathological analysis [10, 12] .
Liver (dead nestlings and failed embryos) and yolk (failed embryos) were used for the determination of the presence of veterinary drugs, including fluoroquinolones (enrofloxacin and ciprofloxacin), other antimicrobials (amoxicillin and oxytetracycline), non-steroideal anti-inflamatories (NSAIDs) such as diclofenac, flunixin meglumine, ketoprofen, ibuprofen, meloxicam, sodium salicylate, acetaminophen, and antiparasitics (metronidazole, diclazuril, fenbendazole, ivermectin) as described previously [12] . The limits of quantification, percentage recoveries, and interand intra-assay reproducibility were adequate [10, 12] .
Other contaminants potentially affecting eggs and embryos were determined in liver, including heavy metals (Cd, Zn, Pb and Hg), following Blanco et al. [23] , dithiocarbamate thiram, disulfuram, polybrominated diphenyl ethers, organochlorines and brominated flame retardants, following Lemus et al. [12] and carbamate and organophosphate pesticides (carbofuran, aldicarb and fenthion) following Elliot et al. [24] . We measured brain cholinesterase activity to assess early exposure to anticholinesterase pesticides [25] . Potential contamination was assessed by comparison with levels from apparently normal wild birds of other species [26] in the absence of basal levels for bearded vultures.
Determination of bacterial and fungal pathogens were conducted by sampling oropharynx, lung, liver, kidney, spleen, and intestine with sterile swabs and cultured using standard microbiology protocols [10, 12, 27, 28] . Salmonella serotypes and phage types were determined in the Spanish Reference Laboratory (Laboratorio Central Veterinario, Algete, Madrid). For confirmation of the identification of the alpha hemolytic Streptococcus pneumoniae we used a specific identification test (Accuprobe, Salem, MA) based on the detection of specific ribosomal RNA sequences. Samples of lesions found in internal organs and tissues during necropsies were taken with sterile swabs and cultured using the same standard microbiology protocols. In addition, we determined the presence of selected avian pathogens, including bacterial, viral, fungal, and protozoan pathogens by means of PCR-based methods (see Table S1 for details). The presence of Chlamyophila psittaci and Mycoplasma sp. in blood was determined as described previously [29, 30] . The presence of poxvirus, the paramyxovirus causing Newcastle disease, the serotypes H5, H7 and H9 of avian influenza, falcon adenovirus, circovirus, herpesvirus, polyomavirus, reovirus and West Nile virus were determined following the PCR-based methods available in the literature [31] [32] [33] [34] [35] [36] [37] [38] [39] . We also searched for helminths and protozoans in the gastrointestinal tract by macroscopic and microscopic observations using standard protocols [40] .
Specific immunocytochemical procedures were used for detection of mielodepressive virus, including the alphaherpesvirus causing Marek disease [41] in kidney and bursa of Fabricius, the gyrovirus causing infectious chicken anaemia [42] in thymus and bone marrow, the birnavirus causing infectious bursal disease (IBD, [43] ) in bursa of Fabricius, and the coronavirus causing chicken infectious bronchitis in kidney [44] . In addition, we conducted a specific immunocytochemical procedure for West Nile virus antigen detection [45] in brain, spinal medulla, thymus and thyroid. All immunohistochemistry analyses were conducted at the Department of Veterinary Anatomy, Veterinary Faculty, Universidad Complutense de Madrid, Spain and at the Pathology Department of the Veterinary Faculty, University of Utrecht, The Netherlands. The presence of these viruses was also determined by PCR-based methods [43, [46] [47] [48] .
All dead nestlings and three of five unhatched embryos showed two to six different veterinary drugs in liver (nestlings) and egg yolk (embryos). In addition, the two embryos with fluoroquinolones in the yolk also had them in the liver (Table 1) . Fluoroquinolones were the most prevalent drugs and showed the highest concentrations (Table 1) . Other drugs such as NSAIDs and antiparasitics were found in most nestlings at variable concentrations, but in no eggs (Table 1) . Other toxic compounds were detected in lower prevalence and concentrations (see Table 1 for those more relevant values; all insecticides were found at concentrations ,0.001 ppb), which was further supported by basal levels of brain cholinesterase (Table 1) .
Dead embryos and nestlings showed a moderate to good nutritional state. Major histopathological lesions were primarily located in the kidney, including glomerulonephritis and/or glomerulonephrosis present in all individuals with fluoroquinolones, but not in those without drugs (Table 1 ). All individuals with fluoroquinolones also showed joint lesions, including arthritis and/ or arthrosis of the long bone articulations, as well as massive osseous stroma of the spongeous bones.
The fungi Candida albicans was isolated from the oral cavity of five individuals. All individuals showed non-specific mixedbacterial flora. Enterotoxigenic Escherichia coli and Salmonella spp. were isolated in four cases ( Table 1 (cont.) *Samples from the same territory in different years. 1 Veterinary drugs. EN: enrofloxacin (mg/g), CI: ciprofloxacin (mg/g), OX: oxytetracyclin (mg/g), FL: flunixin meglumine (mg/g), AS: sodium salicylate (ng/g), IV: Ivermectin (mg/g). 2 Other toxicants. OR: organochlorines (ng/g), Pb: lead (ng/g). 12. z29 (three cases, see Table 1 ). One individual showed infection by Salmonella enterica enteritidis (see above) and enterotoxigenic Escherichia coli O86 in all examined organs (septicaemia) except brain, which rejected the possibility of post-mortem contamination. Pasteurella multocida was isolated in a single individual that also showed enterotoxigenic Escherichia coli O86 (Table 1) ; all of these individuals contained fluoroquinolones. One of the failed embryos without veterinary drugs showed suppurative myocarditis, multiple microabscesses in head muscles, suppurative leptomeningitis, as well as lower jaw gangrenous inflammation with loss of the osseous stroma due to a mixed infection with Streptococcus suis and Streptococcus pneumoniae in brain, meninges and neck muscles; this embryo also showed infection by chicken infectious bronchitis ( Table 1 ). Both immunocytochemistry for the detection of poultry viruses and PCR pathogen survey were positive to IBDV in six individuals with fluoroquinolones (Table 1) . Immunocytochemical procedures failed to detect West Nile virus antigens in individuals in which PCR for this virus had been positive. Parasitology was negative for all helminths, helminth eggs and protozoans.
We found multiple veterinary drugs, primarily fluoroquinolones, in most failed eggs and dead nestling bearded vultures from the Pyrenees. They also showed multiple internal organ damage and pathogens potentially acquired from medicated livestock carrion, especially viruses often infecting poultry. Recorded drug concentrations were among the highest reported in avian scavengers [6, [10] [11] [12] 21] . NSAIDs and antiparasitics were found in lower prevalence than fluoroquinolones, but at higher concentrations than those found in other avian scavengers, especially for flunixin meglumine and sodium salicylate [6, 12, 21] . On the contrary, we found no sterile eggs, poor nutritional conditions or injury in any failed embryo or nestling. Other pollutants were found in low prevalence and concentrations posing low risk to embryo and nestling health.
Fluoroquinolones may cause generalized direct developmental damage precluding embryo hatching, physiological alterations due to their impact on liver and kidney and immunodepression reducing resistance to opportunistic pathogens [6, [10] [11] [12] 21] . These pathogens may be acquired at the same time that drugs used to treat diseased livestock are ingested, as indicated by their high prevalence in embryos and nestlings. Therefore, despite the relatively small sample size resulting from low abundance, endangerment and logistic difficulties in reaching nests in this species, the results provide evidence of a combined impact of veterinary drugs and livestock disease as the primary cause of breeding failure in the sampled individuals.
The presence of West Nile virus is not likely to be associated with nestling disease or mortality because the lack of lesions in target tissues and viral antigen particles in the immunohistochemistry study. Fatal septicaemia caused by Streptococcus suis, one of the most important swine pathogens worldwide [49] , in combination with septicaemia from Streptococcus pneumoniae and infection by chicken infectious bronchitis virus were found in a single embryo. This concentration of livestock pathogens has not been reported before and, to our knowledge, this is the first report of the three pathogens causing disease in a wild bird. Other pathogens recorded in embryos and nestlings, including Salmonella serotypes and phages typical of livestock [50] , and enterotoxigenic Escherichia coli O86 causing septicaemia, were potentially transmitted by consumption of carcasses of infected poultry and other livestock [22, 27, 28] . In addition, we found that the IBD virus infected most individuals alone or together with other pathogens also potentially acquired from livestock carrion. This virus causes a highly contagious immunosuppressive bursal disease in poultry [51] and may be transmitted to wildlife in contact with poultry waste or by ingestion of carcasses [22, 52] . Nestlings are especially susceptible to IBD because of the primary role of bursa of Fabricius in immune function development at this age. In fact, immunosuppression due to IBD was indicated by the inflammation, necrosis and loss of lymphocytes in the bursa of Fabricius together with the presence of viral antigens recorded by means of immunocytochemical procedures. The potential impact of highly pathogenic and contagious poultry viruses has been previously recognized as a threat to wildlife health due to the increasing contact of wildlife with livestock operations in general, and poultry farms and their residues in particular, in natural areas worldwide [52] [53] [54] [55] . However, damage from IBD virus on the bursa of Fabricius represents, to our knowledge, the first evidence of clinical disease compatible with death caused by this poultry virus in wildlife. The presence of IBD has been not previously recorded in embryos of wild birds, probably because vertical transmission has been ruled out in poultry and, as consequence, it has probably not been evaluated in other species until now. This striking and concerning result could be related to the longer egg development and incubation periods of bearded vultures compared with poultry, and/or due to contrasting environmental conditions during incubation between bearded vultures and poultry. Thus, embryo infection with IBD may occurs via the female or during incubation as a consequence of egg contact between the egg and poultry remains in the nests of bearded vultures, which requires more research.
Despite their potential effects on population dynamics and conservation through a reduction of productivity and changes in mating behaviour [13, 14] , habitat saturation processes were apparently not directly related to particular proximate causes of egg and nestling failure in this study or in these sampled individuals. As an alternative non-mutually exclusive explanation, we suggest that the recent decline in productivity could also be linked to the increasing ingestion of veterinary drugs and acquisition of pathogens from medicated stabled livestock carcasses due to decreasing availability of unstabled livestock carcasses -the traditional primary food of bearded vultures [16] since the BSE crisis [21] , accompanied by a possible increasing use of antibiotics in stabled livestock operations. In this sense, it is remarkable that bearded vultures primarily feed upon livestock bones, which are one of the major target tissues of fluoroquinolones in medicated animals [56] , therefore, rendering this species especially sensitive to the consequences of an increase in the consumption of stabled intensively medicated livestock. The presence of veterinary drugs in eggs implies their previous presence at least in breeding females [12] , but also probably in breeding males and non-breeders frequently using artificial feeding sites and livestock carcass dumps [17] , where veterinary drugs may be ingested from medicated livestock carcasses [10, 21] . Therefore, further research is required to determine the impact of veterinary drugs and livestock disease on fitness of full-grown individuals, including the potentially subtle, sublethal or indirect effects of these factors on population dynamics.
The link between veterinary drugs and livestock disease should be further investigated in scavenger species, because both threats may concur in food and because the immunodepressive effects and other physiological alterations caused by drugs may facilitate the acquisition and proliferation of pathogens [6, 11, 21] . Given that both threats acting together may greatly contribute to breeding failure decreasing productivity, their potential as stochastic factors with potentially devastating effects increasing the risk of extinction should be not overlooked in current conservation programs of bearded vultures and other scavenger species, especially regarding dangerous veterinary drugs and highly pathogenic viruses frequently infecting poultry. In addition, restricted geographic distribution and low genetic variability [57] common to many threatened species may favour pathogen transmission and reduce the ability of a naïve immune system to fight against novel pathogens [3, 28, 58] , making them especially vulnerable to the potential cross-species transmission of highly virulent virus strains able to cause important outbreaks, as reported in poultry [59] [60] [61] .
The association of pollution and disease may further increase extinction risk if it interacts with the effects of habitat saturation processes [13, 14, 17] . These processes may facilitate conspecific contact and interactions also likely to increase intra-and interspecific pathogen transmission rates in breeding and feeding areas, especially of highly contagious poultry diseases [22] . This could be further enhanced by the artificially high numbers of bearded vultures and other scavengers attracted to feeding points and carcass refuse dumps, both as a result of management and due to the scarcity of unstabled livestock carcasses since the BSE crisis [17, 21] . Whatever the potential contribution of underlying ultimate mechanisms reducing productivity, our findings highlight the need to determine the proximate causes of breeding failure and mortality in wildlife populations in order to understand the processes regulating demography from an ecological framework perspective.
Table S1
Found at: doi:10.1371/journal.pone.0014163.s001 (0.05 MB DOC)
|
What was the aim of this study?
| 5,269 |
to elucidate the proximate causes of breeding failure behind the recent decline in productivity in the Spanish Pyrenees
| 662 |
1,605 |
Livestock Drugs and Disease: The Fatal Combination behind Breeding Failure in Endangered Bearded Vultures
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2994777/
SHA: f4f804bac7b32c84ad6572776df684df2a2e5fda
Authors: Blanco, Guillermo; Lemus, Jesús A.
Date: 2010-11-30
DOI: 10.1371/journal.pone.0014163
License: cc-by
Abstract: There is increasing concern about the impact of veterinary drugs and livestock pathogens as factors damaging wildlife health, especially of threatened avian scavengers feeding upon medicated livestock carcasses. We conducted a comprehensive study of failed eggs and dead nestlings in bearded vultures (Gypaetus barbatus) to attempt to elucidate the proximate causes of breeding failure behind the recent decline in productivity in the Spanish Pyrenees. We found high concentrations of multiple veterinary drugs, primarily fluoroquinolones, in most failed eggs and nestlings, associated with multiple internal organ damage and livestock pathogens causing disease, especially septicaemia by swine pathogens and infectious bursal disease. The combined impact of drugs and disease as stochastic factors may result in potentially devastating effects exacerbating an already high risk of extinction and should be considered in current conservation programs for bearded vultures and other scavenger species, especially in regards to dangerous veterinary drugs and highly pathogenic poultry viruses.
Text: Environmental pollutants are increasingly documented as a driver of wildlife endangerment due to their roles in organ damage, hormonal disruption and alteration of the immune system [1, 2] . Disease may also facilitate endangerment and extinction at global and local scales, especially when pathogens interact with other drivers such as pollutants [3] . There is increasing concern about the impact of veterinary drugs and livestock pathogens as factors damaging wildlife health [4] [5] [6] , and even causing declines approaching extinction [7] . These threats may be especially detrimental to wildlife as they increasingly concur and interact as a consequence of the elimination of livestock residues containing veterinary pharmaceuticals and resistant pathogens due to growing intensive livestock operations worldwide [6, 8, 9] . In particular, the ingestion of antimicrobials, primarily fluoroquinolones, has been recently related to immunodepression-mediated acquisition of opportunistic pathogens and disease, as well as to organ damage in nestling vultures [6, 10, 11] . Fluoroquinolone residues have also been found in avian scavenger eggs and are associated with severe alterations in the development of embryo cartilage and bones that could preclude embryo movement and subsequently normal development, pre-hatch position and successful hatching [12] . Therefore, antimicrobials and other drugs may negatively affect embryo and nestling health with potentially devastating consequences on breeding success and conservation of vultures and other threatened avian scavengers.
The bearded vulture (Gypaetus barbatus) is one of the most endangered birds in Europe, with a main stronghold in the Pyrenees. Increasing declines in productivity (average number of fledglings raised per territorial pair) have recently been reported in the Spanish Pyrenees associated with habitat saturation processes [13, 14] . Given that bearded vultures may raise only one fledgling per breeding attempt, this productivity decline should be linked to increasing breeding failure when the proportion of territorial pairs that are breeding does not greatly vary with time [15] . The proximate mechanisms by which density can affect productivity have been investigated, including habitat heterogeneity, with progressively poorer territories being used, territory shrinkage and interference with breeders and floaters [13] . However, the proximate causes of breeding failure are poorly known despite the long-term interests in the conservation of this species [16] . To evaluate these causes, the examination of failed eggs and dead nestlings is imperative, including the study of the presence and impact of injury, developmental problems, poor nutritional condition, pollutants, organ damage, pathogens causing disease, etc. in order to determine the most likely cause of breeding failure.
Here, we conducted a comprehensive study of failed eggs and dead nestling bearded vultures collected during recent years in the Pyrenees. Both the productivity and survival rates of adults and young birds have reached the lowest values since the bovine spongiform encephalopathy (BSE) crisis [13, 14, 17] . This temporal decline could be related to illegal poisoning [17] and recent changes in the abundance, distribution and quality of carrion available to avian scavengers as a consequence of EU regulations derived from the BSE crisis [6, [18] [19] [20] . In particular, the BSE crisis caused the lack or scarcity of unstabled livestock available to scavengers and their subsequent increase in the consumption of carrion from stabled livestock, which is intensively medicated [21] . Therefore, we specifically focused on determining whether breeding failure in bearded vultures is related to the ingestion of veterinary drugs from stabled livestock carrion, as documented in other avian scavenger species [12] . We also assessed the potential effects of veterinary drugs on embryo damage and immunodepression increasing the probability of acquisition and proliferation of pathogens causing fatal disease [6, [10] [11] [12] 21] . Because veterinary drugs should be exclusively acquired from the ingestion of carrion from livestock medicated to combat disease, we predict that their presence should be associated with that of pathogens acquired from the same livestock, especially poultry pathogens more likely transmitted between avian species [22] . Alternatively, if the temporal decline in productivity was primarily associated with breeding failure due to the effects of habitat saturation processes [13, 17] , we should expect egg and nestling mortality to be directly related to developmental and nutritional problems indicating progressively lower quality territories (e.g. embryo emaciation, nestling starvation) and interference by both conspecifics and heterospecifics (e.g. incubation failure, injury due to predation attempts or disturbance).
Failed eggs (n = 5) and dead nestlings (n = 4) were collected from bearded vulture nests located in the Spanish Pyrenees between 2005 and 2008. The study of this material did not require of the approval of an ethics committee because it was collected after breeding failure (egg or nestling death) was confirmed in the field. Three of the specimens (two nestlings and one egg) were collected in 2005, 2007 and 2008 from a particular territory. Eggs and nestlings were collected after breeding failure and frozen. Necropsies were performed on all specimens according to standard protocols [12] . The age of embryos and nestlings were estimated according to size and development. Samples of liver, kidney, spleen, large and small intestines, lungs, brain, lymphoid organs (thymus, bursa of Fabricius, Peyer's patches) and knee joints were fixed in 10% buffered formalin, sectioned at 4 mm and stained for histopathological analysis [10, 12] .
Liver (dead nestlings and failed embryos) and yolk (failed embryos) were used for the determination of the presence of veterinary drugs, including fluoroquinolones (enrofloxacin and ciprofloxacin), other antimicrobials (amoxicillin and oxytetracycline), non-steroideal anti-inflamatories (NSAIDs) such as diclofenac, flunixin meglumine, ketoprofen, ibuprofen, meloxicam, sodium salicylate, acetaminophen, and antiparasitics (metronidazole, diclazuril, fenbendazole, ivermectin) as described previously [12] . The limits of quantification, percentage recoveries, and interand intra-assay reproducibility were adequate [10, 12] .
Other contaminants potentially affecting eggs and embryos were determined in liver, including heavy metals (Cd, Zn, Pb and Hg), following Blanco et al. [23] , dithiocarbamate thiram, disulfuram, polybrominated diphenyl ethers, organochlorines and brominated flame retardants, following Lemus et al. [12] and carbamate and organophosphate pesticides (carbofuran, aldicarb and fenthion) following Elliot et al. [24] . We measured brain cholinesterase activity to assess early exposure to anticholinesterase pesticides [25] . Potential contamination was assessed by comparison with levels from apparently normal wild birds of other species [26] in the absence of basal levels for bearded vultures.
Determination of bacterial and fungal pathogens were conducted by sampling oropharynx, lung, liver, kidney, spleen, and intestine with sterile swabs and cultured using standard microbiology protocols [10, 12, 27, 28] . Salmonella serotypes and phage types were determined in the Spanish Reference Laboratory (Laboratorio Central Veterinario, Algete, Madrid). For confirmation of the identification of the alpha hemolytic Streptococcus pneumoniae we used a specific identification test (Accuprobe, Salem, MA) based on the detection of specific ribosomal RNA sequences. Samples of lesions found in internal organs and tissues during necropsies were taken with sterile swabs and cultured using the same standard microbiology protocols. In addition, we determined the presence of selected avian pathogens, including bacterial, viral, fungal, and protozoan pathogens by means of PCR-based methods (see Table S1 for details). The presence of Chlamyophila psittaci and Mycoplasma sp. in blood was determined as described previously [29, 30] . The presence of poxvirus, the paramyxovirus causing Newcastle disease, the serotypes H5, H7 and H9 of avian influenza, falcon adenovirus, circovirus, herpesvirus, polyomavirus, reovirus and West Nile virus were determined following the PCR-based methods available in the literature [31] [32] [33] [34] [35] [36] [37] [38] [39] . We also searched for helminths and protozoans in the gastrointestinal tract by macroscopic and microscopic observations using standard protocols [40] .
Specific immunocytochemical procedures were used for detection of mielodepressive virus, including the alphaherpesvirus causing Marek disease [41] in kidney and bursa of Fabricius, the gyrovirus causing infectious chicken anaemia [42] in thymus and bone marrow, the birnavirus causing infectious bursal disease (IBD, [43] ) in bursa of Fabricius, and the coronavirus causing chicken infectious bronchitis in kidney [44] . In addition, we conducted a specific immunocytochemical procedure for West Nile virus antigen detection [45] in brain, spinal medulla, thymus and thyroid. All immunohistochemistry analyses were conducted at the Department of Veterinary Anatomy, Veterinary Faculty, Universidad Complutense de Madrid, Spain and at the Pathology Department of the Veterinary Faculty, University of Utrecht, The Netherlands. The presence of these viruses was also determined by PCR-based methods [43, [46] [47] [48] .
All dead nestlings and three of five unhatched embryos showed two to six different veterinary drugs in liver (nestlings) and egg yolk (embryos). In addition, the two embryos with fluoroquinolones in the yolk also had them in the liver (Table 1) . Fluoroquinolones were the most prevalent drugs and showed the highest concentrations (Table 1) . Other drugs such as NSAIDs and antiparasitics were found in most nestlings at variable concentrations, but in no eggs (Table 1) . Other toxic compounds were detected in lower prevalence and concentrations (see Table 1 for those more relevant values; all insecticides were found at concentrations ,0.001 ppb), which was further supported by basal levels of brain cholinesterase (Table 1) .
Dead embryos and nestlings showed a moderate to good nutritional state. Major histopathological lesions were primarily located in the kidney, including glomerulonephritis and/or glomerulonephrosis present in all individuals with fluoroquinolones, but not in those without drugs (Table 1 ). All individuals with fluoroquinolones also showed joint lesions, including arthritis and/ or arthrosis of the long bone articulations, as well as massive osseous stroma of the spongeous bones.
The fungi Candida albicans was isolated from the oral cavity of five individuals. All individuals showed non-specific mixedbacterial flora. Enterotoxigenic Escherichia coli and Salmonella spp. were isolated in four cases ( Table 1 (cont.) *Samples from the same territory in different years. 1 Veterinary drugs. EN: enrofloxacin (mg/g), CI: ciprofloxacin (mg/g), OX: oxytetracyclin (mg/g), FL: flunixin meglumine (mg/g), AS: sodium salicylate (ng/g), IV: Ivermectin (mg/g). 2 Other toxicants. OR: organochlorines (ng/g), Pb: lead (ng/g). 12. z29 (three cases, see Table 1 ). One individual showed infection by Salmonella enterica enteritidis (see above) and enterotoxigenic Escherichia coli O86 in all examined organs (septicaemia) except brain, which rejected the possibility of post-mortem contamination. Pasteurella multocida was isolated in a single individual that also showed enterotoxigenic Escherichia coli O86 (Table 1) ; all of these individuals contained fluoroquinolones. One of the failed embryos without veterinary drugs showed suppurative myocarditis, multiple microabscesses in head muscles, suppurative leptomeningitis, as well as lower jaw gangrenous inflammation with loss of the osseous stroma due to a mixed infection with Streptococcus suis and Streptococcus pneumoniae in brain, meninges and neck muscles; this embryo also showed infection by chicken infectious bronchitis ( Table 1 ). Both immunocytochemistry for the detection of poultry viruses and PCR pathogen survey were positive to IBDV in six individuals with fluoroquinolones (Table 1) . Immunocytochemical procedures failed to detect West Nile virus antigens in individuals in which PCR for this virus had been positive. Parasitology was negative for all helminths, helminth eggs and protozoans.
We found multiple veterinary drugs, primarily fluoroquinolones, in most failed eggs and dead nestling bearded vultures from the Pyrenees. They also showed multiple internal organ damage and pathogens potentially acquired from medicated livestock carrion, especially viruses often infecting poultry. Recorded drug concentrations were among the highest reported in avian scavengers [6, [10] [11] [12] 21] . NSAIDs and antiparasitics were found in lower prevalence than fluoroquinolones, but at higher concentrations than those found in other avian scavengers, especially for flunixin meglumine and sodium salicylate [6, 12, 21] . On the contrary, we found no sterile eggs, poor nutritional conditions or injury in any failed embryo or nestling. Other pollutants were found in low prevalence and concentrations posing low risk to embryo and nestling health.
Fluoroquinolones may cause generalized direct developmental damage precluding embryo hatching, physiological alterations due to their impact on liver and kidney and immunodepression reducing resistance to opportunistic pathogens [6, [10] [11] [12] 21] . These pathogens may be acquired at the same time that drugs used to treat diseased livestock are ingested, as indicated by their high prevalence in embryos and nestlings. Therefore, despite the relatively small sample size resulting from low abundance, endangerment and logistic difficulties in reaching nests in this species, the results provide evidence of a combined impact of veterinary drugs and livestock disease as the primary cause of breeding failure in the sampled individuals.
The presence of West Nile virus is not likely to be associated with nestling disease or mortality because the lack of lesions in target tissues and viral antigen particles in the immunohistochemistry study. Fatal septicaemia caused by Streptococcus suis, one of the most important swine pathogens worldwide [49] , in combination with septicaemia from Streptococcus pneumoniae and infection by chicken infectious bronchitis virus were found in a single embryo. This concentration of livestock pathogens has not been reported before and, to our knowledge, this is the first report of the three pathogens causing disease in a wild bird. Other pathogens recorded in embryos and nestlings, including Salmonella serotypes and phages typical of livestock [50] , and enterotoxigenic Escherichia coli O86 causing septicaemia, were potentially transmitted by consumption of carcasses of infected poultry and other livestock [22, 27, 28] . In addition, we found that the IBD virus infected most individuals alone or together with other pathogens also potentially acquired from livestock carrion. This virus causes a highly contagious immunosuppressive bursal disease in poultry [51] and may be transmitted to wildlife in contact with poultry waste or by ingestion of carcasses [22, 52] . Nestlings are especially susceptible to IBD because of the primary role of bursa of Fabricius in immune function development at this age. In fact, immunosuppression due to IBD was indicated by the inflammation, necrosis and loss of lymphocytes in the bursa of Fabricius together with the presence of viral antigens recorded by means of immunocytochemical procedures. The potential impact of highly pathogenic and contagious poultry viruses has been previously recognized as a threat to wildlife health due to the increasing contact of wildlife with livestock operations in general, and poultry farms and their residues in particular, in natural areas worldwide [52] [53] [54] [55] . However, damage from IBD virus on the bursa of Fabricius represents, to our knowledge, the first evidence of clinical disease compatible with death caused by this poultry virus in wildlife. The presence of IBD has been not previously recorded in embryos of wild birds, probably because vertical transmission has been ruled out in poultry and, as consequence, it has probably not been evaluated in other species until now. This striking and concerning result could be related to the longer egg development and incubation periods of bearded vultures compared with poultry, and/or due to contrasting environmental conditions during incubation between bearded vultures and poultry. Thus, embryo infection with IBD may occurs via the female or during incubation as a consequence of egg contact between the egg and poultry remains in the nests of bearded vultures, which requires more research.
Despite their potential effects on population dynamics and conservation through a reduction of productivity and changes in mating behaviour [13, 14] , habitat saturation processes were apparently not directly related to particular proximate causes of egg and nestling failure in this study or in these sampled individuals. As an alternative non-mutually exclusive explanation, we suggest that the recent decline in productivity could also be linked to the increasing ingestion of veterinary drugs and acquisition of pathogens from medicated stabled livestock carcasses due to decreasing availability of unstabled livestock carcasses -the traditional primary food of bearded vultures [16] since the BSE crisis [21] , accompanied by a possible increasing use of antibiotics in stabled livestock operations. In this sense, it is remarkable that bearded vultures primarily feed upon livestock bones, which are one of the major target tissues of fluoroquinolones in medicated animals [56] , therefore, rendering this species especially sensitive to the consequences of an increase in the consumption of stabled intensively medicated livestock. The presence of veterinary drugs in eggs implies their previous presence at least in breeding females [12] , but also probably in breeding males and non-breeders frequently using artificial feeding sites and livestock carcass dumps [17] , where veterinary drugs may be ingested from medicated livestock carcasses [10, 21] . Therefore, further research is required to determine the impact of veterinary drugs and livestock disease on fitness of full-grown individuals, including the potentially subtle, sublethal or indirect effects of these factors on population dynamics.
The link between veterinary drugs and livestock disease should be further investigated in scavenger species, because both threats may concur in food and because the immunodepressive effects and other physiological alterations caused by drugs may facilitate the acquisition and proliferation of pathogens [6, 11, 21] . Given that both threats acting together may greatly contribute to breeding failure decreasing productivity, their potential as stochastic factors with potentially devastating effects increasing the risk of extinction should be not overlooked in current conservation programs of bearded vultures and other scavenger species, especially regarding dangerous veterinary drugs and highly pathogenic viruses frequently infecting poultry. In addition, restricted geographic distribution and low genetic variability [57] common to many threatened species may favour pathogen transmission and reduce the ability of a naïve immune system to fight against novel pathogens [3, 28, 58] , making them especially vulnerable to the potential cross-species transmission of highly virulent virus strains able to cause important outbreaks, as reported in poultry [59] [60] [61] .
The association of pollution and disease may further increase extinction risk if it interacts with the effects of habitat saturation processes [13, 14, 17] . These processes may facilitate conspecific contact and interactions also likely to increase intra-and interspecific pathogen transmission rates in breeding and feeding areas, especially of highly contagious poultry diseases [22] . This could be further enhanced by the artificially high numbers of bearded vultures and other scavengers attracted to feeding points and carcass refuse dumps, both as a result of management and due to the scarcity of unstabled livestock carcasses since the BSE crisis [17, 21] . Whatever the potential contribution of underlying ultimate mechanisms reducing productivity, our findings highlight the need to determine the proximate causes of breeding failure and mortality in wildlife populations in order to understand the processes regulating demography from an ecological framework perspective.
Table S1
Found at: doi:10.1371/journal.pone.0014163.s001 (0.05 MB DOC)
|
Where is the bearded vulture (Gypaetus barbatus) commonly found?
| 5,270 |
the Pyrenees
| 3,127 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What was the focus of this study?
| 5,277 |
the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells
| 856 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What is the third most prevalent cancer in females in the United States?
| 5,278 |
colorectal cancer
| 2,187 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What is the 1-year survival rate for colorectal cancer patients?
| 5,279 |
83.2%
| 2,537 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What is the 5-year survival rate for colorectal cancer patients?
| 5,280 |
64.3%
| 2,547 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
How were nuclear morphological changes in HT-29 cells measured?
| 5,281 |
detection of nuclear condensation
| 16,150 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What is directly related to nuclear condensation?
| 5,282 |
apoptotic chromatin changes
| 16,300 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What morphological cell changes are most associated with apoptosis?
| 5,283 |
membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation
| 18,570 |
1,607 |
A Schiff Base-Derived Copper (II) Complex Is a Potent Inducer of Apoptosis in Colon Cancer Cells by Activating the Intrinsic Pathway
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967396/
SHA: f1f24521928f5d8565a15a17bd7f79239a3d4116
Authors: Hajrezaie, Maryam; Paydar, Mohammadjavad; Zorofchian Moghadamtousi, Soheil; Hassandarvish, Pouya; Gwaram, Nura Suleiman; Zahedifard, Maryam; Rouhollahi, Elham; Karimian, Hamed; Looi, Chung Yeng; Ali, Hapipah Mohd; Abdul Majid, Nazia; Abdulla, Mahmood Ameen
Date: 2014-03-05
DOI: 10.1155/2014/540463
License: cc-by
Abstract: Metal-based drugs with extensive clinical applications hold great promise for the development of cancer chemotherapeutic agents. In the last few decades, Schiff bases and their complexes have become well known for their extensive biological potential. In the present study, we examined the antiproliferative effect of a copper (II) complex on HT-29 colon cancer cells. The Cu(BrHAP)(2 ) Schiff base compound demonstrated a potent antiproliferative effect in HT-29 cells, with an IC(50 )value of 2.87 μg/ml after 72 h of treatment. HT-29 cells treated with Cu (II) complexes underwent apoptosis death, as exhibited by a progressive elevation in the proportion of the G(1 ) cell population. At a concentration of 6.25 μg/ml, the Cu(BrHAP)(2 ) compound caused significant elevation in ROS production following perturbation of mitochondrial membrane potential and cytochrome c release, as assessed by the measurement of fluorescence intensity in stained cells. Furthermore, the activation of caspases 3/7 and 9 was part of the Cu (II) complex-induced apoptosis, which confirmed the involvement of mitochondrial-mediated apoptosis. Meanwhile, there was no significant activation of caspase-8. Taken together, these results imply that the Cu(BrHAP)(2 ) compound is a potential candidate for further in vivo and clinical colon cancer studies to develop novel chemotherapeutic agents derived from metal-based agents.
Text: Cancer is a debilitating disease that afflicts a substantial portion of the world population in all generations and is a major health problem of global concern [1] . Among the various types of cancer, colorectal cancer is the second and third most prevalent cancer among males and females in the United States, respectively. In spite of all the considerable progress in protective methods and recent improvements in screening techniques and chemotherapy, the 1-year and 5-year relative survival rates for patients suffering from colorectal cancer are 83.2% and 64.3%, respectively [2] . In addition, due to bitter controversy over optimal methods for early detection, full compliance of patients with screening recommendations remains a major hindrance for diagnosis at the early stages of cancer development. Development of resistance to chemotherapy also represents a critical issue for which simultaneous treatment with various classes of therapeutics to reduce the resistance has yielded some success [3] . Moreover, the numerous side effects of chemotherapeutic drugs on cancer patients, including hair loss, diarrhea, bleeding, and immunosuppression, have made the process 2
The Scientific World Journal of treatment more complicated [4] . The highly regulated programmed cell death process of apoptosis is a matter of great interest in oncology and cancer therapy and represents a common molecular pathway for drug resistance and carcinogenesis [5] .
Maintenance of a constant cell number in the colonic mucosa is highly regulated through the balance between apoptosis and cell proliferation. The perturbation in this balance leads to an escape from normal cell number homeostasis and is associated with the progression of cancer cells [6, 7] . Thus, suppression of proliferation and elevation of apoptosis in these aberrant cells are suggested to be the essential mechanism for the inhibition of colon cancer. Furthermore, apoptosis and the factors involved in its mechanism of action also present a window that can be exploited for the improvement of potential therapeutic agents with high effectiveness and less adverse side effects [8] . Hence, screening for novel compounds capable of inducing apoptosis in colon cancer cells that can be used alone or in combination with other chemotherapeutic drugs is a significant need and represents a critical challenge in medicinal chemistry.
Metal complexes have been extensively utilized in clinics for centuries and have attracted numerous inorganic chemists to analyze them, with the main focus being medical applications [9, 10] . Copper, an essential trace element with an oxidative nature and bioessential activity in human metabolism, does not exist in an ionic form in biological systems. Thus, measurement of copper in the body is evaluated in the form of complexes with organic compounds [11] . Schiff bases are a critical class of compounds in medical chemistry that have demonstrated significant chemotherapeutic and antibacterial application [12, 13] . Schiff base Cu(II) complexes revealed great potential for antiproliferative, antibacterial, and gastroprotective activity [14] [15] [16] [17] [18] . This study evaluated the anticancer potential of a copper (II) complex derived from N,N -dimethyl ethylene diamine and 2-hydroxyacetophenone Schiff base ligand, Cu(BrHAP) 2 . Furthermore, the possible apoptotic mechanism underlying this activity was also examined. Dulbecco's Modified Eagle Medium (DMEM, Life Technologies, Inc., Rockville, MD) containing 10% fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin G at 37 ∘ C in a humidified atmosphere of 5% CO 2 /95% air. The cells were plated at a fitting density in tissue culture flasks (Corning, USA) according to each experimental scale. Cell viability was measured by a conventional MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide] reduction assay. After 48 h exposure to six concentrations of Cu(BrHAP) 2 , cells were treated with MTT solution (2 mg/mL) for 2 h. The dark formazan crystals formed in intact cells were dissolved in DMSO, and the absorbance was measured at 570 nm and 650 nm as a background using a microplate reader (Hidex, Turku, Finland). The IC 50 value was determined as the concentration of Cu(BrHAP) 2 required to reduce the absorbance of treated cells to 50% of the DMSO-treated control cells. All samples were prepared in triplicates.
Assay. Measurement of lactate dehydrogenase (LDH) release is a biomarker for determining the cytotoxicity of a compound. Briefly, HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 and Triton X-100 (positive control) for 48 h, and the supernatants of the untreated and treated cells were transferred to a new 96-well plate for LDH activity analysis. Next, 100 L of LDH reaction solution was added to each well, the plate was incubated at room temperature for 30 min, and the absorbance was read at 490 nm using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader. The amount of formazan salt and intensity of red color in treated and untreated samples were represented as the LDH activity of cells. The LDH release level in cells treated with Cu(BrHAP) 2 was expressed as a percentage of the positive control.
A propidium iodide (PI) and acridine orange (AO) double staining assay were carried out for detection of apoptosis in the treated cells using a fluorescent microscope (Leica attached with Q-Floro software) according to a standard procedure. HT-29 cells (5 × 10 4 cells/mL in a 25 mL culture flask) were plated, treated with Cu(BrHAP) 2 at the IC 50 concentration, and incubated for 24, 48, and 72 h. After harvesting the cells, they were stained with fluorescent dyes and observed under a UV-fluorescent microscope (Olympus BX51) within 30 min.
In brief, HT-29 cells (1 × 10 4 cells/well in 96-well plate) were supplemented with Cu(BrHAP) 2 (2 g/mL) or DMSO (negative control) for 24 h. The live cells were then incubated with BrdU and Phospho-Histone H3 dyes for 30 min. After the cells were fixed and stained as described by the manufacturer's instructions, they were visualized and analyzed using the Cellomics ArrayScan HCS reader (Thermo Scientific). The fluorescence intensities of the dyes were measured using a target activation bioapplication module.
To confirm the result of the fluorescence cell cycle analysis, HT-29 cells (5 × 10 4 cells/mL) were treated with Cu(BrHAP) 2 for 24, 48, and 72 h for flow cytometry analysis. After incubation, HT-29 cells were spun down at 1800 rpm for 5 min. Next, fixation of a cell population for flow cytometry analysis was carried out to restore integrity. In brief, the cell pellets were fixed by mixing them with 700 L of cold ethanol (90%) and were then kept at 4 ∘ C overnight. Treated HT-29 cells were spun down, and the ethanol was discarded. After washing and suspending the cells in PBS, 25 L of RNase A (10 mg/mL) and 50 L of propidium iodide (PI) (1 mg/mL) were added to the fixed cells for 1 h at 37 ∘ C. The added RNase A limited the ability of PI to bind to only DNA molecules. At the end, the DNA content of the cells was analyzed by a flow cytometer (BD FACSCanto II).
The oxygen radical antioxidant capacity (ORAC) assay was carried out based on the protocols described in detail previously [19] . In brief, Cu(BrHAP) 2 at the concentration of 100 g/mL was used for this assay in a total reaction volume of 200 L. The experiment was performed in a black 96-well microplate with 25 L of compound, blank (solvent/PBS), standard (trolox), or positive control (quercetin). The plate was then supplemented with the working fluorescein solution (150 L), followed by a 5 min incubation at 37 ∘ . The total volume of 200 L was made up by adding 25 L of AAPH working solution. Fluorescence intensity was measured at an excitation wavelength of 485 nm and an emission wavelength of 538 nm every 2 min for 2 h. The result was quantified by calculating the differences of area under the fluorescence decay curve (AUC) of samples and blank. The values were Trolox equivalents (TE).
In brief, HT-29 cells (1 × 10 4 cells/mL) were seeded in 96-well plates and treated with different concentrations of Cu(BrHAP) 2 and DMSO (negative control) for 24 h. After 30 min treatment with dihydroethidium (DHE) dye, cells were fixed and washed with wash buffer as described by the manufacturer's instructions. In the presence of superoxides, DHE dye is oxidized to ethidium. The fluorescence intensity was determined by a fluorescent plate reader at an extension wavelength of 520 nm and an emission wavelength of 620 nm.
The critical factors for monitoring the cell health, namely, cell loss, changes in cell permeability, cytochrome release, mitochondrial membrane potential changes, nuclear size, and morphological changes, were studied using a Cellomics Multiparameter Cytotoxicity 3 Kit as described in detail previously [20] . Plates with stained cells were analyzed using the ArrayScan HCS system (Cellomics, PA, USA).
Caspases 3/7, -8, and 9 activities were determined using the commercial caspase-Glo 3/7, 8, and 9 assay kit (Promega, Madison, WI). HT-29 cells (1.0 × 10 4 cells/well) were seeded overnight in white-walled 96-well plates and treated with different concentrations of Cu(BrHAP) 2 for 24 h. According to the manufacturer's protocol, the treated cells were supplemented with caspase-Glo reagent (100 L) and incubated at room temperature for 30 min. The active caspases from apoptotic cells caused the cleavage of aminoluciferin-labeled synthetic tetrapeptide, leading to the release of substrate for the luciferase enzyme. Caspase activities were analyzed using a Tecan Infinite 200 Pro (Tecan, Männedorf, Switzerland) microplate reader.
In brief, HT-29 cells (1.0 × 10 4 cells/well in a 96-well plate) were treated with different concentrations of Cu(BrHAP) 2 for 3 h, followed by stimulation with TNF-(1 ng/mL) for 30 min. After discarding the medium, cells were fixed and stained using a Cellomics nucleus factor-B (NF-B) activation kit (Thermo Scientific) according to the manufacturer's instructions. Next, an Array Scan HCS Reader was used for evaluation of the plate. Cytoplasmic and nuclear NF-B intensity ratios were calculated using Cytoplasm to Nucleus Translocation Bioapplication software. The average intensity of 200 cells/well was determined. The ratios for untreated, treated, and TNF-stimulated cells were compared.
All the experiments were performed at least three times independently. The results were presented as the mean ± standard deviation (SD) of the number of experiments shown in the legends. An analysis of variance (ANOVA) was carried out using the prism statistical package (GraphPad Software, USA). < 0.05 was considered statistically significant.
Cells of the Colon. Initially, the cytotoxicity of Cu(BrHAP) 2 was tested on HT-29 and CCD 841 cell lines. The IC 50 values of the Schiff base compound were determined based on the result collected from three independent MTT experiments. As indicated in Table 1 , Cu(BrHAP) 2 elicited a significant cytotoxicity and cell inhibitory effect after 24, 48, and 72 h of treatment on HT-29 cell. 2 -Induced LDH Release. Lactate dehydrogenase (LDH) release in the medium is a marker that shows the loss of membrane integrity, apoptosis, or necrosis. The cytotoxicity of the Cu(BrHAP) 2 compound, as determined by the LDH release assay, was quantified on HT-29 cells treated with various concentrations of the Schiff base compound for 48 h. Cu(BrHAP) 2 induced a significant elevation in LDH release, demonstrating cytotoxicity at the 6.25 and 12.5 g/mL concentrations compared to the control cells ( Figure 2 ).
Microscopy and AO/PI Double Staining. Morphological changes in HT-29 cells treated with Cu(BrHAP) 2 compound were observed under a fluorescent microscope at 24, 48, and 72 h. The cells were scored under a fluorescent microscope to analyze viable cells, early apoptosis, and late apoptosis. Early apoptosis, defined as intervening AO within the fragmented DNA, was observed under bright green fluorescence. At the same time, control cells were visualized with a green intact nuclear structure. After 24 and 48 h of treatment with Cu(BrHAP) 2 , moderate apoptosis was observed in the form of blebbing and nuclear chromatin condensation. Furthermore, in the late stage of apoptosis, changes, such as the presence of a reddish-orange color due to binding of PI to denatured DNA, were observed after 72 h of treatment ( Figure 3) . The results showed that the Cu(BrHAP) 2 compound induced morphological features of apoptosis in a time-dependent manner. Figure 4 , demonstrated that there is no cell cycle arrest in the S/M phases. The lack of cell cycle arrest in the S/M phases suggested possible cell cycle arrest in the G 1 /G 2 phases. To determine the exact arrested phase, treated HT-29 cells were analyzed for cell cycle progression using flow cytometry. As expected, there was no significant arrest in the S/M phases. Meanwhile, significant cell cycle arrest in the G 1 phase was observed for HT-29 cells after 24 and 48 h of treatment ( Figure 5 ).
Assay. Antioxidant capacity was measured by ORAC assay, which is the only assay that involves the use of peroxyl radical as a prooxidant and quantifies activity via the area under the curve (AUC) technique. In our experiment, quercetin was used as a positive control. The result demonstrated that Cu(BrHAP) 2 exhibited low to moderate antioxidant activity compared to quercetin ( Table 2) .
Formation. HT-29 cells were treated with different concentrations of Cu(BrHAP) 2 for 24 h and stained with DHE dye to determine the influence of the Schiff base compound on ROS production. The fluorescence intensities of DHE oxidization by ROS were quantified using a fluorescence microplate reader. As depicted in Figure 6 , exposure to the Schiff base compound caused a significant elevation in the ROS levels of treated HT-29 cells at the 6.25 g/mL concentration.
To investigate the induction of apoptosis by Cu(BrHAP) 2 , nuclear morphological changes in HT-29 cells were analyzed by detection of nuclear condensation. As shown in Figure 7 , Hoechst 33342 staining demonstrated that nuclear condensation, which is directly related to apoptotic chromatin changes, emerged in some cells after treatment with Cu(BrHAP) 2 . Meanwhile, the permeability of treated cells was also elevated. Mitochondria are the main source for the production of ROS and adenosine triphosphate (ATP) and are critical in controlling the death and survival of cells. The reduction in fluorescence intensity depicted in Figure 6 Cu(BrHAP) 2 triggered the translocation of cytochrome from mitochondria into the cytosol during apoptosis in HT-29 cells.
Activation. The elevation in ROS production associated with a collapse in MMP may lead to the activation of the caspase cascade. To investigate caspase activation, the bioluminescent intensities representing caspases 3/7, 8, and 9 activities were quantified in HT-29 cells treated with different concentrations of Cu(BrHAP) 2 for 24 h. As shown in Figure 8 , significant elevation in the activity of caspase-3/7 at the 6.25 g/mL concentration and caspase-9 at the 6.25 and 12.5 g/mL concentrations was observed in Cu(BrHAP) 2treated cells, while no significant change in the activity of caspase-8 was detected between treated and untreated HT-29 cells. Thus, the apoptosis induced by the Schiff base compound in HT-29 cells is possibly mediated via the intrinsic pathway, but not the extrinsic pathway.
is a transcription factor that has a critical role in cytokine gene expression. NF-B activation and translocation to the nucleus to enable DNA-binding activity and facilitate target gene expression are mediated by inflammatory cytokines such as tumor necrosis factor-(TNF-). The Cu(BrHAP) 2 Schiff base compound did not exhibit any inhibitory effect on translocation of TNF--stimulated NF-B in HT-29 treated cells, and TNF--stimulation led to NF-B translocation from the cytoplasm to the nucleus (Figure 9 ).
Carcinogenesis is a multistage process in which unregulated cell proliferation as well as a reduction in apoptosis incidence serves as initial characterizations for its progression [21] . One of the defense procedures in multicellular organisms is the destruction of undesirable cell development, which is defined as programmed cell death. Apoptosis is the most noticed programmed cell death mechanism and is characterized by distinct morphological changes such as membrane permeability, cell shrinkage, disruption of the mitochondrial membrane, and chromatin condensation [22, 23]. The disruption of cellular homeostasis between cell death and cell proliferation leads to cancer incidence [24] , and agents that can induce apoptosis are known to have potential anticancer effects [25, 26] . Apoptosis pathways are effective targets for cancer therapy as well as chemoprevention. Numerous chemopreventive drugs have been determined to regulate key events or molecules in apoptosis-inducing signal transduction pathways [27] . In the present study, the Cu(BrHAP) 2 Schiff base compound was evaluated for its ability to inhibit the growth of HT-29 cells using an MTT assay. HT-29 cells have recently been characterized as a suitable model for colon cancer studies [28] [29] [30] . human colon cancer cells in a time-and dose-dependent manner. Meanwhile, the nontumorigenic colon cell line (CCD 841) showed no cytotoxicity after treatment with the compound. The cytotoxic effect of the Cu(II) compound was also confirmed by measuring the level of LDH release from treated cells. Considerably elevated LDH release showed that the cytotoxicity of the Cu(BrHAP) 2 compound possibly occurred via the loss of membrane integrity, whether through activation of apoptosis or the necrosis pathway [31] . The observation of early apoptosis and late apoptosis by fluorescent microscopy analysis and AO/PI double staining following treatment of HT-29 cells with the compound included some signs of apoptosis, namely, cytoplasmic shrinkage, membrane blebbing, and DNA fragmentation [32, 33] . We found that the number of cells with early apoptosis features was higher at earlier stages of treatment. However, when treatment time increased to 72 h, late apoptosis or necrosis characterizations were dominant among treated HT-29 cells. Concurrent detection of late apoptosis or necrosis is scientifically possible because treated HT-29 cells undergoing apoptosis may have progressed into necrosis due to the prolonged incubation with the Schiff base compound.
To elucidate the mechanisms underlying the observed antiproliferative effect of the Cu(II) complex on cancer cells, cell cycle distribution was analyzed using BrdU and Phospho-Histone H3 staining along with flow cytometry [34] [35] [36] . BrdU dye can attach to the synthesized DNA of replicating cells during the S phase of the cell cycle, while Phospho-Histone H3 dye stains cells in different mitotic stages. The cell cycle results from the BrdU and Phospho-Histone H3 double staining assay indicated that there were no significant changes in the number of cells in the S/M phases after the exposure of HT-29 cells to the Schiff base compound. This result suggests the possibility that the cells were arrested in the G 1 or G 2 phase of the cell cycle. Thus, the flow cytometry analysis of the cell cycle was performed to determine the exact arrested phase, and the results demonstrated significant cell cycle arrest at G 1 after 24 and 48 h of treatment, suggesting proliferative suppression via induction of apoptosis [37, 38] .
Perturbation of mitochondrial membrane potential is one of the earliest intracellular events that occur following the induction of apoptosis [39] . As the main source of cellular ROS and adenosine triphosphate (ATP), mitochondria are the key regulators of mechanisms controlling the survival or death of cells. After confirming that the Cu(BrHAP) 2 Schiff base compound did not have significant antioxidant capacity in HT-29 cancer cells using the ORAC assay, the induction of ROS production in treated cells was analyzed. According to our study, after exposing the Cu(II) compound to HT-29 cells and analyzing the levels of ROS, it was demonstrated that the level of ROS in treated HT-29 cells was significantly elevated at a compound concentration of 6.25 g/mL.
In metal-induced apoptosis, the mitochondria have the crucial role in mediating apoptosis through metal-induced ROS [40] . The intrinsic or mitochondrial-dependent signaling pathway involves different factors of nonreceptor-mediated stimuli that induce intracellular signals. These signals, mainly through the p53 protein, act on the mitochondrialinitiated events. Excessive ROS production is a negative signal that can result in the failure of suppression of antiapoptotic factors, thereby triggering apoptosis. Therefore, we used mitochondrial membrane potential (MMP) fluorescent probes to examine the effect of elevated ROS production on the function of mitochondria in treated HT-29 cells. As shown in Figure 7 , changes in MMP after treatment with the Cu(BrHAP) 2 Schiff base compound leading to the membrane depolarization of the mitochondria were demonstrated by Rhodamine 123 release to the cytoplasm from the mitochondria matrix. The result implies that the induction of apoptosis by Cu(II) Schiff base complexes may be associated with the mitochondrial pathway [26, 41, 42] . One of the important signals to initiate the procedure of apoptosis is cytosolic cytochrome . The release of cytochrome into the cytosol and reduction of its levels in the mitochondria have been shown to occur as a result of changes in MMP [30] . As the result illustrated, the synthetic Schiff base compound also led to an increase in the level of cytochrome in the cytosol compared to the control.
The excessive production of ROS from mitochondria and the collapse of MMP may activate the downstream caspase molecules and consequently lead to apoptotic cell death. After the binding of cytochrome to apoptotic activating factor-1, caspase-9 is activated via apoptosome formation, which leads to active caspase-3/7, the most effective caspase with many cellular targets [43] . In the extrinsic pathway, apoptosis is mediated by death receptors. As an example, FAS ligand interacts with the FAS receptor, leading to the activation of caspase-8 [44] . Caspase-8 activation cleaves and activates downstream executioner caspases such as caspase-3/7 [45, 46] . In our study, the Cu(BrHAP) 2 schiff base compound induced significant elevation in the caspases 3/7 and 9 activities compared to the control. Meanwhile, there was no activation of caspase-8, suggesting that the apoptosis induced in HT-29 cells was mediated via the intrinsic mitochondrial pathway but not the extrinsic, death receptor-linked caspase-8 pathway.
The supporting evidence of LDH release, ROS production, MMP suppression, elevation in the level of cytochrome , and activation of caspases 3/7 and 9 demonstrated the promising anticancer activity of the Cu(BrHAP) 2 Schiff base compound against the HT-29 colon cancer cell line via the intrinsic mitochondrial pathway.
|
What types of cells are suitable for colon cancer studies?
| 5,284 |
HT-29 cells
| 19,277 |
1,608 |
Multivalent HA DNA Vaccination Protects against Highly Pathogenic H5N1 Avian Influenza Infection in Chickens and Mice
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2657001/
SHA: ef872b80cf38917f64c42bfa52a57beb4399897a
Authors: Rao, Srinivas; Kong, Wing-Pui; Wei, Chih-Jen; Yang, Zhi-Yong; Nason, Martha; Styles, Darrel; DeTolla, Louis J.; Sorrell, Erin M.; Song, Haichen; Wan, Hongquan; Ramirez-Nieto, Gloria C.; Perez, Daniel; Nabel, Gary J.
Date: 2008-06-18
DOI: 10.1371/journal.pone.0002432
License: cc0
Abstract: BACKGROUND: Sustained outbreaks of highly pathogenic avian influenza (HPAI) H5N1 in avian species increase the risk of reassortment and adaptation to humans. The ability to contain its spread in chickens would reduce this threat and help maintain the capacity for egg-based vaccine production. While vaccines offer the potential to control avian disease, a major concern of current vaccines is their potency and inability to protect against evolving avian influenza viruses. METHODOLOGY / PRINCIPAL FINDINGS: The ability of DNA vaccines encoding hemagglutinin (HA) proteins from different HPAI H5N1 serotypes was evaluated for its ability to elicit neutralizing antibodies and to protect against homologous and heterologous HPAI H5N1 strain challenge in mice and chickens after DNA immunization by needle and syringe or with a pressure injection device. These vaccines elicited antibodies that neutralized multiple strains of HPAI H5N1 when given in combinations containing up to 10 HAs. The response was dose-dependent, and breadth was determined by the choice of the influenza virus HA in the vaccine. Monovalent and trivalent HA vaccines were tested first in mice and conferred protection against lethal H5N1 A/Vietnam/1203/2004 challenge 68 weeks after vaccination. In chickens, protection was observed against heterologous strains of HPAI H5N1 after vaccination with a trivalent H5 serotype DNA vaccine with doses as low as 5 µg DNA given twice either by intramuscular needle injection or with a needle-free device. CONCLUSIONS/SIGNIFICANCE: DNA vaccines offer a generic approach to influenza virus immunization applicable to multiple animal species. In addition, the ability to substitute plasmids encoding different strains enables rapid adaptation of the vaccine to newly evolving field isolates.
Text: The highly pathogenic H5N1 influenza virus causes lethal multi-organ disease in poultry, resulting in significant economic losses and a public health concern in many parts of the world. The greatest threats posed by this virus are its ability to cause mortality in humans, its potential to compromise food supplies, and its possible economic impacts. Viral maintenance in poultry potentiates the risk of human-to-human transmission and the emergence of a pandemic strain through reassortment. An effective, safe poultry vaccine that elicits broadly protective immune responses to evolving flu strains would provide a countermeasure to reduce the likelihood of transmission of this virus from domestic birds to humans and simultaneously would protect commercial poultry operations and subsistence farmers.
DNA vaccines have been shown to elicit robust immune responses in various animal species, from mice to nonhuman primates [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] . In human trials, these vaccines elicit cellular and humoral immune responses against various infectious agents, including influenza, SARS, SIV and HIV. In addition to their ability to elicit antibody responses, they also stimulate antigenspecific and sustained T cell responses [1] [2] [3] 6, 12, 13] . DNA vaccination has been used experimentally against various infectious agents in a variety of mammals, including cattle (against infectious bovine rhinotracheitis/bovine diarrhea virus, leptospirosis and mycobacteriosis) [14, 15] , pigs (against classical swine fever virus and mycoplasmosis) [16] , and horses (against West Nile virus and rabies) [17] . In addition, DNA vaccines have been tested against avian plasmodium infection in penguins [18] and against influenza and infectious bursal disease in chickens [7, 8, 19] , duck hepatitis B virus in ducks [6] , and avian metapneumovirus and Chlamydia psittaci in turkeys [20, 21] (reviewed in ref. [22] ). While they have been used in chickens to generate antisera to specific influenza viruses and confer protection against the low pathogenicity H5N2 strain [23] , there is only one previous report of a monovalent DNA vaccine effective against H5N1 (and that only against a matched H5N1 isolate) [24] ; no protection with multivalent DNA vaccines against heterologous strains has been reported.
Development and characterization of a DNA vaccine modality for use in poultry offers a potential countermeasure against HPAI H5N1 avian influenza outbreaks. The virus can infect humans, typically from animal sources, including commercial and wild avian species, livestock, and possibly other non-domesticated animal species [25] [26] [27] . While there is marked diversity in the host range of type A influenza viruses, many experts have speculated that a pandemic strain of type A influenza could evolve in avian species or avian influenza viruses could contribute virulent genes to a pandemic strain through reassortment [28, 29] . Thus, there is reason to consider vaccination of poultry that would stimulate potent and broad protective immune responses [7, 30, 31] . In undertaking such efforts, it is important that there be a differentiation of infected from vaccinated animals [32] so that animals can be protected and permit monitoring of new infections using proven and sensitive methodologies.
In this study, we used an automated high capacity needle-free injection device, Agro-JetH (Medical International Technology, Inc., Denver, CO) to explore the feasibility of DNA vaccination of poultry. After optimization of injection conditions, alternative multivalent DNA vaccine regimens were analyzed and compared for magnitude and breadth of neutralizing antibodies, as well as protective efficacy after challenge in mouse and chicken models of HPAI H5N1 infection. The findings suggest that it is possible to develop a multivalent DNA vaccine for poultry that can protect against multiple HPAI H5N1 strains and that could keep pace with the continued evolution of avian influenza viruses.
Immunogenicity and neutralizing antibody specificity of alternative HA DNA vaccines in mice To evaluate the efficacy of multivalent DNA vaccines, initial studies were performed in mice. Expression vectors encoding HAs from ten phylogenetically diverse strains of influenza viruses [33] were generated by synthesis of cDNAs (see Materials and Methods) in plasmid expression vectors, pCMV/R or pCMV/R 8kB, which mediates high level expression and immunogenicity in vivo [34, 35, 36] . Animals were immunized with each expression vector intramuscularly (IM) at three week intervals, and the antisera were evaluated on day 14 after the third immunization for their ability to neutralize HPAI H5N1 pseudotyped lentiviral vectors as previously described [35, 36] . We have previously shown that lentiviral assay inhibition (LAI) yields similar results to microneutralization and HAI analyses with higher sensitivity in mice [35, 36]
To determine whether immunization with multiple HAs simultaneously could expand the breadth of the neutralizing antibody response without significant loss of magnitude, a combination of 10 HA DNA vaccine immunogens was administered IM at proportionally lower concentration (1.5 mg per immunogen) into groups of 10 mice (see Materials and Methods). Remarkably, despite a log lower DNA concentration of each component, significant neutralizing antibody titers were generated to each of the 10 immunogens, with .80% neutralization against 6 out of 12 H5 HA pseudoviruses at dilutions of up to 1:400 ( Fig. 2A) .
To evaluate whether similar breadth of immunity could be generated with fewer immunogens, two different combinations of 5 immunogens were selected, based on the phylogenetic diversity of HA among the avian influenza viruses [33] and the crossreactivity of the neutralizing antibody responses of select individual immunogens (Fig. 1 ). As expected, there were substantial differences in the breadth of neutralization between these two sets of 5 immunogen multivalent vaccines (Fig. 2 , B vs. C). In one set, while neutralization of homologous strains was comparable to the monovalent and the 10 immunogen multivalent immune response, fewer cross-reactive antibodies were detected, directed most prominently against A/Iraq Protection of DNA-vaccinated mice against challenge with heterologous H5N1 A/Vietnam/1203/2004 influenza virus Mice immunized as described above were challenged with a heterologous H5N1 virus 68 weeks after the final DNA vaccination. Animals were then challenged with 10 LD 50 of the highly pathogenic A/Vietnam/1203/2004 virus intranasally, and morbidity and mortality were monitored for 21 days after the viral challenge. The control animals, injected with the plasmid expression vector with no insert, died within 10 days of infection. Complete survival was observed in the groups immunized with the 10 component and set 2 of the 5 component multivalent DNA vaccines (Fig. 3) . Immunization with HA derived from the A/ Indonesia/05/2005 strain or set 1 of the 5 component multivalent DNA vaccine showed a survival rate approaching 90%. In contrast, animals injected with HA plasmid DNA derived from A/ Anhui/1/2005, which has diverged more from A/Vietnam/ 1203/2004, showed a lower percent survival (70%) after lethal viral challenge. Survival differences between groups were assessed using a log-rank test and the Gehan-Wilcoxon test on the survival curves for pairs of groups. A test was deemed significant if the pvalue was ,0.01. Mice injected IM with different HAs, A/ Indonesia/5/05, A/Anhui/1/05, 10HA, 5 HA (Set 1), or 5 HA (Set 2) showed a significant difference compared to control (all p values,0.001). Among the HA-immunized groups, there was no significant difference between any two groups (p.0.08 for all comparisons). For example, no significant difference was observed between the A/Anhui/1/05 group, which had the least survival among the HA immunized groups (7 out of 10), and other HA groups: A/Indonesia/5/05 (p = 0.377), 10 HA (p = 0.082), 5 HA (Set 1) (p = 0.101), or 5 HA (Set 2) (p = .411). Therefore, we cannot exclude the possibility that the 3 deaths in the A/Anhui/1/05 group may have been due to random chance.
Since it is desirable to confer protective immunity in poultry and HA DNA vaccination was effective in mice, we next examined the breadth and potency of single or multiple HA plasmid immunization in chickens. The ability of chickens to generate specific antibodies was assessed with three strains that showed broad cross protection in mouse studies (A/Vietnam/1203/2004, A/Anhui/ 1/2005 and A/Indonesia/05/2005), administered individually or in combination, by different injection methods. In addition to needle injection, a needle-free repetitive injection device, Agro-JetH (Medical International Technology, Inc., Denver, CO), was analyzed. This device disperses the 0.1 to 5 ml injection doses into the dermal, subcutaneous, or intramuscular tissue depending upon the pressure adjustments, powered by a CO 2 gas pressure plunger [39] . The injection conditions were determined by histologic analysis of tissues that received injections of India ink; a pressure of 48 psi was chosen since it enabled consistent delivery into intradermal and subcutaneous tissues (Fig. S1 ).
Immunization of chickens with the control plasmid (CMV/R) without an HA gene insert elicited minimal neutralizing antibody titers compared to HA-immunized animals 1 week after 3 DNA immunizations. Nearly all chickens immunized with either monovalent or multivalent HA DNA vaccines generated significant neutralization titers ( Fig. 4 and Table S1 ). In general, there was a progressive increase in the amount of neutralization after each successive DNA vaccination (data not shown) with maximal response at 1 week after the 3 rd DNA immunization, with highest and most consistent levels in the trivalent vaccine group delivered with the Agro-JetH device. Neutralization of the Indonesia HA strain was the most robust, with neutralization nearing 100% at titers greater than 1:3200. Both the monovalent and multivalent vaccines elicited robust homologous ( vaccine (Fig. 4 ). Even though one chicken (238) in the multivalent vaccine group produced almost the same degree of neutralization at each time point and was protected, it did not produce a high neutralizing antibody titer for reasons that were uncertain but possibly related to a non-specific inhibitor in the sera.
To determine whether chickens immunized with single or multiple DNA vaccines were protected from a lethal challenge of a heterologous HPAI H5N1 virus, vaccinated chickens were In panels B and C, mice (n = 10) were immunized with 15 mg of plasmid (3 mg each) three times at 3 week intervals. Serum pools from the immunized animals were collected 14 days after the third immunization. The antisera were tested against the 12 indicated pseudotyped lentiviral vectors at varying dilutions. Error bars at each point indicate the standard deviation; each sample was evaluated in triplicate. In general, the immunized serum neutralized all tested pseudotyped lentiviruses at low dilutions while differences were often observed at high dilution. doi:10.1371/journal.pone.0002432.g002 inoculated with 20 LD 50 of highly pathogenic A/Vietnam/1203/ 2004 heterologous virus intranasally using standard methods [25, 40] and monitored for morbidity, mortality, viral shedding and serum antibodies. While all the control animals died within 2 days of infection, 100% survival was noted in the rest of the chickens (Fig. 5A ). The animals that were healthy, showing no signs of clinical disease or malaise, were euthanized on day 14. There was no evidence for viral shedding monitored via tracheal and cloacal swabs of infected chickens 2-14 days after challenge as determined by embryonal inoculation (data not shown: egg infectious dose 50 (EID 50 ) limit of detection ,100 virus particles).
To compare the relative efficacy of DNA vaccines delivered IM by needle and syringe versus the needle-free Agro-JetH device injection, a dose-response study was performed with amounts of DNA vaccine ranging from 500 to 0.5 mg with two inoculations. In these experiments, the HA derived from A/chicken/Nigeria/641/ 2006 was substituted for A/Vietnam/1203/2004 since it represented a more contemporary isolate. The observed rate of protection was higher among the animals receiving 5 mg by Agro-Jet (8/8) than by IM injection (6/8) (Fig. 5, B vs. C). Both modes provided complete protection for all animals at doses higher than this, and 25% protection for the animals receiving 0.5 mg doses (Fig. 5B, C) . Survival differences between consecutive doses were assessed using a log-rank test on the survival curves for pairs of groups. A test was deemed significant if the p-value was ,.01, and marginally significant if the p-value was ,.05 but ..01. Chickens injected IM showed a marginally significant difference between 0.5 and 5 mg (p = .047). In the same group there was a significant difference between control and 5, 50 and 500 mg (p,.001 for all comparisons) and the difference between control and 0.5 mg was marginally significant (p = .016). Chickens that were injected using Agro-JetH showed a significant difference between 0.5 and 5 mg (p = .004) and between control and 5, 50, and 500 mg (p,.001 for all comparisons). There were no differences between control and 0.5 mg or between 5, 50, and 500 mg. Lastly, the survival differences between Agro-JetH and IM for each dose group were not significant. The neutralizing antibody response to homologous and heterologous HAs corresponded with protection and correlated with dose, with higher titers elicited by injection with Agro-JetH compared to needle (Table S2) . We assessed viable viral shedding after inoculation by chick embryo inoculation three days after virus challenge (Week 8). While we noted some embryonic lethality at the 0.5 mg dose, there was no embryonic lethality at 5, 50 or 500 mg groups (data not shown).
Since the HPAI H5N1 virus first appeared ten years ago, this highly pathogenic avian influenza virus has shown increasing diversification and dissemination in Asia, Africa, and Europe [28, [41] [42] [43] [44] . In addition to its effects on human health by crossspecies transmission [28, 45, 46] and ability to compromise food sources, it poses a continuing threat to public health as it evolves and adapts in different species. The pandemic potential of this virus, especially as it relates to the poultry industry and for reservoir avian hosts, underscores the need for a vaccine that offers broad spectrum immunity and protection against lethal viral challenge. While the virus remains restricted in its ability to infect humans and undergo efficient human-to-human transmission [28, 47] , its persistence and spread in poultry increases the risk of the emergence of a pandemic strain. One approach to pandemic risk reduction is to limit the propagation of the virus in poultry and other relevant avian species.
We have previously reported that DNA vaccines encoding HA can confer protection against a highly lethal human pandemic influenza virus, the 1918 H1N1 virus, in mice [36] . DNA vaccines offer several advantages, including the ability to express diverse antigens, tolerability in various hosts, ease of delivery, and stability for storage and distribution without the necessity of maintaining a cold chain; they have been shown to be safe and efficacious in a variety of animal models [2, 4, 12, 22, 48] . Because they do not contain other viral proteins used to screen for infection, they also address the need to differentiate vaccinated from infected animals. There is evidence that DNA vaccination elicits cell-mediated immunity against influenza HA in addition to inducing an antibody response [36] , an effect that could significantly contribute to protective immunity as viruses show genetic drift and reduced susceptibility to neutralization.
Ideally, a highly effective influenza vaccine should not only be able to let the host develop a protective immune response against a matching live virus challenge but also elicit robust protective immune responses against a broad range of homologous and heterologous H5 influenza strains. A multivalent H5 vaccine containing diverse serotypes could expand the antigenic breadth sufficiently to provide protection against heterologous challenge and may preclude the emergence of vaccine-resistant strains that may arise due to evolutionary vaccine pressure on the virus. Due to the antigenic drift and shift of the influenza virus genome, it has been very difficult to predict the next dominant strain of an avian endemic outbreak. DNA vaccines can be synthesized in a relatively short period of time, and the targeted mutations can be tailored to specific viral serotypes. The mutations promote a focused and enhanced immune response [3, 49, 50] that may be particularly important in the event of an outbreak where specificity is the key to epidemic control. The use of modified codons ensures maximal expression in the host and eliminates the possibility of recombination with influenza viruses that might potentially generate new strains.
A more broadly protective murine vaccine was developed here by including more HAs from varying strains in the multivalent vaccine (Figs. 2 and 3) . However, it is less practical to include large numbers of different HAs in one vaccine due to the cost and complexity of manufacturing such a vaccine. Therefore, we simplified the vaccine regimen based on cross-neutralization studies and phylogenetic relationships. A trivalent vaccine was subsequently identified for further studies. Due [51] .
While three DNA immunizations were used initially to demonstrate protective immunity and have been used previously to elicit protection in mice [36] , we found that effective protective immunity could be induced with two DNA vaccinations and as little as 5 mg trivalent DNA immunization using the ID/SC route with the Agro-JetH device. In addition, based on the chick embryo inoculation data, we believe that there is effective neutralization of the virus and lack of infectious viral shedding in chicken vaccinated with as little as 5 mg of DNA. The device's capacity for rapid repetitive injection and the lower quantity and stability of DNA enhance the practicality and utility of this approach for vaccination of endangered species in captivity or administration to poultry or other animals.
A/Vietnam/1203/2004 (H5N1) (A/VN/1203/04) was obtained from the repository at the Centers for Disease Control and Prevention (CDC), Atlanta, Georgia. The virus was propagated in 10-day old embryonated chicken eggs at 35uC and stored at 270uC until use. The virus was titrated by the Reed and Muench method to determine EID 50 [52] . GenBank ABD28180) were synthesized using human-preferred codons (GeneArt, Regensburg, Germany) [36] . HA cDNAs from diverse strains of influenza viruses were then inserted into plasmid expression vectors, pCMV/R or pCMV/R 8kB, which mediates high level expression and immunogenicity in vivo [34, 35, 36] . For initial trivalent immunizations in chickens, the A/Vietnam/1203/ 2004, A/Anhui/1/2005 and A/Indonesia/05/2005 strains were used and in the dose response study, the Vietnam strain was replaced with A/chicken/Nigeria/641/2006. The immunogens used in DNA vaccination contained a cleavage site mutation (PQRERRRKKRG to PQRETRG) as previously described [35, 36] . This mutation was generated by site-directed mutagenesis using a QuickChange kit (Stratagene, La Jolla, CA).
DNA immunization of mice [6] [7] [8] week old female BALB/c mice were purchased from The Jackson Laboratory and maintained in the AAALAC-accredited Vaccine Research Center Animal Care Facility (Bethesda, MD) under specific pathogen-free conditions. All experiments were approved by the Vaccine Research Center Animal Care and Use Committee. The mice were immunized as previously described [5] . Briefly, mice (10 animals for all test groups, 20 animals for the
The study was carried out in the AAALAC-accredited animal facility at the University of Maryland School of Medicine. Six groups of 8 one-day-old male and female SPAFAS White Leghorn Chickens, Gallus domesticus, were obtained from Charles River Laboratories (Connecticut). The animals were housed in brooder and grower cages (McMurray Hatcheries, Iowa). Feed (Teklad Japanese Quail Diet -3050, Harlan-Teklad, WI) and water were provided to the animals ad libitum. The study was performed in strict accordance with the ''Guide'' after approvals from the Animal Care and Use Committees of the Vaccine Research Center, NIH and the University of Maryland. DNA immunizations were performed as described at 0, 3 and 6 weeks. A total dose of 500 mg of one or a combination of the following DNA plasmids in a volume of 250 ml was administered to each animal: pCMV/ R, pCMV/R-HA
Agro-JetH is a needle-free device used for mass delivery of vaccines and drugs in livestock and poultry. The device is semiautomatic and requires a small CO 2 tank or compressed air for low pressure delivery. Upon trigger activation, CO 2 disperses the injectate at a precise dose into the muscle, dermis or subcutaneous tissue depending on the setting that was standardized for our use. We used an effective volume of 0.1 ml in our injectate [39] . In this study we were able to effectively deliver 0.1 ml of injectate into the animal's dermis/subcutaneous tissue at a pressure of 48-55 psi.
Sixty-eight weeks after the last immunization, female BALB/c mice were lightly anesthetized with Ketamine/Xylazine and inoculated intranasally with 10 LD 50 of A/Vietnam/1203/2004 virus diluted in phosphate-buffered saline in a 50 ul volume. Mice were monitored daily for morbidity and measured for weight loss and mortality for 21 days post infection. Any mouse that had lost more than 25% of its body weight was euthanized. All experiments involving the HPAI virus were conducted in an AAALAC accredited facility (BioQual Inc., Gaithersburg, MD) under BSL 3 conditions that included enhancements required by the USDA and the Select Agent Program.
White Leghorn chickens were challenged one week after the last immunization with 20 lethal dose 50 (LD 50 ) of A/Vietnam/1203/04 (H5N1) influenza A virus, equivalent to 2610 4 EID 50 based on previous challenges [53] . Chickens were infected with 200 ml virus intranasally. Tracheal and cloacal swabs were collected days 3 and 5 post-challenge and stored in glass vials containing BHI medium (BBL TM Brain Heart Infusion, Becton Dickinson) at 280uC. Blood was collected 14 days post-challenge and serum was titered by microneutralization assay. Chickens were observed and scored daily for clinical signs of infection, morbidity and mortality. Chickens that survived the study were bled and humanely euthanized at day 14 post-challenge. Lungs, heart, intestine and kidney were collected and samples were stored in formalin for histopathology. Experiments were carried out under BSL3+ conditions with investigators wearing appropriate protective equipment and compliant with all Institutional Animal Care and Use Committee-approved protocols and under Animal Welfare Act regulations at the University of Maryland, College Park, Maryland.
Representative tracheal and cloacal swabs were chosen to run an EID 50 assay for comparison and virus titers were determine by the method of Reed and Meunch [52] . Briefly, swabs were used to infect 10 day-old embryonated chicken eggs in 10-fold dilutions. Three eggs were inoculated per dilution and incubated for 48 hours before titration.
Neutralizing antibodies were titrated from serum samples collected week 5 and 7 post-vaccination and day 14 post-challenge. The microneutralization assay was performed using a 96-well plate format. Serum was treated with receptor-destroying enzyme (Denka Seiken Co.) and treated at 37uC per the manufacturer's instructions. After an overnight incubation and subsequent inactivation samples were brought to a final dilution of 1:10 using PBS and each sample was serially diluted and virus, diluted to 100 TCID 50 , was added to each well. The plates were then incubated at 37uC, 5% CO 2 for 1-2 hours. Following incubation, supernatants were used to infect a second 96-well plate of MDCK cells. Microplates were incubated at 4uC for 15 minutes and then 37uC, 5% CO 2 for 45 minutes. Supernatants of serum and virus were then discarded and 200 ml of OptiMEM (containing 1X antibiotics/antimycotics, 1 mg/ml TPCK-trypsin) was added and incubated at 37uC, 5% CO 2 for 3 days. After 3 days, 50 ml of the supernatant from each well was transferred into a new 96-well microplate, and an HA assay was performed to calculate the antibody titers. Virus and cell controls were included in the assay.
Two-fold dilutions of heat-inactivated sera were tested in a microneutralization assay as previously described [54] for the presence of antibodies that neutralized the infectivity of 100 TCID 50 (50% tissue culture infectious dose) of the A/Vietnam/ 1203/2004 H5N1 virus on MDCK cell monolayers by using two wells per dilution on a 96-well plate.
The recombinant lentiviral vectors expressing a luciferase reporter gene were produced as previously described [35, 36] . For the neutralization assay, antisera from immunized animals were heat-inactivated at 55uC for 30 minutes and mixed with 50 ml of pseudovirus at various dilutions. The sera/virus mixture was then added to 293A cells in 96-well B&W TC Isoplates (Wallac, Turku, Finland; 12,000 cells/well). Two hours later, the plates were washed and fresh medium was added. Cells were lysed in mammalian cell lysis buffer (Promega, Madison, WI) 24 hrs after infection and luciferase activity was measured using the Luciferase Assay System (Promega, Madison, WI).
The following strains were used for the production of pseudotyped viruses: for HA we used A/Thailand/1(KAN-
The HA/HI titers were determined as previously described [54] . Briefly, HA titers were calculated using 50 ml of 0.5% chicken red blood cell suspension in PBS added to 50 ml of twofold dilutions of virus in PBS. This mix was incubated at room temperature for 30 minutes. The HA titers were calculated as the reciprocal value of the highest dilution that caused complete hemagglutination. HI titers were calculated by titrating 50 ml of antiserum treated with receptor-destroying enzyme and an equivalent amount of A/Vietnam/1203/2004 virus (four hemagglutinating doses) was added to each well. Wells were incubated at room temperature for 30 minutes and 50 ml of a 0.5% suspension of chicken red blood cells was added. HI titers were calculated after 30 minutes as the reciprocal of the serum dilution that inhibited hemagglutination.
Table S1 Hemagglutination inhibition (HI), microneutralization titer (NT), and LAI of sera from individual chickens immunized with different vaccines. Sera from immunized animals were obtained at week 5 or 7, a week before or after the final boost, and neutralization was assessed by HI, microneutralization (NT) and LAI (shown as IC 50 ). Individual animal serum of each group is shown and was analyzed as described in the Materials and Methods section. Figure S1 Characterization of needle-free (Agro-JetH) DNA immunization in chickens. To evaluate the distribution of fluid into superficial or deep layers of subcutaneous tissues after delivery by AgroJetH, 4 or 7 week old chickens were injected with a solution containing India ink with this needle-free device at various pressures, ranging from 45 to 55 mm Hg. Three sites (thigh, wing and breast) were used, and biopsies were taken for routine hematoxylin and eosin staining. Representative sections of thigh injections are shown from 7 week old chickens and were similar at 4 weeks (data not shown). While the 48 mm Hg pressure deposited the injectate into the dermis/subcutaneous region (left), the higher pressure injections, 52 and 58 mm Hg, deposited the injectate into the subcutaneous and muscle layers (middle, right). 48 mm Hg consistently provided an optimal pressure to deposit the injectate into the dermis and subcutaneous tissue and was chosen for all AgroJetH immunizations. Found at: doi:10.1371/journal.pone.0002432.s003 (10.74 MB DOC)
|
What was the goal of this study?
| 5,285 |
to explore the feasibility of DNA vaccination of poultry
| 5,812 |
1,608 |
Multivalent HA DNA Vaccination Protects against Highly Pathogenic H5N1 Avian Influenza Infection in Chickens and Mice
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2657001/
SHA: ef872b80cf38917f64c42bfa52a57beb4399897a
Authors: Rao, Srinivas; Kong, Wing-Pui; Wei, Chih-Jen; Yang, Zhi-Yong; Nason, Martha; Styles, Darrel; DeTolla, Louis J.; Sorrell, Erin M.; Song, Haichen; Wan, Hongquan; Ramirez-Nieto, Gloria C.; Perez, Daniel; Nabel, Gary J.
Date: 2008-06-18
DOI: 10.1371/journal.pone.0002432
License: cc0
Abstract: BACKGROUND: Sustained outbreaks of highly pathogenic avian influenza (HPAI) H5N1 in avian species increase the risk of reassortment and adaptation to humans. The ability to contain its spread in chickens would reduce this threat and help maintain the capacity for egg-based vaccine production. While vaccines offer the potential to control avian disease, a major concern of current vaccines is their potency and inability to protect against evolving avian influenza viruses. METHODOLOGY / PRINCIPAL FINDINGS: The ability of DNA vaccines encoding hemagglutinin (HA) proteins from different HPAI H5N1 serotypes was evaluated for its ability to elicit neutralizing antibodies and to protect against homologous and heterologous HPAI H5N1 strain challenge in mice and chickens after DNA immunization by needle and syringe or with a pressure injection device. These vaccines elicited antibodies that neutralized multiple strains of HPAI H5N1 when given in combinations containing up to 10 HAs. The response was dose-dependent, and breadth was determined by the choice of the influenza virus HA in the vaccine. Monovalent and trivalent HA vaccines were tested first in mice and conferred protection against lethal H5N1 A/Vietnam/1203/2004 challenge 68 weeks after vaccination. In chickens, protection was observed against heterologous strains of HPAI H5N1 after vaccination with a trivalent H5 serotype DNA vaccine with doses as low as 5 µg DNA given twice either by intramuscular needle injection or with a needle-free device. CONCLUSIONS/SIGNIFICANCE: DNA vaccines offer a generic approach to influenza virus immunization applicable to multiple animal species. In addition, the ability to substitute plasmids encoding different strains enables rapid adaptation of the vaccine to newly evolving field isolates.
Text: The highly pathogenic H5N1 influenza virus causes lethal multi-organ disease in poultry, resulting in significant economic losses and a public health concern in many parts of the world. The greatest threats posed by this virus are its ability to cause mortality in humans, its potential to compromise food supplies, and its possible economic impacts. Viral maintenance in poultry potentiates the risk of human-to-human transmission and the emergence of a pandemic strain through reassortment. An effective, safe poultry vaccine that elicits broadly protective immune responses to evolving flu strains would provide a countermeasure to reduce the likelihood of transmission of this virus from domestic birds to humans and simultaneously would protect commercial poultry operations and subsistence farmers.
DNA vaccines have been shown to elicit robust immune responses in various animal species, from mice to nonhuman primates [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] . In human trials, these vaccines elicit cellular and humoral immune responses against various infectious agents, including influenza, SARS, SIV and HIV. In addition to their ability to elicit antibody responses, they also stimulate antigenspecific and sustained T cell responses [1] [2] [3] 6, 12, 13] . DNA vaccination has been used experimentally against various infectious agents in a variety of mammals, including cattle (against infectious bovine rhinotracheitis/bovine diarrhea virus, leptospirosis and mycobacteriosis) [14, 15] , pigs (against classical swine fever virus and mycoplasmosis) [16] , and horses (against West Nile virus and rabies) [17] . In addition, DNA vaccines have been tested against avian plasmodium infection in penguins [18] and against influenza and infectious bursal disease in chickens [7, 8, 19] , duck hepatitis B virus in ducks [6] , and avian metapneumovirus and Chlamydia psittaci in turkeys [20, 21] (reviewed in ref. [22] ). While they have been used in chickens to generate antisera to specific influenza viruses and confer protection against the low pathogenicity H5N2 strain [23] , there is only one previous report of a monovalent DNA vaccine effective against H5N1 (and that only against a matched H5N1 isolate) [24] ; no protection with multivalent DNA vaccines against heterologous strains has been reported.
Development and characterization of a DNA vaccine modality for use in poultry offers a potential countermeasure against HPAI H5N1 avian influenza outbreaks. The virus can infect humans, typically from animal sources, including commercial and wild avian species, livestock, and possibly other non-domesticated animal species [25] [26] [27] . While there is marked diversity in the host range of type A influenza viruses, many experts have speculated that a pandemic strain of type A influenza could evolve in avian species or avian influenza viruses could contribute virulent genes to a pandemic strain through reassortment [28, 29] . Thus, there is reason to consider vaccination of poultry that would stimulate potent and broad protective immune responses [7, 30, 31] . In undertaking such efforts, it is important that there be a differentiation of infected from vaccinated animals [32] so that animals can be protected and permit monitoring of new infections using proven and sensitive methodologies.
In this study, we used an automated high capacity needle-free injection device, Agro-JetH (Medical International Technology, Inc., Denver, CO) to explore the feasibility of DNA vaccination of poultry. After optimization of injection conditions, alternative multivalent DNA vaccine regimens were analyzed and compared for magnitude and breadth of neutralizing antibodies, as well as protective efficacy after challenge in mouse and chicken models of HPAI H5N1 infection. The findings suggest that it is possible to develop a multivalent DNA vaccine for poultry that can protect against multiple HPAI H5N1 strains and that could keep pace with the continued evolution of avian influenza viruses.
Immunogenicity and neutralizing antibody specificity of alternative HA DNA vaccines in mice To evaluate the efficacy of multivalent DNA vaccines, initial studies were performed in mice. Expression vectors encoding HAs from ten phylogenetically diverse strains of influenza viruses [33] were generated by synthesis of cDNAs (see Materials and Methods) in plasmid expression vectors, pCMV/R or pCMV/R 8kB, which mediates high level expression and immunogenicity in vivo [34, 35, 36] . Animals were immunized with each expression vector intramuscularly (IM) at three week intervals, and the antisera were evaluated on day 14 after the third immunization for their ability to neutralize HPAI H5N1 pseudotyped lentiviral vectors as previously described [35, 36] . We have previously shown that lentiviral assay inhibition (LAI) yields similar results to microneutralization and HAI analyses with higher sensitivity in mice [35, 36]
To determine whether immunization with multiple HAs simultaneously could expand the breadth of the neutralizing antibody response without significant loss of magnitude, a combination of 10 HA DNA vaccine immunogens was administered IM at proportionally lower concentration (1.5 mg per immunogen) into groups of 10 mice (see Materials and Methods). Remarkably, despite a log lower DNA concentration of each component, significant neutralizing antibody titers were generated to each of the 10 immunogens, with .80% neutralization against 6 out of 12 H5 HA pseudoviruses at dilutions of up to 1:400 ( Fig. 2A) .
To evaluate whether similar breadth of immunity could be generated with fewer immunogens, two different combinations of 5 immunogens were selected, based on the phylogenetic diversity of HA among the avian influenza viruses [33] and the crossreactivity of the neutralizing antibody responses of select individual immunogens (Fig. 1 ). As expected, there were substantial differences in the breadth of neutralization between these two sets of 5 immunogen multivalent vaccines (Fig. 2 , B vs. C). In one set, while neutralization of homologous strains was comparable to the monovalent and the 10 immunogen multivalent immune response, fewer cross-reactive antibodies were detected, directed most prominently against A/Iraq Protection of DNA-vaccinated mice against challenge with heterologous H5N1 A/Vietnam/1203/2004 influenza virus Mice immunized as described above were challenged with a heterologous H5N1 virus 68 weeks after the final DNA vaccination. Animals were then challenged with 10 LD 50 of the highly pathogenic A/Vietnam/1203/2004 virus intranasally, and morbidity and mortality were monitored for 21 days after the viral challenge. The control animals, injected with the plasmid expression vector with no insert, died within 10 days of infection. Complete survival was observed in the groups immunized with the 10 component and set 2 of the 5 component multivalent DNA vaccines (Fig. 3) . Immunization with HA derived from the A/ Indonesia/05/2005 strain or set 1 of the 5 component multivalent DNA vaccine showed a survival rate approaching 90%. In contrast, animals injected with HA plasmid DNA derived from A/ Anhui/1/2005, which has diverged more from A/Vietnam/ 1203/2004, showed a lower percent survival (70%) after lethal viral challenge. Survival differences between groups were assessed using a log-rank test and the Gehan-Wilcoxon test on the survival curves for pairs of groups. A test was deemed significant if the pvalue was ,0.01. Mice injected IM with different HAs, A/ Indonesia/5/05, A/Anhui/1/05, 10HA, 5 HA (Set 1), or 5 HA (Set 2) showed a significant difference compared to control (all p values,0.001). Among the HA-immunized groups, there was no significant difference between any two groups (p.0.08 for all comparisons). For example, no significant difference was observed between the A/Anhui/1/05 group, which had the least survival among the HA immunized groups (7 out of 10), and other HA groups: A/Indonesia/5/05 (p = 0.377), 10 HA (p = 0.082), 5 HA (Set 1) (p = 0.101), or 5 HA (Set 2) (p = .411). Therefore, we cannot exclude the possibility that the 3 deaths in the A/Anhui/1/05 group may have been due to random chance.
Since it is desirable to confer protective immunity in poultry and HA DNA vaccination was effective in mice, we next examined the breadth and potency of single or multiple HA plasmid immunization in chickens. The ability of chickens to generate specific antibodies was assessed with three strains that showed broad cross protection in mouse studies (A/Vietnam/1203/2004, A/Anhui/ 1/2005 and A/Indonesia/05/2005), administered individually or in combination, by different injection methods. In addition to needle injection, a needle-free repetitive injection device, Agro-JetH (Medical International Technology, Inc., Denver, CO), was analyzed. This device disperses the 0.1 to 5 ml injection doses into the dermal, subcutaneous, or intramuscular tissue depending upon the pressure adjustments, powered by a CO 2 gas pressure plunger [39] . The injection conditions were determined by histologic analysis of tissues that received injections of India ink; a pressure of 48 psi was chosen since it enabled consistent delivery into intradermal and subcutaneous tissues (Fig. S1 ).
Immunization of chickens with the control plasmid (CMV/R) without an HA gene insert elicited minimal neutralizing antibody titers compared to HA-immunized animals 1 week after 3 DNA immunizations. Nearly all chickens immunized with either monovalent or multivalent HA DNA vaccines generated significant neutralization titers ( Fig. 4 and Table S1 ). In general, there was a progressive increase in the amount of neutralization after each successive DNA vaccination (data not shown) with maximal response at 1 week after the 3 rd DNA immunization, with highest and most consistent levels in the trivalent vaccine group delivered with the Agro-JetH device. Neutralization of the Indonesia HA strain was the most robust, with neutralization nearing 100% at titers greater than 1:3200. Both the monovalent and multivalent vaccines elicited robust homologous ( vaccine (Fig. 4 ). Even though one chicken (238) in the multivalent vaccine group produced almost the same degree of neutralization at each time point and was protected, it did not produce a high neutralizing antibody titer for reasons that were uncertain but possibly related to a non-specific inhibitor in the sera.
To determine whether chickens immunized with single or multiple DNA vaccines were protected from a lethal challenge of a heterologous HPAI H5N1 virus, vaccinated chickens were In panels B and C, mice (n = 10) were immunized with 15 mg of plasmid (3 mg each) three times at 3 week intervals. Serum pools from the immunized animals were collected 14 days after the third immunization. The antisera were tested against the 12 indicated pseudotyped lentiviral vectors at varying dilutions. Error bars at each point indicate the standard deviation; each sample was evaluated in triplicate. In general, the immunized serum neutralized all tested pseudotyped lentiviruses at low dilutions while differences were often observed at high dilution. doi:10.1371/journal.pone.0002432.g002 inoculated with 20 LD 50 of highly pathogenic A/Vietnam/1203/ 2004 heterologous virus intranasally using standard methods [25, 40] and monitored for morbidity, mortality, viral shedding and serum antibodies. While all the control animals died within 2 days of infection, 100% survival was noted in the rest of the chickens (Fig. 5A ). The animals that were healthy, showing no signs of clinical disease or malaise, were euthanized on day 14. There was no evidence for viral shedding monitored via tracheal and cloacal swabs of infected chickens 2-14 days after challenge as determined by embryonal inoculation (data not shown: egg infectious dose 50 (EID 50 ) limit of detection ,100 virus particles).
To compare the relative efficacy of DNA vaccines delivered IM by needle and syringe versus the needle-free Agro-JetH device injection, a dose-response study was performed with amounts of DNA vaccine ranging from 500 to 0.5 mg with two inoculations. In these experiments, the HA derived from A/chicken/Nigeria/641/ 2006 was substituted for A/Vietnam/1203/2004 since it represented a more contemporary isolate. The observed rate of protection was higher among the animals receiving 5 mg by Agro-Jet (8/8) than by IM injection (6/8) (Fig. 5, B vs. C). Both modes provided complete protection for all animals at doses higher than this, and 25% protection for the animals receiving 0.5 mg doses (Fig. 5B, C) . Survival differences between consecutive doses were assessed using a log-rank test on the survival curves for pairs of groups. A test was deemed significant if the p-value was ,.01, and marginally significant if the p-value was ,.05 but ..01. Chickens injected IM showed a marginally significant difference between 0.5 and 5 mg (p = .047). In the same group there was a significant difference between control and 5, 50 and 500 mg (p,.001 for all comparisons) and the difference between control and 0.5 mg was marginally significant (p = .016). Chickens that were injected using Agro-JetH showed a significant difference between 0.5 and 5 mg (p = .004) and between control and 5, 50, and 500 mg (p,.001 for all comparisons). There were no differences between control and 0.5 mg or between 5, 50, and 500 mg. Lastly, the survival differences between Agro-JetH and IM for each dose group were not significant. The neutralizing antibody response to homologous and heterologous HAs corresponded with protection and correlated with dose, with higher titers elicited by injection with Agro-JetH compared to needle (Table S2) . We assessed viable viral shedding after inoculation by chick embryo inoculation three days after virus challenge (Week 8). While we noted some embryonic lethality at the 0.5 mg dose, there was no embryonic lethality at 5, 50 or 500 mg groups (data not shown).
Since the HPAI H5N1 virus first appeared ten years ago, this highly pathogenic avian influenza virus has shown increasing diversification and dissemination in Asia, Africa, and Europe [28, [41] [42] [43] [44] . In addition to its effects on human health by crossspecies transmission [28, 45, 46] and ability to compromise food sources, it poses a continuing threat to public health as it evolves and adapts in different species. The pandemic potential of this virus, especially as it relates to the poultry industry and for reservoir avian hosts, underscores the need for a vaccine that offers broad spectrum immunity and protection against lethal viral challenge. While the virus remains restricted in its ability to infect humans and undergo efficient human-to-human transmission [28, 47] , its persistence and spread in poultry increases the risk of the emergence of a pandemic strain. One approach to pandemic risk reduction is to limit the propagation of the virus in poultry and other relevant avian species.
We have previously reported that DNA vaccines encoding HA can confer protection against a highly lethal human pandemic influenza virus, the 1918 H1N1 virus, in mice [36] . DNA vaccines offer several advantages, including the ability to express diverse antigens, tolerability in various hosts, ease of delivery, and stability for storage and distribution without the necessity of maintaining a cold chain; they have been shown to be safe and efficacious in a variety of animal models [2, 4, 12, 22, 48] . Because they do not contain other viral proteins used to screen for infection, they also address the need to differentiate vaccinated from infected animals. There is evidence that DNA vaccination elicits cell-mediated immunity against influenza HA in addition to inducing an antibody response [36] , an effect that could significantly contribute to protective immunity as viruses show genetic drift and reduced susceptibility to neutralization.
Ideally, a highly effective influenza vaccine should not only be able to let the host develop a protective immune response against a matching live virus challenge but also elicit robust protective immune responses against a broad range of homologous and heterologous H5 influenza strains. A multivalent H5 vaccine containing diverse serotypes could expand the antigenic breadth sufficiently to provide protection against heterologous challenge and may preclude the emergence of vaccine-resistant strains that may arise due to evolutionary vaccine pressure on the virus. Due to the antigenic drift and shift of the influenza virus genome, it has been very difficult to predict the next dominant strain of an avian endemic outbreak. DNA vaccines can be synthesized in a relatively short period of time, and the targeted mutations can be tailored to specific viral serotypes. The mutations promote a focused and enhanced immune response [3, 49, 50] that may be particularly important in the event of an outbreak where specificity is the key to epidemic control. The use of modified codons ensures maximal expression in the host and eliminates the possibility of recombination with influenza viruses that might potentially generate new strains.
A more broadly protective murine vaccine was developed here by including more HAs from varying strains in the multivalent vaccine (Figs. 2 and 3) . However, it is less practical to include large numbers of different HAs in one vaccine due to the cost and complexity of manufacturing such a vaccine. Therefore, we simplified the vaccine regimen based on cross-neutralization studies and phylogenetic relationships. A trivalent vaccine was subsequently identified for further studies. Due [51] .
While three DNA immunizations were used initially to demonstrate protective immunity and have been used previously to elicit protection in mice [36] , we found that effective protective immunity could be induced with two DNA vaccinations and as little as 5 mg trivalent DNA immunization using the ID/SC route with the Agro-JetH device. In addition, based on the chick embryo inoculation data, we believe that there is effective neutralization of the virus and lack of infectious viral shedding in chicken vaccinated with as little as 5 mg of DNA. The device's capacity for rapid repetitive injection and the lower quantity and stability of DNA enhance the practicality and utility of this approach for vaccination of endangered species in captivity or administration to poultry or other animals.
A/Vietnam/1203/2004 (H5N1) (A/VN/1203/04) was obtained from the repository at the Centers for Disease Control and Prevention (CDC), Atlanta, Georgia. The virus was propagated in 10-day old embryonated chicken eggs at 35uC and stored at 270uC until use. The virus was titrated by the Reed and Muench method to determine EID 50 [52] . GenBank ABD28180) were synthesized using human-preferred codons (GeneArt, Regensburg, Germany) [36] . HA cDNAs from diverse strains of influenza viruses were then inserted into plasmid expression vectors, pCMV/R or pCMV/R 8kB, which mediates high level expression and immunogenicity in vivo [34, 35, 36] . For initial trivalent immunizations in chickens, the A/Vietnam/1203/ 2004, A/Anhui/1/2005 and A/Indonesia/05/2005 strains were used and in the dose response study, the Vietnam strain was replaced with A/chicken/Nigeria/641/2006. The immunogens used in DNA vaccination contained a cleavage site mutation (PQRERRRKKRG to PQRETRG) as previously described [35, 36] . This mutation was generated by site-directed mutagenesis using a QuickChange kit (Stratagene, La Jolla, CA).
DNA immunization of mice [6] [7] [8] week old female BALB/c mice were purchased from The Jackson Laboratory and maintained in the AAALAC-accredited Vaccine Research Center Animal Care Facility (Bethesda, MD) under specific pathogen-free conditions. All experiments were approved by the Vaccine Research Center Animal Care and Use Committee. The mice were immunized as previously described [5] . Briefly, mice (10 animals for all test groups, 20 animals for the
The study was carried out in the AAALAC-accredited animal facility at the University of Maryland School of Medicine. Six groups of 8 one-day-old male and female SPAFAS White Leghorn Chickens, Gallus domesticus, were obtained from Charles River Laboratories (Connecticut). The animals were housed in brooder and grower cages (McMurray Hatcheries, Iowa). Feed (Teklad Japanese Quail Diet -3050, Harlan-Teklad, WI) and water were provided to the animals ad libitum. The study was performed in strict accordance with the ''Guide'' after approvals from the Animal Care and Use Committees of the Vaccine Research Center, NIH and the University of Maryland. DNA immunizations were performed as described at 0, 3 and 6 weeks. A total dose of 500 mg of one or a combination of the following DNA plasmids in a volume of 250 ml was administered to each animal: pCMV/ R, pCMV/R-HA
Agro-JetH is a needle-free device used for mass delivery of vaccines and drugs in livestock and poultry. The device is semiautomatic and requires a small CO 2 tank or compressed air for low pressure delivery. Upon trigger activation, CO 2 disperses the injectate at a precise dose into the muscle, dermis or subcutaneous tissue depending on the setting that was standardized for our use. We used an effective volume of 0.1 ml in our injectate [39] . In this study we were able to effectively deliver 0.1 ml of injectate into the animal's dermis/subcutaneous tissue at a pressure of 48-55 psi.
Sixty-eight weeks after the last immunization, female BALB/c mice were lightly anesthetized with Ketamine/Xylazine and inoculated intranasally with 10 LD 50 of A/Vietnam/1203/2004 virus diluted in phosphate-buffered saline in a 50 ul volume. Mice were monitored daily for morbidity and measured for weight loss and mortality for 21 days post infection. Any mouse that had lost more than 25% of its body weight was euthanized. All experiments involving the HPAI virus were conducted in an AAALAC accredited facility (BioQual Inc., Gaithersburg, MD) under BSL 3 conditions that included enhancements required by the USDA and the Select Agent Program.
White Leghorn chickens were challenged one week after the last immunization with 20 lethal dose 50 (LD 50 ) of A/Vietnam/1203/04 (H5N1) influenza A virus, equivalent to 2610 4 EID 50 based on previous challenges [53] . Chickens were infected with 200 ml virus intranasally. Tracheal and cloacal swabs were collected days 3 and 5 post-challenge and stored in glass vials containing BHI medium (BBL TM Brain Heart Infusion, Becton Dickinson) at 280uC. Blood was collected 14 days post-challenge and serum was titered by microneutralization assay. Chickens were observed and scored daily for clinical signs of infection, morbidity and mortality. Chickens that survived the study were bled and humanely euthanized at day 14 post-challenge. Lungs, heart, intestine and kidney were collected and samples were stored in formalin for histopathology. Experiments were carried out under BSL3+ conditions with investigators wearing appropriate protective equipment and compliant with all Institutional Animal Care and Use Committee-approved protocols and under Animal Welfare Act regulations at the University of Maryland, College Park, Maryland.
Representative tracheal and cloacal swabs were chosen to run an EID 50 assay for comparison and virus titers were determine by the method of Reed and Meunch [52] . Briefly, swabs were used to infect 10 day-old embryonated chicken eggs in 10-fold dilutions. Three eggs were inoculated per dilution and incubated for 48 hours before titration.
Neutralizing antibodies were titrated from serum samples collected week 5 and 7 post-vaccination and day 14 post-challenge. The microneutralization assay was performed using a 96-well plate format. Serum was treated with receptor-destroying enzyme (Denka Seiken Co.) and treated at 37uC per the manufacturer's instructions. After an overnight incubation and subsequent inactivation samples were brought to a final dilution of 1:10 using PBS and each sample was serially diluted and virus, diluted to 100 TCID 50 , was added to each well. The plates were then incubated at 37uC, 5% CO 2 for 1-2 hours. Following incubation, supernatants were used to infect a second 96-well plate of MDCK cells. Microplates were incubated at 4uC for 15 minutes and then 37uC, 5% CO 2 for 45 minutes. Supernatants of serum and virus were then discarded and 200 ml of OptiMEM (containing 1X antibiotics/antimycotics, 1 mg/ml TPCK-trypsin) was added and incubated at 37uC, 5% CO 2 for 3 days. After 3 days, 50 ml of the supernatant from each well was transferred into a new 96-well microplate, and an HA assay was performed to calculate the antibody titers. Virus and cell controls were included in the assay.
Two-fold dilutions of heat-inactivated sera were tested in a microneutralization assay as previously described [54] for the presence of antibodies that neutralized the infectivity of 100 TCID 50 (50% tissue culture infectious dose) of the A/Vietnam/ 1203/2004 H5N1 virus on MDCK cell monolayers by using two wells per dilution on a 96-well plate.
The recombinant lentiviral vectors expressing a luciferase reporter gene were produced as previously described [35, 36] . For the neutralization assay, antisera from immunized animals were heat-inactivated at 55uC for 30 minutes and mixed with 50 ml of pseudovirus at various dilutions. The sera/virus mixture was then added to 293A cells in 96-well B&W TC Isoplates (Wallac, Turku, Finland; 12,000 cells/well). Two hours later, the plates were washed and fresh medium was added. Cells were lysed in mammalian cell lysis buffer (Promega, Madison, WI) 24 hrs after infection and luciferase activity was measured using the Luciferase Assay System (Promega, Madison, WI).
The following strains were used for the production of pseudotyped viruses: for HA we used A/Thailand/1(KAN-
The HA/HI titers were determined as previously described [54] . Briefly, HA titers were calculated using 50 ml of 0.5% chicken red blood cell suspension in PBS added to 50 ml of twofold dilutions of virus in PBS. This mix was incubated at room temperature for 30 minutes. The HA titers were calculated as the reciprocal value of the highest dilution that caused complete hemagglutination. HI titers were calculated by titrating 50 ml of antiserum treated with receptor-destroying enzyme and an equivalent amount of A/Vietnam/1203/2004 virus (four hemagglutinating doses) was added to each well. Wells were incubated at room temperature for 30 minutes and 50 ml of a 0.5% suspension of chicken red blood cells was added. HI titers were calculated after 30 minutes as the reciprocal of the serum dilution that inhibited hemagglutination.
Table S1 Hemagglutination inhibition (HI), microneutralization titer (NT), and LAI of sera from individual chickens immunized with different vaccines. Sera from immunized animals were obtained at week 5 or 7, a week before or after the final boost, and neutralization was assessed by HI, microneutralization (NT) and LAI (shown as IC 50 ). Individual animal serum of each group is shown and was analyzed as described in the Materials and Methods section. Figure S1 Characterization of needle-free (Agro-JetH) DNA immunization in chickens. To evaluate the distribution of fluid into superficial or deep layers of subcutaneous tissues after delivery by AgroJetH, 4 or 7 week old chickens were injected with a solution containing India ink with this needle-free device at various pressures, ranging from 45 to 55 mm Hg. Three sites (thigh, wing and breast) were used, and biopsies were taken for routine hematoxylin and eosin staining. Representative sections of thigh injections are shown from 7 week old chickens and were similar at 4 weeks (data not shown). While the 48 mm Hg pressure deposited the injectate into the dermis/subcutaneous region (left), the higher pressure injections, 52 and 58 mm Hg, deposited the injectate into the subcutaneous and muscle layers (middle, right). 48 mm Hg consistently provided an optimal pressure to deposit the injectate into the dermis and subcutaneous tissue and was chosen for all AgroJetH immunizations. Found at: doi:10.1371/journal.pone.0002432.s003 (10.74 MB DOC)
|
What is the conclusion of this study?
| 5,286 |
it is possible to develop a multivalent DNA vaccine for poultry that can protect against multiple HPAI H5N1 strains and that could keep pace with the continued evolution of avian influenza viruses
| 6,165 |
1,632 |
Metabolic engineering of Escherichia coli into a versatile glycosylation platform: production of bio-active quercetin glycosides
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573293/
SHA: f4cd52975e6aa33e8c947082eda9b261952b0f8f
Authors: De Bruyn, Frederik; Van Brempt, Maarten; Maertens, Jo; Van Bellegem, Wouter; Duchi, Dries; De Mey, Marjan
Date: 2015-09-16
DOI: 10.1186/s12934-015-0326-1
License: cc-by
Abstract: BACKGROUND: Flavonoids are bio-active specialized plant metabolites which mainly occur as different glycosides. Due to the increasing market demand, various biotechnological approaches have been developed which use Escherichia coli as a microbial catalyst for the stereospecific glycosylation of flavonoids. Despite these efforts, most processes still display low production rates and titers, which render them unsuitable for large-scale applications. RESULTS: In this contribution, we expanded a previously developed in vivo glucosylation platform in E. coli W, into an efficient system for selective galactosylation and rhamnosylation. The rational of the novel metabolic engineering strategy constitutes of the introduction of an alternative sucrose metabolism in the form of a sucrose phosphorylase, which cleaves sucrose into fructose and glucose 1-phosphate as precursor for UDP-glucose. To preserve these intermediates for glycosylation purposes, metabolization reactions were knocked-out. Due to the pivotal role of UDP-glucose, overexpression of the interconverting enzymes galE and MUM4 ensured the formation of both UDP-galactose and UDP-rhamnose, respectively. By additionally supplying exogenously fed quercetin and overexpressing a flavonol galactosyltransferase (F3GT) or a rhamnosyltransferase (RhaGT), 0.94 g/L hyperoside (quercetin 3-O-galactoside) and 1.12 g/L quercitrin (quercetin 3-O-rhamnoside) could be produced, respectively. In addition, both strains showed activity towards other promising dietary flavonols like kaempferol, fisetin, morin and myricetin. CONCLUSIONS: Two E. coli W mutants were engineered that could effectively produce the bio-active flavonol glycosides hyperoside and quercitrin starting from the cheap substrates sucrose and quercetin. This novel fermentation-based glycosylation strategy will allow the economically viable production of various glycosides. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12934-015-0326-1) contains supplementary material, which is available to authorized users.
Text: Flavonoids are a class of plant secondary metabolites, which are chemically characterized by a 15-carbon backbone that consists of two phenyl rings and a heterocyclic ring. To date, over 10,000 flavonoids have been characterized from various plants, which are classified according to their chemical structure, i.e., the number and presence of hydroxyl groups and further functional group modifications into various subgroups, such as anthoxanthins, flavanones, and flavanonols [1, 2] .
In recent years flavonoids have garnered much attention from various application domains because of the various beneficial effects on human health that have been attributed to them, such as anticancer [3] and antioxidant [4] to anti-inflammatory [5] , antimicrobial [6] and antiviral [6, 7] effects. As final step in their biosynthesis, flavonoids are often glycosylated which has a profound effect on their solubility, stability and bio-activity [8, 9] . For example, the best studied flavonol quercetin, which makes up to 75 % of our daily flavonoid intake, predominantly occurs as different glycosides. Over 350 different quercetin glycoforms have been reported to date with varying pharmacological properties [10, 11] . In this context, hyperoside (quercetin 3-O-galactoside) and quercitrin (quercetin 3-O-rhamnoside) ( Fig. 1) have gained a lot of attention as valuable products for the pharmaceutical industry e.g., as powerful antioxidants with cytoprotective effects [12] [13] [14] [15] and as promising antiviral agents that block replication of the influenza virus [16] or inhibit the viruses hepatitis B [17] and SARS [18] . Furthermore, they have been attributed with anti-inflammatory [19, 20] , antidepressant [21, 22] , apoptotic [23] and antifungal [24] activities, rendering them interesting therapeutics resulting in a steadily increasing market demand.
To date, the majority of quercetin and its glycosides are extracted from plant material, which is generally a laborious and low-yielding process requiring many purification steps [25] . In vitro plant cell cultures or engineered plants can be used to overcome the low yields and improve production [26] [27] [28] , however since metabolic engineering of plants is both very controversial and still in its infancy [29] , this approach is often restricted to small-scale production. Although chemical synthesis of quercetin (glycosides) has proven to be feasible [30] [31] [32] , stereoselective formation of glycosidic linkages is often hampered by the presence of various reactive groups [33] , which requires many protecting and deprotecting steps [34] . In addition, the generation of toxic waste and a low atomefficiency [35] render these production processes neither sustainable nor economically viable.
As a result, in the last two decades enormous efforts have been invested in the development of alternative production methods for these specialized (secondary) plant metabolites [36] . Advances in the fields of protein engineering, systems and synthetic biology have accelerated these efforts to transform model organisms like Escherichia coli and Saccharomyces cerevisiae in real microbial cell factories for the sustainable production of flavonoids [37] [38] [39] . Subsequently, strategies for the in vivo glycosylation of flavonoids have also been developed. These are typically based on both the overexpression of specific glycosyltransferases, which transfer a sugar residue from an activated nucleotide sugar to an aglycon in a stereoand regioselective way, and the engineering or introduction of the targeted nucleotide sugar pathway. In this way, Fig. 1 Transformation of E. coli W into a sucrose-based galactosylation and rhamnosylation platform. The metabolic engineering strategy applied makes use of several gene deletions (indicated in red) and overexpressions of genes (indicated in green). The rational of a split metabolism is applied, whereby sucrose is divided by sucrose phosphorylase (BaSP) in fructose to be used for growth and a glucose 1-phosphate as activated precursor for UDP-glucose. The latter is a universal pivot molecule for the formation of UDP-galactose and UDP-rhamnose, interconversions catalyzed by the enzymes GalE and MUM4, respectively. To ensure growth-coupled production, various genes, involved in the metabolization of these UDPsugars and their precursors, were knocked out (shown in red). The production of the bioactive quercetin glycosides hyperoside and quercitrin was chosen to evaluate the versatility of the engineered production platform. Finally, the introduction of either the glycosyltransferase F3GT or RhaGT ensures efficient galactosylation or rhamnosylation, respectively various quercetin glycosides have already been produced in E. coli such as the naturally occurring 3-O-glucoside [40] , 3-O-xyloside [41] and 3,7-O-bisrhamnoside [42] , or the new-to-nature quercetin 3-O-(6-deoxytalose) [43] . However, despite these engineering efforts, the reported product rates and titers are still in the milligram range, rendering these microbial production hosts unsuitable for industrial applications. The developed production processes are typically biphasic bioconversion processes using resting cells, which makes it difficult to improve production rates [44] . Furthermore, such systems often entail expensive growth media or the addition of enzyme inducers, making the overall process very costly.
To tackle these problems, we previously developed an efficient platform for the glucosylation of small molecules in E. coli W [45] . Through metabolic engineering, a mutant was created which couples the production of glucosides to growth, using sucrose as a cheap and sustainable carbon source. By introducing the sucrose phosphorylase from Bifidobacterium adolescentis (BaSP) sucrose can be split into fructose to be used for growth purposes and glucose 1-phosphate (glc1P) to be used as precursor for UDP-glucose (UDP-glc) formation ( Fig. 1) . To impede the conversion of glc1P into biomass precursors, several endogenous genes involved in its metabolization such as phosphoglucomutase (pgm) and glucose-1-phosphatase (agp) were knocked out. Subsequently, glc1P can efficiently be channeled towards UDP-glc by overexpressing the uridylyltransferase from Bifidobacterium bifidum (ugpA). Metabolization of UDP-glc is prevented by knocking out the UDP-sugar hydrolase (ushA) and the galactose operon (galETKM). However, in view of the pivotal role of UDP-glc in the production of a large variety of UDP-sugars, this glucosylation system can easily be extended towards other UDP-sugars, such as UDP-galactose (UDP-gal), UDPrhamnose (UDP-rha) and UDP-glucuronate.
In the present contribution, this previously developed E. coli W-based glucosylation platform is transformed into a platform for galactosylation and rhamnosylation ( Fig. 1) , whose potential is demonstrated using the galactosylation and rhamnosylation of exogenously fed quercetin yielding hyperoside and quercitrin, respectively, as case study.
Escherichia coli W is a fast-growing non-pathogenic strain which tolerates osmotic stress, acidic conditions, and can be cultured to high cell densities, making it an attractive host for industrial fermentations [46] . Moreover, E. coli W is able to grow on sucrose as sole carbon source [46] , which is an emerging feedstock for the production of bio-products. Hence, E. coli W was selected as host for sucrose-based in vivo glycosylation. Prior to the production of the glycosides hyperoside and quercitrin in E. coli W, the toxicity of their aglycon quercetin was investigated. To this end, the wild type (WT) strain was grown on minimal sucrose medium containing different concentrations of quercetin (0, 0.15 and 1.5 g/L). The specific growth rates (h −1 ) (0.96 ± 0.06, 0.92 ± 0.05 and 0.87 ± 0.06, respectively) were not significantly different (p ANOVA = 0.12) nor from the one previously determined for the WT [45] (p = 0.69, p = 0.98 and p = 0.68, respectively). On the other hand, the optical density at 600 nm after 24 h incubation (6.36 ± 0.12, 5.18 ± 0.16 and 4.77 ± 0.20, respectively) was lower (about 20 %) when quercetin was added (p = 0.0002 and p = 0.0001). No significant difference in optical density could be observed between 0.15 and 1.5 g/L quercetin (p = 0.14). In view of the above, it was opted to add 1.5 g/L quercetin to evaluate the potential of the developed glycosylation platform.
To evaluate the in vivo glycosylation potential, strains sGAL1 and sRHA1, which constitutively express the flavonol 3-O-galactosyltransferase from Petunia hybrida and the flavonol 3-O-rhamnosyltransferase from A. thaliana, respectively, were cultured in minimal medium with 1.5 g/L of quercetin for 16 h. TLC analysis of the supernatants of both cultures yielded two new yellow product spots. The TLC spot obtained from the sGAL1 culture, which had the same retention time as the hyperoside standard (R f = 0.5), was subsequently purified and analyzed. Both NMR and MS analysis confirmed the production of quercetin 3-O-galactoside. However, the product spot obtained from the sRHA1 culture had a different retention factor (R f = 0.55) than the quercitrin standard (R f = 0.74), and was identified as isoquercitrin (quercetin 3-O-glucoside). As opposed to other reports on wild type E. coli strains expressing RhaGT, which simultaneously produced quercitrin (quercetin 3-O-rhamnoside) and isoquercitrin [47, 48] , no rhamnoside could be detected. Examination of the E. coli W genome revealed that the gene cluster responsible for the endogenous production of dTDP-rhamnose, which functions as an alternative rhamnosyldonor for RhaGT in E. coli B and K12 derivatives [47] , was not present [46, 49] .
In a follow-up experiment, sGAL1 and sRHA1 were grown on minimal medium with two different concentrations (0.15 and 1.5 g/L) of quercetin. Growth and glycoside formation were monitored during 30 h. The final titers (C p ) and specific productivities (q p ) are shown in Fig. 2 . Remarkably, an increase in quercetin concentration resulted in a two to threefold increase in productivity and titer, indicating that quercetin supply is rate-limiting and crucial for efficient in vivo glycosylation. However, while sGAL1 continuously produced hyperoside during the exponential phase, which is also reflected in the relatively high specific productivity, sRHA1 only started to accumulate significant amounts of isoquercitrin at the end of the exponential phase. This production start coincides with a reduction in specific growth rate, which dropped from 0.35 ± 0.04 to 0.06 ± 0.01 h −1 .
As described in detail in the Background section, we previously metabolically engineered E. coli W to create a platform for in vivo glucosylation of small molecules [45] . In the original base glucosylation strain, sucrose phosphorylase encoded by BaSP was located on a mediumcopy plasmid and transcribed from a medium-strong constitutive promoter (P22) [50] . For reasons of comparison and flexibility, it was opted to integrate BaSP in the genome of E. coli W. In addition, chromosomal integration is advantageous because of a significant increase in gene stability. Since the level of gene expression can considerably be impacted by the genome integration site [51] due to structural differences such as supercoiling DNA regions, two different DNA sites were assessed for BaSP integration, i.e., melA and glgC, which encode an α-galactosidase and a glucose-1-phosphate adenylyltransferase, respectively. To this end, an adapted knockin-knockout procedure for large DNA fragments was applied, which is schematically shown in Additional file 1: Figure S2 . BaSP under control of promoter P22 was knocked in at the two different loci in E. coli W ΔcscAR, which resulted in the E. coli W strains ΔcscAR ΔmelA::L4-P22-BaSP-L5 and ΔcscAR ΔglgC::L4-P22-BaSP-L5. Their maximal specific growth rate (µ max ) on minimal sucrose medium, which is shown in Fig. 3 , was compared to the original strain ΔcscAR + pBaSP. The influence of the knockin locus on the maximal specific growth rate is clear. Interestingly, integration at the melA locus resulted in a strain with a µ max which was not significantly different from the reference strain ΔcscAR + pBaSP. In view of the latter and considering the aimed growth-coupled production, it was opted to integrate BaSP at the melA locus leading to the final production base strain E. coli W ΔcscAR Δpgm Δagp ΔushA ΔlacZYA::P22-lacY ΔgalETKM ΔmelA::L4-P22-BaSP-L5 (sGLYC) as shown in Table 1 .
In nature, UDP-glc serves as a pivot molecule in the formation of a variety of UDP-sugars [44] . For example, using the interconverting enzymes UDP-glucose 4-epimerase (GalE) and UDP-rhamnose synthase (MUM4) UDP-glc can be converted to UDP-gal and UDP-rha, respectively. Though GalE is natively present in E. coli W an alternative homologous epimerase (GalE2) from B. bifidum was also selected and cloned due to the Fig. 2 Comparison of the specific glycoside productivities (q p ) and glycoside titers (C p ) for strains sGAL1, which produces the 3-O-galactoside, and sRHA1, which produces the 3-O-glucoside, when grown for 30 h on minimal medium containing 0.15 or 1.5 g/L of quercetin. Error bars represent standard deviations tight and complex regulation of GalE expression in E. coli W. On the other hand, UDP-rhamnose synthesis is restricted to plants. Due to lack of the rfb cluster [46] E. coli W is even unable to form endogenous dTDP-rhamnose as alternative rhamnosyl donor. Hence, the MUM4 gene from A. thaliana was expressed from plasmid pMUM4 to achieve UDP-rhamnose formation in E. coli (Fig. 1) .
The constructed galactosylation (sGAL) and rhamnosylation (sRHA) strains were grown on minimal medium with two levels (0.15 and 1.5 g/L) of quercetin. Growth and production were monitored to determine the specific productivities, as shown in Fig. 4 . Again, higher extracellular quercetin concentrations resulted in a fivefold increase in q p . However, no significant difference in productivity was observed between sGAL2 and sGAL3 at 1.5 g/L quercetin, indicating that UDP-galactose formation is as efficient with both GalE homologs and not likely the rate limiting step. With sGAL3, the highest hyperoside productivity (68.7 mg/g CDW/h) and titer (0.94 g/L) were obtained, the latter being 3.5-fold higher compared to sGAL1.
In contrast to sRHA1, TLC analysis of the supernatant of the cultures of sRHA2 and sRHA3 resulted in a product spot with a retention factor that corresponds to quercitrin, which was confirmed by MS analysis, thus showing in vivo activity of MUM4. A quercitrin titer of 1.18 g/L and specific productivity of 47.8 mg/g CDW/h were obtained after 30 h incubation of sRHA3 when 1.5 g/L quercetin was added to the medium, which corresponded to a 53 % conversion. Also 51 mg/L of isoquercitrin was produced extracellularly which corresponds with a quercitrin:isoquercitrin production ratio of 24:1. This suggests the preference of RhaGT for UDP-rhamnose when different UDP-sugar donors are present. Possible explanations for the significantly lower specific productivity (fivefold decrease) of sRHA2 as compared to sRHA3 are either a higher metabolic burden [52] caused by the two plasmid system or a too limited activity of the native GalU, which could be insufficient for adequate UDP-glc formation [45] .
To demonstrate the scalability of the developed bioprocess, strain sGAL3 was cultured in a 1-L bioreactor, which also ensures a constant pH set at 6.80 and avoid oxygen limitation. A detailed overview of the consumption of sucrose, growth and hyperoside production is given in Fig. 5 . After a lag-phase, the strain displayed a growth rate of 0.32 ± 0.02 h −1 while simultaneously producing hyperoside. The observed specific productivity (65.9 ± 2.6 mg/g CDW/h) was comparable to the one obtained on shake flask scale. When nearly all quercetin was converted, hyperoside formation slowed down, which can be explained either by the observed correlation between quercetin concentration and q p , or by the reported reversibility of F3GT [53] . It is likely that further improvements in titer and productivity can be realized by optimizing the supply of quercetin using a fed-batch system.
To the best of our knowledge, the results obtained in this study with the engineered sGAL and sRHA strains for the production of hyperoside and quercitrin are the highest reported to date both in terms of titer and production rate. The maximal production rate obtained in this contribution was 6 to 50-fold higher compared to the maximal production rates (r p,max ) of processes reported in the literature [47, 54] as is illustrated in Fig. 6 .
The increased performance, in terms of titer and productivity, obtained with the developed platform can be attributed to the use of a split metabolism in combination with optimally rerouting the flux from glucose 1-phosphate towards UDP-galactose and UDP-rhamnose. The undesired conversion of the activated sugars into biomass Fig. 3 Effect of the chromosomal integration locus of the knockin of BaSP on the growth rate. Strains were grown in shake flasks and the resulting maximal growth rates (µ max ) were compared with E. coli W ΔcscAR with plasmid-based BaSP expression (+pBaSP). Error bars represent standard deviations is impeded by gene deletions, which guarantees a high product yield. In addition, since biomass formation, which is fueled by the fructose moiety of sucrose, and glycoside synthesis go hand in hand and subsequently are performed at the same time at a high rate, a high productivity is equally guaranteed (one-step fermentation process).
Besides quercetin also other flavonols such as kaempferol, fisetin, morin and myricetin significantly contribute to our daily flavonoid intake, which also have extremely diverse beneficial effects [55, 56] . As the sugar moiety is a major determinant of the intestinal absorption of dietary flavonoids and their subsequent bioactivity [57, 58] , the To this end, strains sGAL3 and sRHA3 were grown in tubes with 5 mL minimal medium, each containing 1.5 g/L of either kaempferol, myricetin, morin or fisetin. Growth and production were monitored over 48 h and various spots were observed on TLC with similar retention factors as hyperoside and quercitrin. Mass spectrometry was used to identify the compounds produced, which confirmed the in vivo galactosylation of myricetin, kaempferol, morin and fisetin ( Table 2 ). All compounds occurred with an m/z of [M + 114], due to complexation with trifluoroacetic acid from the mobile phase. The galactoside of morin was produced at a slow rate, which is in accordance to the very low in vitro activity of F3GT towards this flavonol [53] . A possible explanation for this limited activity may be the presence of an unusual hydroxyl group at the 2′ position, which may sterically hinder deprotonation and consequent galactosylation of morin at hydroxyl group 3 [59] .
Incubation of sRHA3 with the different flavonols investigated showed two distinct glycoside spots on TLC, which corresponded to the 3-O-rhamnoside and 3-O-glucoside. Kaempferol proved to be the best substrate for RhaGT and was predominantly rhamnosylated (8:1 ratio), with a titer exceeding 400 mg/L, which is twofold higher than previously reported [47] . Fisetin on the other hand was efficiently glucosylated, yet the formation of its rhamnoside was not as efficient, with a titer below 5 mg/L. A similar preference towards glucoside formation was also observed with myricetin and morin, which indicates that the positioning of the hydroxyl groups is the determining factor for glycosylation with RhaGT. The production of the desired rhamnosides, galactosides or glucosides may be improved considerably by using UGTs that are more specific towards certain flavonols and UDP-sugars. Transformation of the corresponding UGTs in the developed in vivo glycosylation strains presents a promising alternative for the large-scale production of various flavonol glycoforms, which are to date mainly extracted from plant material.
On the other hand, due to the pivotal role of UDP-glc, various other UDP-sugars can be formed in vivo (e.g. UDP-glucuronate, UDP-xylose, UDP-arabinose). In combination with the modularity of the developed glycosylation platform, which permits rapid introduction of any UGT or UDP-sugar pathway, virtually any glycoside can be produced. Hence, this demonstrates that the proposed microbial platform is a robust, versatile and efficient microbial cell factory for the glycosylation (e.g. glucosylation, rhamnosylation, galactosylation) of small molecules. Although obtained productivities are the highest reported today and compete with the current production processes, further improvement can be limited due to solubility issues of the aglycon or of the glycoside. To this end follow-up research can focus on further metabolic engineering (e.g. introduction of the aglycon pathway allowing in vivo gradually production of the aglycon) or on process optimization [e.g. 2-phase (bilayer) fermentation which enables in situ recovery] to improve these issues.
In this contribution, a biotechnological platform was developed for the galactosylation and rhamnosylation of small molecules, such as secondary metabolite natural products, starting from a previously created glucosylation host. To this end, the routes to convert UDP-glucose into UDP-galactose and UDP-rhamnose were introduced by expressing a UDP-glucose epimerase (galE) and a UDP-rhamnose synthase (MUM4), respectively. As a proof of concept, the bio-active flavonol quercetin was selected for galactosylation and rhamnosylation, yielding hyperoside (quercetin 3-O-galactoside) and quercitrin (quercetin 3-O-rhamnoside), respectively. Next, the flavonol 3-O-galactosyltransferase (F3GT) from Petunia hybrida and the flavonol 3-O-rhamnosyltransferase from Arabidopsis thaliana (RhaGT) were overexpressed in the metabolically engineered E. coli W mutants. The strains created were able to produce 940 mg/L of hyperoside and 1176 mg/L of quercitrin at specific production rates of 68.7 mg/g CDW/h and 47.8 mg/g CDW/h, respectively, which are the highest reported to date. Interestingly, both GTs showed in vivo activity towards other dietary flavonols, whereby for example over 400 mg/L of kaempferol 3-O-rhamnoside could be formed extracellularly.
All plasmids used were constructed using Gibson assembly [60] or CLIVA [61] . All PCR fragments were amplified using Q5 polymerase from New England Biolabs (Ipswich, Massachusetts). Oligonucleotides were purchased from IDT (Leuven, Belgium). The plasmids and bacterial strains used in this study are listed in Table 1 . A list of primers for the creation of gene knockouts/knockins and for the cloning of the expression plasmids is given in Additional file 2: Table S1 . E. coli DH5α was used for plasmid cloning and propagation, while E. coli W was used for expression of the production plasmids and the creation of gene knockouts and knockins. Hyperoside, quercitrin, isoquercitrin, kaempferol and myricetin were purchased from Carbosynth (Berkshire, UK). All other chemicals used were purchased from Sigma Aldrich (Germany) unless otherwise indicated.
The expression plasmids for the prod uction of hyperoside and quercitrin were constructed as depicted in Additional file 3: Figure S1A Figure S1D ). The galE [Genbank: JW0742] and galE2 [Genbank: KJ543703] sequences were amplified from the genomic DNA of E. coli and Bifidobacterium bifidum, respectively. CLIVA assembly resulted in the intermediary plasmid pBaSP/F3GT/UgpA ( Figure S1A ), which was subsequently used for the amplification of the F3GT/ UgpA backbone. Gibson assembly of the GalE or GalE2 inserts with this backbone resulted in the final galactosylation plasmids pGalE/F3GT/UgpA and pGalE2/F3GT/ UgpA, respectively ( Figure S1B ). Similarly, MUM4 and RhaGT were introduced using a 3-pieces Gibson assembly ( Figure S1C ), resulting in the final rhamnosylation plasmid pMUM4/RhaGT/UgpA.
The overall E. coli W knockout mutants were created using the one step deletion system of Datsenko and Wanner [62] . The strategy for chromosomal integration of BaSP under control of the constitutive promoter P22 flanked by L4 and L5 at the melA and glgC loci is depicted and explained in Additional file 1: Figure S2 . Transformants were plated on minimal sucrose medium agar plates and grown overnight for screening. The in-house strain E. coli W ΔcscAR Δpgm Δagp ΔushA ΔlacZYA::P22-lacY ΔgalETKM [45] was used for the chromosomal integration of L4-P22-BaSP-L5 at the melA site, yielding the base strain sGLYC (Table 1) . This strain and the E. coli W wild type were transformed with the production plasmids described above, resulting in the galactosylation (sGAL) and rhamnosylation (sRHA) strains given in Table 1 .
Composition of LB and minimal sucrose medium was described previously [45] . Minimal medium agar plates with sucrose (50 g/L) had the same composition as minimal sucrose medium, but contained additionally 15 g/L of agarose. The agarose and salts were autoclaved separately at 121 °C for 21 min. Sucrose was filter sterilized through a 0.22 µm corning filter (Fisher, Belgium) and heated for 1 min in a microwave oven at 800 W prior to mixing it with the warm agarose and salt solutions. 1 mL/L of mineral solution [45] was sterilely added prior to pouring the plates.
Escherichia coli W mutant precultures were grown in 5 mL LB medium with the antibiotics (50 μg/mL kanamycin or carbenicillin) required for maintenance and selection of the plasmids. The cultures were grown for 16 h at 37 °C and 200 rpm and used for the 2 % inoculation of 100 mL minimal sucrose medium in 500 mL shake flasks. For the production of hyperoside and quercitrin, quercetin was added to the minimal medium at a concentration of 0.15 or 1.5 g/L. Growth conditions were the same as previously described [45] . Samples were taken at regular intervals from the broth and, after centrifugation, the supernatant was used for the analysis and quantification of sugars. For the analysis of quercetin and its glycosides, 200 µL of the culture was collected and extracted with 800 µL ethyl acetate. The organic layer was collected, evaporated in a SpeedVac ™ vacuum concentrator (Thermo Fisher, USA) and dissolved in 200 µL of DMSO for HPLC quantification.
The bioreactor set-up and fermentation conditions used are the same as previously described [45] . Production experiments were performed on minimal sucrose medium without MOPS buffer and with the addition of quercetin as acceptor.
Culture samples were primarily analyzed by TLC on Silica gel 60 F 254 precoated plates (Merck, Germany). All plates were run in a closed TLC chamber and developed using standard visualization techniques and agents: UV fluorescence (254 nm) or by staining with 10 % (v/v) H 2 SO 4 and subsequent charring. The mobile phase for detecting the various flavonols and corresponding glycosides consisted of an ethyl acetate:acetic acid:formic acid:water (100:11:11:27 v/v) mixture [63] . Product spot intensities of other flavonol glycosides were processed and quantified using ImageJ [64] . HPLC quantification of sucrose, fructose and glucose was performed using an X-bridge Amide column (35 μm, Waters, USA) as described previously [45] . Quercetin, hyperoside, quercitrin and isoquercitrin were detected with the method described by Pandey et al. [41] using a Varian HPLC system (Agilent technologies, California). Mass spectrometry for determination of the various flavonol glycosides was performed with a Micromass Quattro LC (McKinley Scientific, USA). Detection was performed in negative mode ESI-224 MS with a capillary voltage of 2.53 kV, a cone voltage of 20 V, cone and desolvation gas flows of 93 and 420 L/h, and source and cone temperatures of 150 and 350 °C, respectively.
Quercetin glycosides were extracted from the broth with an equal volume of ethyl acetate after which the organic layer was evaporated to dryness. The remaining product was dissolved in the solvent system described above and run on a preparative TLC plate. The band containing hyperoside (R f 0.53) or quercitrin (R f 0.75) was scraped off, extracted with ethyl acetate and evaporated to yield a bright yellow powder. Products were confirmed by NMR. Spectra were reported elsewhere [47, 65] .
|
What was the conclusion of this study?
| 5,298 |
Two E. coli W mutants were engineered that could effectively produce the bio-active flavonol glycosides hyperoside and quercitrin
| 2,016 |
1,632 |
Metabolic engineering of Escherichia coli into a versatile glycosylation platform: production of bio-active quercetin glycosides
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573293/
SHA: f4cd52975e6aa33e8c947082eda9b261952b0f8f
Authors: De Bruyn, Frederik; Van Brempt, Maarten; Maertens, Jo; Van Bellegem, Wouter; Duchi, Dries; De Mey, Marjan
Date: 2015-09-16
DOI: 10.1186/s12934-015-0326-1
License: cc-by
Abstract: BACKGROUND: Flavonoids are bio-active specialized plant metabolites which mainly occur as different glycosides. Due to the increasing market demand, various biotechnological approaches have been developed which use Escherichia coli as a microbial catalyst for the stereospecific glycosylation of flavonoids. Despite these efforts, most processes still display low production rates and titers, which render them unsuitable for large-scale applications. RESULTS: In this contribution, we expanded a previously developed in vivo glucosylation platform in E. coli W, into an efficient system for selective galactosylation and rhamnosylation. The rational of the novel metabolic engineering strategy constitutes of the introduction of an alternative sucrose metabolism in the form of a sucrose phosphorylase, which cleaves sucrose into fructose and glucose 1-phosphate as precursor for UDP-glucose. To preserve these intermediates for glycosylation purposes, metabolization reactions were knocked-out. Due to the pivotal role of UDP-glucose, overexpression of the interconverting enzymes galE and MUM4 ensured the formation of both UDP-galactose and UDP-rhamnose, respectively. By additionally supplying exogenously fed quercetin and overexpressing a flavonol galactosyltransferase (F3GT) or a rhamnosyltransferase (RhaGT), 0.94 g/L hyperoside (quercetin 3-O-galactoside) and 1.12 g/L quercitrin (quercetin 3-O-rhamnoside) could be produced, respectively. In addition, both strains showed activity towards other promising dietary flavonols like kaempferol, fisetin, morin and myricetin. CONCLUSIONS: Two E. coli W mutants were engineered that could effectively produce the bio-active flavonol glycosides hyperoside and quercitrin starting from the cheap substrates sucrose and quercetin. This novel fermentation-based glycosylation strategy will allow the economically viable production of various glycosides. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12934-015-0326-1) contains supplementary material, which is available to authorized users.
Text: Flavonoids are a class of plant secondary metabolites, which are chemically characterized by a 15-carbon backbone that consists of two phenyl rings and a heterocyclic ring. To date, over 10,000 flavonoids have been characterized from various plants, which are classified according to their chemical structure, i.e., the number and presence of hydroxyl groups and further functional group modifications into various subgroups, such as anthoxanthins, flavanones, and flavanonols [1, 2] .
In recent years flavonoids have garnered much attention from various application domains because of the various beneficial effects on human health that have been attributed to them, such as anticancer [3] and antioxidant [4] to anti-inflammatory [5] , antimicrobial [6] and antiviral [6, 7] effects. As final step in their biosynthesis, flavonoids are often glycosylated which has a profound effect on their solubility, stability and bio-activity [8, 9] . For example, the best studied flavonol quercetin, which makes up to 75 % of our daily flavonoid intake, predominantly occurs as different glycosides. Over 350 different quercetin glycoforms have been reported to date with varying pharmacological properties [10, 11] . In this context, hyperoside (quercetin 3-O-galactoside) and quercitrin (quercetin 3-O-rhamnoside) ( Fig. 1) have gained a lot of attention as valuable products for the pharmaceutical industry e.g., as powerful antioxidants with cytoprotective effects [12] [13] [14] [15] and as promising antiviral agents that block replication of the influenza virus [16] or inhibit the viruses hepatitis B [17] and SARS [18] . Furthermore, they have been attributed with anti-inflammatory [19, 20] , antidepressant [21, 22] , apoptotic [23] and antifungal [24] activities, rendering them interesting therapeutics resulting in a steadily increasing market demand.
To date, the majority of quercetin and its glycosides are extracted from plant material, which is generally a laborious and low-yielding process requiring many purification steps [25] . In vitro plant cell cultures or engineered plants can be used to overcome the low yields and improve production [26] [27] [28] , however since metabolic engineering of plants is both very controversial and still in its infancy [29] , this approach is often restricted to small-scale production. Although chemical synthesis of quercetin (glycosides) has proven to be feasible [30] [31] [32] , stereoselective formation of glycosidic linkages is often hampered by the presence of various reactive groups [33] , which requires many protecting and deprotecting steps [34] . In addition, the generation of toxic waste and a low atomefficiency [35] render these production processes neither sustainable nor economically viable.
As a result, in the last two decades enormous efforts have been invested in the development of alternative production methods for these specialized (secondary) plant metabolites [36] . Advances in the fields of protein engineering, systems and synthetic biology have accelerated these efforts to transform model organisms like Escherichia coli and Saccharomyces cerevisiae in real microbial cell factories for the sustainable production of flavonoids [37] [38] [39] . Subsequently, strategies for the in vivo glycosylation of flavonoids have also been developed. These are typically based on both the overexpression of specific glycosyltransferases, which transfer a sugar residue from an activated nucleotide sugar to an aglycon in a stereoand regioselective way, and the engineering or introduction of the targeted nucleotide sugar pathway. In this way, Fig. 1 Transformation of E. coli W into a sucrose-based galactosylation and rhamnosylation platform. The metabolic engineering strategy applied makes use of several gene deletions (indicated in red) and overexpressions of genes (indicated in green). The rational of a split metabolism is applied, whereby sucrose is divided by sucrose phosphorylase (BaSP) in fructose to be used for growth and a glucose 1-phosphate as activated precursor for UDP-glucose. The latter is a universal pivot molecule for the formation of UDP-galactose and UDP-rhamnose, interconversions catalyzed by the enzymes GalE and MUM4, respectively. To ensure growth-coupled production, various genes, involved in the metabolization of these UDPsugars and their precursors, were knocked out (shown in red). The production of the bioactive quercetin glycosides hyperoside and quercitrin was chosen to evaluate the versatility of the engineered production platform. Finally, the introduction of either the glycosyltransferase F3GT or RhaGT ensures efficient galactosylation or rhamnosylation, respectively various quercetin glycosides have already been produced in E. coli such as the naturally occurring 3-O-glucoside [40] , 3-O-xyloside [41] and 3,7-O-bisrhamnoside [42] , or the new-to-nature quercetin 3-O-(6-deoxytalose) [43] . However, despite these engineering efforts, the reported product rates and titers are still in the milligram range, rendering these microbial production hosts unsuitable for industrial applications. The developed production processes are typically biphasic bioconversion processes using resting cells, which makes it difficult to improve production rates [44] . Furthermore, such systems often entail expensive growth media or the addition of enzyme inducers, making the overall process very costly.
To tackle these problems, we previously developed an efficient platform for the glucosylation of small molecules in E. coli W [45] . Through metabolic engineering, a mutant was created which couples the production of glucosides to growth, using sucrose as a cheap and sustainable carbon source. By introducing the sucrose phosphorylase from Bifidobacterium adolescentis (BaSP) sucrose can be split into fructose to be used for growth purposes and glucose 1-phosphate (glc1P) to be used as precursor for UDP-glucose (UDP-glc) formation ( Fig. 1) . To impede the conversion of glc1P into biomass precursors, several endogenous genes involved in its metabolization such as phosphoglucomutase (pgm) and glucose-1-phosphatase (agp) were knocked out. Subsequently, glc1P can efficiently be channeled towards UDP-glc by overexpressing the uridylyltransferase from Bifidobacterium bifidum (ugpA). Metabolization of UDP-glc is prevented by knocking out the UDP-sugar hydrolase (ushA) and the galactose operon (galETKM). However, in view of the pivotal role of UDP-glc in the production of a large variety of UDP-sugars, this glucosylation system can easily be extended towards other UDP-sugars, such as UDP-galactose (UDP-gal), UDPrhamnose (UDP-rha) and UDP-glucuronate.
In the present contribution, this previously developed E. coli W-based glucosylation platform is transformed into a platform for galactosylation and rhamnosylation ( Fig. 1) , whose potential is demonstrated using the galactosylation and rhamnosylation of exogenously fed quercetin yielding hyperoside and quercitrin, respectively, as case study.
Escherichia coli W is a fast-growing non-pathogenic strain which tolerates osmotic stress, acidic conditions, and can be cultured to high cell densities, making it an attractive host for industrial fermentations [46] . Moreover, E. coli W is able to grow on sucrose as sole carbon source [46] , which is an emerging feedstock for the production of bio-products. Hence, E. coli W was selected as host for sucrose-based in vivo glycosylation. Prior to the production of the glycosides hyperoside and quercitrin in E. coli W, the toxicity of their aglycon quercetin was investigated. To this end, the wild type (WT) strain was grown on minimal sucrose medium containing different concentrations of quercetin (0, 0.15 and 1.5 g/L). The specific growth rates (h −1 ) (0.96 ± 0.06, 0.92 ± 0.05 and 0.87 ± 0.06, respectively) were not significantly different (p ANOVA = 0.12) nor from the one previously determined for the WT [45] (p = 0.69, p = 0.98 and p = 0.68, respectively). On the other hand, the optical density at 600 nm after 24 h incubation (6.36 ± 0.12, 5.18 ± 0.16 and 4.77 ± 0.20, respectively) was lower (about 20 %) when quercetin was added (p = 0.0002 and p = 0.0001). No significant difference in optical density could be observed between 0.15 and 1.5 g/L quercetin (p = 0.14). In view of the above, it was opted to add 1.5 g/L quercetin to evaluate the potential of the developed glycosylation platform.
To evaluate the in vivo glycosylation potential, strains sGAL1 and sRHA1, which constitutively express the flavonol 3-O-galactosyltransferase from Petunia hybrida and the flavonol 3-O-rhamnosyltransferase from A. thaliana, respectively, were cultured in minimal medium with 1.5 g/L of quercetin for 16 h. TLC analysis of the supernatants of both cultures yielded two new yellow product spots. The TLC spot obtained from the sGAL1 culture, which had the same retention time as the hyperoside standard (R f = 0.5), was subsequently purified and analyzed. Both NMR and MS analysis confirmed the production of quercetin 3-O-galactoside. However, the product spot obtained from the sRHA1 culture had a different retention factor (R f = 0.55) than the quercitrin standard (R f = 0.74), and was identified as isoquercitrin (quercetin 3-O-glucoside). As opposed to other reports on wild type E. coli strains expressing RhaGT, which simultaneously produced quercitrin (quercetin 3-O-rhamnoside) and isoquercitrin [47, 48] , no rhamnoside could be detected. Examination of the E. coli W genome revealed that the gene cluster responsible for the endogenous production of dTDP-rhamnose, which functions as an alternative rhamnosyldonor for RhaGT in E. coli B and K12 derivatives [47] , was not present [46, 49] .
In a follow-up experiment, sGAL1 and sRHA1 were grown on minimal medium with two different concentrations (0.15 and 1.5 g/L) of quercetin. Growth and glycoside formation were monitored during 30 h. The final titers (C p ) and specific productivities (q p ) are shown in Fig. 2 . Remarkably, an increase in quercetin concentration resulted in a two to threefold increase in productivity and titer, indicating that quercetin supply is rate-limiting and crucial for efficient in vivo glycosylation. However, while sGAL1 continuously produced hyperoside during the exponential phase, which is also reflected in the relatively high specific productivity, sRHA1 only started to accumulate significant amounts of isoquercitrin at the end of the exponential phase. This production start coincides with a reduction in specific growth rate, which dropped from 0.35 ± 0.04 to 0.06 ± 0.01 h −1 .
As described in detail in the Background section, we previously metabolically engineered E. coli W to create a platform for in vivo glucosylation of small molecules [45] . In the original base glucosylation strain, sucrose phosphorylase encoded by BaSP was located on a mediumcopy plasmid and transcribed from a medium-strong constitutive promoter (P22) [50] . For reasons of comparison and flexibility, it was opted to integrate BaSP in the genome of E. coli W. In addition, chromosomal integration is advantageous because of a significant increase in gene stability. Since the level of gene expression can considerably be impacted by the genome integration site [51] due to structural differences such as supercoiling DNA regions, two different DNA sites were assessed for BaSP integration, i.e., melA and glgC, which encode an α-galactosidase and a glucose-1-phosphate adenylyltransferase, respectively. To this end, an adapted knockin-knockout procedure for large DNA fragments was applied, which is schematically shown in Additional file 1: Figure S2 . BaSP under control of promoter P22 was knocked in at the two different loci in E. coli W ΔcscAR, which resulted in the E. coli W strains ΔcscAR ΔmelA::L4-P22-BaSP-L5 and ΔcscAR ΔglgC::L4-P22-BaSP-L5. Their maximal specific growth rate (µ max ) on minimal sucrose medium, which is shown in Fig. 3 , was compared to the original strain ΔcscAR + pBaSP. The influence of the knockin locus on the maximal specific growth rate is clear. Interestingly, integration at the melA locus resulted in a strain with a µ max which was not significantly different from the reference strain ΔcscAR + pBaSP. In view of the latter and considering the aimed growth-coupled production, it was opted to integrate BaSP at the melA locus leading to the final production base strain E. coli W ΔcscAR Δpgm Δagp ΔushA ΔlacZYA::P22-lacY ΔgalETKM ΔmelA::L4-P22-BaSP-L5 (sGLYC) as shown in Table 1 .
In nature, UDP-glc serves as a pivot molecule in the formation of a variety of UDP-sugars [44] . For example, using the interconverting enzymes UDP-glucose 4-epimerase (GalE) and UDP-rhamnose synthase (MUM4) UDP-glc can be converted to UDP-gal and UDP-rha, respectively. Though GalE is natively present in E. coli W an alternative homologous epimerase (GalE2) from B. bifidum was also selected and cloned due to the Fig. 2 Comparison of the specific glycoside productivities (q p ) and glycoside titers (C p ) for strains sGAL1, which produces the 3-O-galactoside, and sRHA1, which produces the 3-O-glucoside, when grown for 30 h on minimal medium containing 0.15 or 1.5 g/L of quercetin. Error bars represent standard deviations tight and complex regulation of GalE expression in E. coli W. On the other hand, UDP-rhamnose synthesis is restricted to plants. Due to lack of the rfb cluster [46] E. coli W is even unable to form endogenous dTDP-rhamnose as alternative rhamnosyl donor. Hence, the MUM4 gene from A. thaliana was expressed from plasmid pMUM4 to achieve UDP-rhamnose formation in E. coli (Fig. 1) .
The constructed galactosylation (sGAL) and rhamnosylation (sRHA) strains were grown on minimal medium with two levels (0.15 and 1.5 g/L) of quercetin. Growth and production were monitored to determine the specific productivities, as shown in Fig. 4 . Again, higher extracellular quercetin concentrations resulted in a fivefold increase in q p . However, no significant difference in productivity was observed between sGAL2 and sGAL3 at 1.5 g/L quercetin, indicating that UDP-galactose formation is as efficient with both GalE homologs and not likely the rate limiting step. With sGAL3, the highest hyperoside productivity (68.7 mg/g CDW/h) and titer (0.94 g/L) were obtained, the latter being 3.5-fold higher compared to sGAL1.
In contrast to sRHA1, TLC analysis of the supernatant of the cultures of sRHA2 and sRHA3 resulted in a product spot with a retention factor that corresponds to quercitrin, which was confirmed by MS analysis, thus showing in vivo activity of MUM4. A quercitrin titer of 1.18 g/L and specific productivity of 47.8 mg/g CDW/h were obtained after 30 h incubation of sRHA3 when 1.5 g/L quercetin was added to the medium, which corresponded to a 53 % conversion. Also 51 mg/L of isoquercitrin was produced extracellularly which corresponds with a quercitrin:isoquercitrin production ratio of 24:1. This suggests the preference of RhaGT for UDP-rhamnose when different UDP-sugar donors are present. Possible explanations for the significantly lower specific productivity (fivefold decrease) of sRHA2 as compared to sRHA3 are either a higher metabolic burden [52] caused by the two plasmid system or a too limited activity of the native GalU, which could be insufficient for adequate UDP-glc formation [45] .
To demonstrate the scalability of the developed bioprocess, strain sGAL3 was cultured in a 1-L bioreactor, which also ensures a constant pH set at 6.80 and avoid oxygen limitation. A detailed overview of the consumption of sucrose, growth and hyperoside production is given in Fig. 5 . After a lag-phase, the strain displayed a growth rate of 0.32 ± 0.02 h −1 while simultaneously producing hyperoside. The observed specific productivity (65.9 ± 2.6 mg/g CDW/h) was comparable to the one obtained on shake flask scale. When nearly all quercetin was converted, hyperoside formation slowed down, which can be explained either by the observed correlation between quercetin concentration and q p , or by the reported reversibility of F3GT [53] . It is likely that further improvements in titer and productivity can be realized by optimizing the supply of quercetin using a fed-batch system.
To the best of our knowledge, the results obtained in this study with the engineered sGAL and sRHA strains for the production of hyperoside and quercitrin are the highest reported to date both in terms of titer and production rate. The maximal production rate obtained in this contribution was 6 to 50-fold higher compared to the maximal production rates (r p,max ) of processes reported in the literature [47, 54] as is illustrated in Fig. 6 .
The increased performance, in terms of titer and productivity, obtained with the developed platform can be attributed to the use of a split metabolism in combination with optimally rerouting the flux from glucose 1-phosphate towards UDP-galactose and UDP-rhamnose. The undesired conversion of the activated sugars into biomass Fig. 3 Effect of the chromosomal integration locus of the knockin of BaSP on the growth rate. Strains were grown in shake flasks and the resulting maximal growth rates (µ max ) were compared with E. coli W ΔcscAR with plasmid-based BaSP expression (+pBaSP). Error bars represent standard deviations is impeded by gene deletions, which guarantees a high product yield. In addition, since biomass formation, which is fueled by the fructose moiety of sucrose, and glycoside synthesis go hand in hand and subsequently are performed at the same time at a high rate, a high productivity is equally guaranteed (one-step fermentation process).
Besides quercetin also other flavonols such as kaempferol, fisetin, morin and myricetin significantly contribute to our daily flavonoid intake, which also have extremely diverse beneficial effects [55, 56] . As the sugar moiety is a major determinant of the intestinal absorption of dietary flavonoids and their subsequent bioactivity [57, 58] , the To this end, strains sGAL3 and sRHA3 were grown in tubes with 5 mL minimal medium, each containing 1.5 g/L of either kaempferol, myricetin, morin or fisetin. Growth and production were monitored over 48 h and various spots were observed on TLC with similar retention factors as hyperoside and quercitrin. Mass spectrometry was used to identify the compounds produced, which confirmed the in vivo galactosylation of myricetin, kaempferol, morin and fisetin ( Table 2 ). All compounds occurred with an m/z of [M + 114], due to complexation with trifluoroacetic acid from the mobile phase. The galactoside of morin was produced at a slow rate, which is in accordance to the very low in vitro activity of F3GT towards this flavonol [53] . A possible explanation for this limited activity may be the presence of an unusual hydroxyl group at the 2′ position, which may sterically hinder deprotonation and consequent galactosylation of morin at hydroxyl group 3 [59] .
Incubation of sRHA3 with the different flavonols investigated showed two distinct glycoside spots on TLC, which corresponded to the 3-O-rhamnoside and 3-O-glucoside. Kaempferol proved to be the best substrate for RhaGT and was predominantly rhamnosylated (8:1 ratio), with a titer exceeding 400 mg/L, which is twofold higher than previously reported [47] . Fisetin on the other hand was efficiently glucosylated, yet the formation of its rhamnoside was not as efficient, with a titer below 5 mg/L. A similar preference towards glucoside formation was also observed with myricetin and morin, which indicates that the positioning of the hydroxyl groups is the determining factor for glycosylation with RhaGT. The production of the desired rhamnosides, galactosides or glucosides may be improved considerably by using UGTs that are more specific towards certain flavonols and UDP-sugars. Transformation of the corresponding UGTs in the developed in vivo glycosylation strains presents a promising alternative for the large-scale production of various flavonol glycoforms, which are to date mainly extracted from plant material.
On the other hand, due to the pivotal role of UDP-glc, various other UDP-sugars can be formed in vivo (e.g. UDP-glucuronate, UDP-xylose, UDP-arabinose). In combination with the modularity of the developed glycosylation platform, which permits rapid introduction of any UGT or UDP-sugar pathway, virtually any glycoside can be produced. Hence, this demonstrates that the proposed microbial platform is a robust, versatile and efficient microbial cell factory for the glycosylation (e.g. glucosylation, rhamnosylation, galactosylation) of small molecules. Although obtained productivities are the highest reported today and compete with the current production processes, further improvement can be limited due to solubility issues of the aglycon or of the glycoside. To this end follow-up research can focus on further metabolic engineering (e.g. introduction of the aglycon pathway allowing in vivo gradually production of the aglycon) or on process optimization [e.g. 2-phase (bilayer) fermentation which enables in situ recovery] to improve these issues.
In this contribution, a biotechnological platform was developed for the galactosylation and rhamnosylation of small molecules, such as secondary metabolite natural products, starting from a previously created glucosylation host. To this end, the routes to convert UDP-glucose into UDP-galactose and UDP-rhamnose were introduced by expressing a UDP-glucose epimerase (galE) and a UDP-rhamnose synthase (MUM4), respectively. As a proof of concept, the bio-active flavonol quercetin was selected for galactosylation and rhamnosylation, yielding hyperoside (quercetin 3-O-galactoside) and quercitrin (quercetin 3-O-rhamnoside), respectively. Next, the flavonol 3-O-galactosyltransferase (F3GT) from Petunia hybrida and the flavonol 3-O-rhamnosyltransferase from Arabidopsis thaliana (RhaGT) were overexpressed in the metabolically engineered E. coli W mutants. The strains created were able to produce 940 mg/L of hyperoside and 1176 mg/L of quercitrin at specific production rates of 68.7 mg/g CDW/h and 47.8 mg/g CDW/h, respectively, which are the highest reported to date. Interestingly, both GTs showed in vivo activity towards other dietary flavonols, whereby for example over 400 mg/L of kaempferol 3-O-rhamnoside could be formed extracellularly.
All plasmids used were constructed using Gibson assembly [60] or CLIVA [61] . All PCR fragments were amplified using Q5 polymerase from New England Biolabs (Ipswich, Massachusetts). Oligonucleotides were purchased from IDT (Leuven, Belgium). The plasmids and bacterial strains used in this study are listed in Table 1 . A list of primers for the creation of gene knockouts/knockins and for the cloning of the expression plasmids is given in Additional file 2: Table S1 . E. coli DH5α was used for plasmid cloning and propagation, while E. coli W was used for expression of the production plasmids and the creation of gene knockouts and knockins. Hyperoside, quercitrin, isoquercitrin, kaempferol and myricetin were purchased from Carbosynth (Berkshire, UK). All other chemicals used were purchased from Sigma Aldrich (Germany) unless otherwise indicated.
The expression plasmids for the prod uction of hyperoside and quercitrin were constructed as depicted in Additional file 3: Figure S1A Figure S1D ). The galE [Genbank: JW0742] and galE2 [Genbank: KJ543703] sequences were amplified from the genomic DNA of E. coli and Bifidobacterium bifidum, respectively. CLIVA assembly resulted in the intermediary plasmid pBaSP/F3GT/UgpA ( Figure S1A ), which was subsequently used for the amplification of the F3GT/ UgpA backbone. Gibson assembly of the GalE or GalE2 inserts with this backbone resulted in the final galactosylation plasmids pGalE/F3GT/UgpA and pGalE2/F3GT/ UgpA, respectively ( Figure S1B ). Similarly, MUM4 and RhaGT were introduced using a 3-pieces Gibson assembly ( Figure S1C ), resulting in the final rhamnosylation plasmid pMUM4/RhaGT/UgpA.
The overall E. coli W knockout mutants were created using the one step deletion system of Datsenko and Wanner [62] . The strategy for chromosomal integration of BaSP under control of the constitutive promoter P22 flanked by L4 and L5 at the melA and glgC loci is depicted and explained in Additional file 1: Figure S2 . Transformants were plated on minimal sucrose medium agar plates and grown overnight for screening. The in-house strain E. coli W ΔcscAR Δpgm Δagp ΔushA ΔlacZYA::P22-lacY ΔgalETKM [45] was used for the chromosomal integration of L4-P22-BaSP-L5 at the melA site, yielding the base strain sGLYC (Table 1) . This strain and the E. coli W wild type were transformed with the production plasmids described above, resulting in the galactosylation (sGAL) and rhamnosylation (sRHA) strains given in Table 1 .
Composition of LB and minimal sucrose medium was described previously [45] . Minimal medium agar plates with sucrose (50 g/L) had the same composition as minimal sucrose medium, but contained additionally 15 g/L of agarose. The agarose and salts were autoclaved separately at 121 °C for 21 min. Sucrose was filter sterilized through a 0.22 µm corning filter (Fisher, Belgium) and heated for 1 min in a microwave oven at 800 W prior to mixing it with the warm agarose and salt solutions. 1 mL/L of mineral solution [45] was sterilely added prior to pouring the plates.
Escherichia coli W mutant precultures were grown in 5 mL LB medium with the antibiotics (50 μg/mL kanamycin or carbenicillin) required for maintenance and selection of the plasmids. The cultures were grown for 16 h at 37 °C and 200 rpm and used for the 2 % inoculation of 100 mL minimal sucrose medium in 500 mL shake flasks. For the production of hyperoside and quercitrin, quercetin was added to the minimal medium at a concentration of 0.15 or 1.5 g/L. Growth conditions were the same as previously described [45] . Samples were taken at regular intervals from the broth and, after centrifugation, the supernatant was used for the analysis and quantification of sugars. For the analysis of quercetin and its glycosides, 200 µL of the culture was collected and extracted with 800 µL ethyl acetate. The organic layer was collected, evaporated in a SpeedVac ™ vacuum concentrator (Thermo Fisher, USA) and dissolved in 200 µL of DMSO for HPLC quantification.
The bioreactor set-up and fermentation conditions used are the same as previously described [45] . Production experiments were performed on minimal sucrose medium without MOPS buffer and with the addition of quercetin as acceptor.
Culture samples were primarily analyzed by TLC on Silica gel 60 F 254 precoated plates (Merck, Germany). All plates were run in a closed TLC chamber and developed using standard visualization techniques and agents: UV fluorescence (254 nm) or by staining with 10 % (v/v) H 2 SO 4 and subsequent charring. The mobile phase for detecting the various flavonols and corresponding glycosides consisted of an ethyl acetate:acetic acid:formic acid:water (100:11:11:27 v/v) mixture [63] . Product spot intensities of other flavonol glycosides were processed and quantified using ImageJ [64] . HPLC quantification of sucrose, fructose and glucose was performed using an X-bridge Amide column (35 μm, Waters, USA) as described previously [45] . Quercetin, hyperoside, quercitrin and isoquercitrin were detected with the method described by Pandey et al. [41] using a Varian HPLC system (Agilent technologies, California). Mass spectrometry for determination of the various flavonol glycosides was performed with a Micromass Quattro LC (McKinley Scientific, USA). Detection was performed in negative mode ESI-224 MS with a capillary voltage of 2.53 kV, a cone voltage of 20 V, cone and desolvation gas flows of 93 and 420 L/h, and source and cone temperatures of 150 and 350 °C, respectively.
Quercetin glycosides were extracted from the broth with an equal volume of ethyl acetate after which the organic layer was evaporated to dryness. The remaining product was dissolved in the solvent system described above and run on a preparative TLC plate. The band containing hyperoside (R f 0.53) or quercitrin (R f 0.75) was scraped off, extracted with ethyl acetate and evaporated to yield a bright yellow powder. Products were confirmed by NMR. Spectra were reported elsewhere [47, 65] .
|
What are the implications of the novel fermentation-based glycosylation strategy described in this study?
| 5,299 |
the economically viable production of various glycosides
| 2,268 |
1,632 |
Metabolic engineering of Escherichia coli into a versatile glycosylation platform: production of bio-active quercetin glycosides
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573293/
SHA: f4cd52975e6aa33e8c947082eda9b261952b0f8f
Authors: De Bruyn, Frederik; Van Brempt, Maarten; Maertens, Jo; Van Bellegem, Wouter; Duchi, Dries; De Mey, Marjan
Date: 2015-09-16
DOI: 10.1186/s12934-015-0326-1
License: cc-by
Abstract: BACKGROUND: Flavonoids are bio-active specialized plant metabolites which mainly occur as different glycosides. Due to the increasing market demand, various biotechnological approaches have been developed which use Escherichia coli as a microbial catalyst for the stereospecific glycosylation of flavonoids. Despite these efforts, most processes still display low production rates and titers, which render them unsuitable for large-scale applications. RESULTS: In this contribution, we expanded a previously developed in vivo glucosylation platform in E. coli W, into an efficient system for selective galactosylation and rhamnosylation. The rational of the novel metabolic engineering strategy constitutes of the introduction of an alternative sucrose metabolism in the form of a sucrose phosphorylase, which cleaves sucrose into fructose and glucose 1-phosphate as precursor for UDP-glucose. To preserve these intermediates for glycosylation purposes, metabolization reactions were knocked-out. Due to the pivotal role of UDP-glucose, overexpression of the interconverting enzymes galE and MUM4 ensured the formation of both UDP-galactose and UDP-rhamnose, respectively. By additionally supplying exogenously fed quercetin and overexpressing a flavonol galactosyltransferase (F3GT) or a rhamnosyltransferase (RhaGT), 0.94 g/L hyperoside (quercetin 3-O-galactoside) and 1.12 g/L quercitrin (quercetin 3-O-rhamnoside) could be produced, respectively. In addition, both strains showed activity towards other promising dietary flavonols like kaempferol, fisetin, morin and myricetin. CONCLUSIONS: Two E. coli W mutants were engineered that could effectively produce the bio-active flavonol glycosides hyperoside and quercitrin starting from the cheap substrates sucrose and quercetin. This novel fermentation-based glycosylation strategy will allow the economically viable production of various glycosides. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12934-015-0326-1) contains supplementary material, which is available to authorized users.
Text: Flavonoids are a class of plant secondary metabolites, which are chemically characterized by a 15-carbon backbone that consists of two phenyl rings and a heterocyclic ring. To date, over 10,000 flavonoids have been characterized from various plants, which are classified according to their chemical structure, i.e., the number and presence of hydroxyl groups and further functional group modifications into various subgroups, such as anthoxanthins, flavanones, and flavanonols [1, 2] .
In recent years flavonoids have garnered much attention from various application domains because of the various beneficial effects on human health that have been attributed to them, such as anticancer [3] and antioxidant [4] to anti-inflammatory [5] , antimicrobial [6] and antiviral [6, 7] effects. As final step in their biosynthesis, flavonoids are often glycosylated which has a profound effect on their solubility, stability and bio-activity [8, 9] . For example, the best studied flavonol quercetin, which makes up to 75 % of our daily flavonoid intake, predominantly occurs as different glycosides. Over 350 different quercetin glycoforms have been reported to date with varying pharmacological properties [10, 11] . In this context, hyperoside (quercetin 3-O-galactoside) and quercitrin (quercetin 3-O-rhamnoside) ( Fig. 1) have gained a lot of attention as valuable products for the pharmaceutical industry e.g., as powerful antioxidants with cytoprotective effects [12] [13] [14] [15] and as promising antiviral agents that block replication of the influenza virus [16] or inhibit the viruses hepatitis B [17] and SARS [18] . Furthermore, they have been attributed with anti-inflammatory [19, 20] , antidepressant [21, 22] , apoptotic [23] and antifungal [24] activities, rendering them interesting therapeutics resulting in a steadily increasing market demand.
To date, the majority of quercetin and its glycosides are extracted from plant material, which is generally a laborious and low-yielding process requiring many purification steps [25] . In vitro plant cell cultures or engineered plants can be used to overcome the low yields and improve production [26] [27] [28] , however since metabolic engineering of plants is both very controversial and still in its infancy [29] , this approach is often restricted to small-scale production. Although chemical synthesis of quercetin (glycosides) has proven to be feasible [30] [31] [32] , stereoselective formation of glycosidic linkages is often hampered by the presence of various reactive groups [33] , which requires many protecting and deprotecting steps [34] . In addition, the generation of toxic waste and a low atomefficiency [35] render these production processes neither sustainable nor economically viable.
As a result, in the last two decades enormous efforts have been invested in the development of alternative production methods for these specialized (secondary) plant metabolites [36] . Advances in the fields of protein engineering, systems and synthetic biology have accelerated these efforts to transform model organisms like Escherichia coli and Saccharomyces cerevisiae in real microbial cell factories for the sustainable production of flavonoids [37] [38] [39] . Subsequently, strategies for the in vivo glycosylation of flavonoids have also been developed. These are typically based on both the overexpression of specific glycosyltransferases, which transfer a sugar residue from an activated nucleotide sugar to an aglycon in a stereoand regioselective way, and the engineering or introduction of the targeted nucleotide sugar pathway. In this way, Fig. 1 Transformation of E. coli W into a sucrose-based galactosylation and rhamnosylation platform. The metabolic engineering strategy applied makes use of several gene deletions (indicated in red) and overexpressions of genes (indicated in green). The rational of a split metabolism is applied, whereby sucrose is divided by sucrose phosphorylase (BaSP) in fructose to be used for growth and a glucose 1-phosphate as activated precursor for UDP-glucose. The latter is a universal pivot molecule for the formation of UDP-galactose and UDP-rhamnose, interconversions catalyzed by the enzymes GalE and MUM4, respectively. To ensure growth-coupled production, various genes, involved in the metabolization of these UDPsugars and their precursors, were knocked out (shown in red). The production of the bioactive quercetin glycosides hyperoside and quercitrin was chosen to evaluate the versatility of the engineered production platform. Finally, the introduction of either the glycosyltransferase F3GT or RhaGT ensures efficient galactosylation or rhamnosylation, respectively various quercetin glycosides have already been produced in E. coli such as the naturally occurring 3-O-glucoside [40] , 3-O-xyloside [41] and 3,7-O-bisrhamnoside [42] , or the new-to-nature quercetin 3-O-(6-deoxytalose) [43] . However, despite these engineering efforts, the reported product rates and titers are still in the milligram range, rendering these microbial production hosts unsuitable for industrial applications. The developed production processes are typically biphasic bioconversion processes using resting cells, which makes it difficult to improve production rates [44] . Furthermore, such systems often entail expensive growth media or the addition of enzyme inducers, making the overall process very costly.
To tackle these problems, we previously developed an efficient platform for the glucosylation of small molecules in E. coli W [45] . Through metabolic engineering, a mutant was created which couples the production of glucosides to growth, using sucrose as a cheap and sustainable carbon source. By introducing the sucrose phosphorylase from Bifidobacterium adolescentis (BaSP) sucrose can be split into fructose to be used for growth purposes and glucose 1-phosphate (glc1P) to be used as precursor for UDP-glucose (UDP-glc) formation ( Fig. 1) . To impede the conversion of glc1P into biomass precursors, several endogenous genes involved in its metabolization such as phosphoglucomutase (pgm) and glucose-1-phosphatase (agp) were knocked out. Subsequently, glc1P can efficiently be channeled towards UDP-glc by overexpressing the uridylyltransferase from Bifidobacterium bifidum (ugpA). Metabolization of UDP-glc is prevented by knocking out the UDP-sugar hydrolase (ushA) and the galactose operon (galETKM). However, in view of the pivotal role of UDP-glc in the production of a large variety of UDP-sugars, this glucosylation system can easily be extended towards other UDP-sugars, such as UDP-galactose (UDP-gal), UDPrhamnose (UDP-rha) and UDP-glucuronate.
In the present contribution, this previously developed E. coli W-based glucosylation platform is transformed into a platform for galactosylation and rhamnosylation ( Fig. 1) , whose potential is demonstrated using the galactosylation and rhamnosylation of exogenously fed quercetin yielding hyperoside and quercitrin, respectively, as case study.
Escherichia coli W is a fast-growing non-pathogenic strain which tolerates osmotic stress, acidic conditions, and can be cultured to high cell densities, making it an attractive host for industrial fermentations [46] . Moreover, E. coli W is able to grow on sucrose as sole carbon source [46] , which is an emerging feedstock for the production of bio-products. Hence, E. coli W was selected as host for sucrose-based in vivo glycosylation. Prior to the production of the glycosides hyperoside and quercitrin in E. coli W, the toxicity of their aglycon quercetin was investigated. To this end, the wild type (WT) strain was grown on minimal sucrose medium containing different concentrations of quercetin (0, 0.15 and 1.5 g/L). The specific growth rates (h −1 ) (0.96 ± 0.06, 0.92 ± 0.05 and 0.87 ± 0.06, respectively) were not significantly different (p ANOVA = 0.12) nor from the one previously determined for the WT [45] (p = 0.69, p = 0.98 and p = 0.68, respectively). On the other hand, the optical density at 600 nm after 24 h incubation (6.36 ± 0.12, 5.18 ± 0.16 and 4.77 ± 0.20, respectively) was lower (about 20 %) when quercetin was added (p = 0.0002 and p = 0.0001). No significant difference in optical density could be observed between 0.15 and 1.5 g/L quercetin (p = 0.14). In view of the above, it was opted to add 1.5 g/L quercetin to evaluate the potential of the developed glycosylation platform.
To evaluate the in vivo glycosylation potential, strains sGAL1 and sRHA1, which constitutively express the flavonol 3-O-galactosyltransferase from Petunia hybrida and the flavonol 3-O-rhamnosyltransferase from A. thaliana, respectively, were cultured in minimal medium with 1.5 g/L of quercetin for 16 h. TLC analysis of the supernatants of both cultures yielded two new yellow product spots. The TLC spot obtained from the sGAL1 culture, which had the same retention time as the hyperoside standard (R f = 0.5), was subsequently purified and analyzed. Both NMR and MS analysis confirmed the production of quercetin 3-O-galactoside. However, the product spot obtained from the sRHA1 culture had a different retention factor (R f = 0.55) than the quercitrin standard (R f = 0.74), and was identified as isoquercitrin (quercetin 3-O-glucoside). As opposed to other reports on wild type E. coli strains expressing RhaGT, which simultaneously produced quercitrin (quercetin 3-O-rhamnoside) and isoquercitrin [47, 48] , no rhamnoside could be detected. Examination of the E. coli W genome revealed that the gene cluster responsible for the endogenous production of dTDP-rhamnose, which functions as an alternative rhamnosyldonor for RhaGT in E. coli B and K12 derivatives [47] , was not present [46, 49] .
In a follow-up experiment, sGAL1 and sRHA1 were grown on minimal medium with two different concentrations (0.15 and 1.5 g/L) of quercetin. Growth and glycoside formation were monitored during 30 h. The final titers (C p ) and specific productivities (q p ) are shown in Fig. 2 . Remarkably, an increase in quercetin concentration resulted in a two to threefold increase in productivity and titer, indicating that quercetin supply is rate-limiting and crucial for efficient in vivo glycosylation. However, while sGAL1 continuously produced hyperoside during the exponential phase, which is also reflected in the relatively high specific productivity, sRHA1 only started to accumulate significant amounts of isoquercitrin at the end of the exponential phase. This production start coincides with a reduction in specific growth rate, which dropped from 0.35 ± 0.04 to 0.06 ± 0.01 h −1 .
As described in detail in the Background section, we previously metabolically engineered E. coli W to create a platform for in vivo glucosylation of small molecules [45] . In the original base glucosylation strain, sucrose phosphorylase encoded by BaSP was located on a mediumcopy plasmid and transcribed from a medium-strong constitutive promoter (P22) [50] . For reasons of comparison and flexibility, it was opted to integrate BaSP in the genome of E. coli W. In addition, chromosomal integration is advantageous because of a significant increase in gene stability. Since the level of gene expression can considerably be impacted by the genome integration site [51] due to structural differences such as supercoiling DNA regions, two different DNA sites were assessed for BaSP integration, i.e., melA and glgC, which encode an α-galactosidase and a glucose-1-phosphate adenylyltransferase, respectively. To this end, an adapted knockin-knockout procedure for large DNA fragments was applied, which is schematically shown in Additional file 1: Figure S2 . BaSP under control of promoter P22 was knocked in at the two different loci in E. coli W ΔcscAR, which resulted in the E. coli W strains ΔcscAR ΔmelA::L4-P22-BaSP-L5 and ΔcscAR ΔglgC::L4-P22-BaSP-L5. Their maximal specific growth rate (µ max ) on minimal sucrose medium, which is shown in Fig. 3 , was compared to the original strain ΔcscAR + pBaSP. The influence of the knockin locus on the maximal specific growth rate is clear. Interestingly, integration at the melA locus resulted in a strain with a µ max which was not significantly different from the reference strain ΔcscAR + pBaSP. In view of the latter and considering the aimed growth-coupled production, it was opted to integrate BaSP at the melA locus leading to the final production base strain E. coli W ΔcscAR Δpgm Δagp ΔushA ΔlacZYA::P22-lacY ΔgalETKM ΔmelA::L4-P22-BaSP-L5 (sGLYC) as shown in Table 1 .
In nature, UDP-glc serves as a pivot molecule in the formation of a variety of UDP-sugars [44] . For example, using the interconverting enzymes UDP-glucose 4-epimerase (GalE) and UDP-rhamnose synthase (MUM4) UDP-glc can be converted to UDP-gal and UDP-rha, respectively. Though GalE is natively present in E. coli W an alternative homologous epimerase (GalE2) from B. bifidum was also selected and cloned due to the Fig. 2 Comparison of the specific glycoside productivities (q p ) and glycoside titers (C p ) for strains sGAL1, which produces the 3-O-galactoside, and sRHA1, which produces the 3-O-glucoside, when grown for 30 h on minimal medium containing 0.15 or 1.5 g/L of quercetin. Error bars represent standard deviations tight and complex regulation of GalE expression in E. coli W. On the other hand, UDP-rhamnose synthesis is restricted to plants. Due to lack of the rfb cluster [46] E. coli W is even unable to form endogenous dTDP-rhamnose as alternative rhamnosyl donor. Hence, the MUM4 gene from A. thaliana was expressed from plasmid pMUM4 to achieve UDP-rhamnose formation in E. coli (Fig. 1) .
The constructed galactosylation (sGAL) and rhamnosylation (sRHA) strains were grown on minimal medium with two levels (0.15 and 1.5 g/L) of quercetin. Growth and production were monitored to determine the specific productivities, as shown in Fig. 4 . Again, higher extracellular quercetin concentrations resulted in a fivefold increase in q p . However, no significant difference in productivity was observed between sGAL2 and sGAL3 at 1.5 g/L quercetin, indicating that UDP-galactose formation is as efficient with both GalE homologs and not likely the rate limiting step. With sGAL3, the highest hyperoside productivity (68.7 mg/g CDW/h) and titer (0.94 g/L) were obtained, the latter being 3.5-fold higher compared to sGAL1.
In contrast to sRHA1, TLC analysis of the supernatant of the cultures of sRHA2 and sRHA3 resulted in a product spot with a retention factor that corresponds to quercitrin, which was confirmed by MS analysis, thus showing in vivo activity of MUM4. A quercitrin titer of 1.18 g/L and specific productivity of 47.8 mg/g CDW/h were obtained after 30 h incubation of sRHA3 when 1.5 g/L quercetin was added to the medium, which corresponded to a 53 % conversion. Also 51 mg/L of isoquercitrin was produced extracellularly which corresponds with a quercitrin:isoquercitrin production ratio of 24:1. This suggests the preference of RhaGT for UDP-rhamnose when different UDP-sugar donors are present. Possible explanations for the significantly lower specific productivity (fivefold decrease) of sRHA2 as compared to sRHA3 are either a higher metabolic burden [52] caused by the two plasmid system or a too limited activity of the native GalU, which could be insufficient for adequate UDP-glc formation [45] .
To demonstrate the scalability of the developed bioprocess, strain sGAL3 was cultured in a 1-L bioreactor, which also ensures a constant pH set at 6.80 and avoid oxygen limitation. A detailed overview of the consumption of sucrose, growth and hyperoside production is given in Fig. 5 . After a lag-phase, the strain displayed a growth rate of 0.32 ± 0.02 h −1 while simultaneously producing hyperoside. The observed specific productivity (65.9 ± 2.6 mg/g CDW/h) was comparable to the one obtained on shake flask scale. When nearly all quercetin was converted, hyperoside formation slowed down, which can be explained either by the observed correlation between quercetin concentration and q p , or by the reported reversibility of F3GT [53] . It is likely that further improvements in titer and productivity can be realized by optimizing the supply of quercetin using a fed-batch system.
To the best of our knowledge, the results obtained in this study with the engineered sGAL and sRHA strains for the production of hyperoside and quercitrin are the highest reported to date both in terms of titer and production rate. The maximal production rate obtained in this contribution was 6 to 50-fold higher compared to the maximal production rates (r p,max ) of processes reported in the literature [47, 54] as is illustrated in Fig. 6 .
The increased performance, in terms of titer and productivity, obtained with the developed platform can be attributed to the use of a split metabolism in combination with optimally rerouting the flux from glucose 1-phosphate towards UDP-galactose and UDP-rhamnose. The undesired conversion of the activated sugars into biomass Fig. 3 Effect of the chromosomal integration locus of the knockin of BaSP on the growth rate. Strains were grown in shake flasks and the resulting maximal growth rates (µ max ) were compared with E. coli W ΔcscAR with plasmid-based BaSP expression (+pBaSP). Error bars represent standard deviations is impeded by gene deletions, which guarantees a high product yield. In addition, since biomass formation, which is fueled by the fructose moiety of sucrose, and glycoside synthesis go hand in hand and subsequently are performed at the same time at a high rate, a high productivity is equally guaranteed (one-step fermentation process).
Besides quercetin also other flavonols such as kaempferol, fisetin, morin and myricetin significantly contribute to our daily flavonoid intake, which also have extremely diverse beneficial effects [55, 56] . As the sugar moiety is a major determinant of the intestinal absorption of dietary flavonoids and their subsequent bioactivity [57, 58] , the To this end, strains sGAL3 and sRHA3 were grown in tubes with 5 mL minimal medium, each containing 1.5 g/L of either kaempferol, myricetin, morin or fisetin. Growth and production were monitored over 48 h and various spots were observed on TLC with similar retention factors as hyperoside and quercitrin. Mass spectrometry was used to identify the compounds produced, which confirmed the in vivo galactosylation of myricetin, kaempferol, morin and fisetin ( Table 2 ). All compounds occurred with an m/z of [M + 114], due to complexation with trifluoroacetic acid from the mobile phase. The galactoside of morin was produced at a slow rate, which is in accordance to the very low in vitro activity of F3GT towards this flavonol [53] . A possible explanation for this limited activity may be the presence of an unusual hydroxyl group at the 2′ position, which may sterically hinder deprotonation and consequent galactosylation of morin at hydroxyl group 3 [59] .
Incubation of sRHA3 with the different flavonols investigated showed two distinct glycoside spots on TLC, which corresponded to the 3-O-rhamnoside and 3-O-glucoside. Kaempferol proved to be the best substrate for RhaGT and was predominantly rhamnosylated (8:1 ratio), with a titer exceeding 400 mg/L, which is twofold higher than previously reported [47] . Fisetin on the other hand was efficiently glucosylated, yet the formation of its rhamnoside was not as efficient, with a titer below 5 mg/L. A similar preference towards glucoside formation was also observed with myricetin and morin, which indicates that the positioning of the hydroxyl groups is the determining factor for glycosylation with RhaGT. The production of the desired rhamnosides, galactosides or glucosides may be improved considerably by using UGTs that are more specific towards certain flavonols and UDP-sugars. Transformation of the corresponding UGTs in the developed in vivo glycosylation strains presents a promising alternative for the large-scale production of various flavonol glycoforms, which are to date mainly extracted from plant material.
On the other hand, due to the pivotal role of UDP-glc, various other UDP-sugars can be formed in vivo (e.g. UDP-glucuronate, UDP-xylose, UDP-arabinose). In combination with the modularity of the developed glycosylation platform, which permits rapid introduction of any UGT or UDP-sugar pathway, virtually any glycoside can be produced. Hence, this demonstrates that the proposed microbial platform is a robust, versatile and efficient microbial cell factory for the glycosylation (e.g. glucosylation, rhamnosylation, galactosylation) of small molecules. Although obtained productivities are the highest reported today and compete with the current production processes, further improvement can be limited due to solubility issues of the aglycon or of the glycoside. To this end follow-up research can focus on further metabolic engineering (e.g. introduction of the aglycon pathway allowing in vivo gradually production of the aglycon) or on process optimization [e.g. 2-phase (bilayer) fermentation which enables in situ recovery] to improve these issues.
In this contribution, a biotechnological platform was developed for the galactosylation and rhamnosylation of small molecules, such as secondary metabolite natural products, starting from a previously created glucosylation host. To this end, the routes to convert UDP-glucose into UDP-galactose and UDP-rhamnose were introduced by expressing a UDP-glucose epimerase (galE) and a UDP-rhamnose synthase (MUM4), respectively. As a proof of concept, the bio-active flavonol quercetin was selected for galactosylation and rhamnosylation, yielding hyperoside (quercetin 3-O-galactoside) and quercitrin (quercetin 3-O-rhamnoside), respectively. Next, the flavonol 3-O-galactosyltransferase (F3GT) from Petunia hybrida and the flavonol 3-O-rhamnosyltransferase from Arabidopsis thaliana (RhaGT) were overexpressed in the metabolically engineered E. coli W mutants. The strains created were able to produce 940 mg/L of hyperoside and 1176 mg/L of quercitrin at specific production rates of 68.7 mg/g CDW/h and 47.8 mg/g CDW/h, respectively, which are the highest reported to date. Interestingly, both GTs showed in vivo activity towards other dietary flavonols, whereby for example over 400 mg/L of kaempferol 3-O-rhamnoside could be formed extracellularly.
All plasmids used were constructed using Gibson assembly [60] or CLIVA [61] . All PCR fragments were amplified using Q5 polymerase from New England Biolabs (Ipswich, Massachusetts). Oligonucleotides were purchased from IDT (Leuven, Belgium). The plasmids and bacterial strains used in this study are listed in Table 1 . A list of primers for the creation of gene knockouts/knockins and for the cloning of the expression plasmids is given in Additional file 2: Table S1 . E. coli DH5α was used for plasmid cloning and propagation, while E. coli W was used for expression of the production plasmids and the creation of gene knockouts and knockins. Hyperoside, quercitrin, isoquercitrin, kaempferol and myricetin were purchased from Carbosynth (Berkshire, UK). All other chemicals used were purchased from Sigma Aldrich (Germany) unless otherwise indicated.
The expression plasmids for the prod uction of hyperoside and quercitrin were constructed as depicted in Additional file 3: Figure S1A Figure S1D ). The galE [Genbank: JW0742] and galE2 [Genbank: KJ543703] sequences were amplified from the genomic DNA of E. coli and Bifidobacterium bifidum, respectively. CLIVA assembly resulted in the intermediary plasmid pBaSP/F3GT/UgpA ( Figure S1A ), which was subsequently used for the amplification of the F3GT/ UgpA backbone. Gibson assembly of the GalE or GalE2 inserts with this backbone resulted in the final galactosylation plasmids pGalE/F3GT/UgpA and pGalE2/F3GT/ UgpA, respectively ( Figure S1B ). Similarly, MUM4 and RhaGT were introduced using a 3-pieces Gibson assembly ( Figure S1C ), resulting in the final rhamnosylation plasmid pMUM4/RhaGT/UgpA.
The overall E. coli W knockout mutants were created using the one step deletion system of Datsenko and Wanner [62] . The strategy for chromosomal integration of BaSP under control of the constitutive promoter P22 flanked by L4 and L5 at the melA and glgC loci is depicted and explained in Additional file 1: Figure S2 . Transformants were plated on minimal sucrose medium agar plates and grown overnight for screening. The in-house strain E. coli W ΔcscAR Δpgm Δagp ΔushA ΔlacZYA::P22-lacY ΔgalETKM [45] was used for the chromosomal integration of L4-P22-BaSP-L5 at the melA site, yielding the base strain sGLYC (Table 1) . This strain and the E. coli W wild type were transformed with the production plasmids described above, resulting in the galactosylation (sGAL) and rhamnosylation (sRHA) strains given in Table 1 .
Composition of LB and minimal sucrose medium was described previously [45] . Minimal medium agar plates with sucrose (50 g/L) had the same composition as minimal sucrose medium, but contained additionally 15 g/L of agarose. The agarose and salts were autoclaved separately at 121 °C for 21 min. Sucrose was filter sterilized through a 0.22 µm corning filter (Fisher, Belgium) and heated for 1 min in a microwave oven at 800 W prior to mixing it with the warm agarose and salt solutions. 1 mL/L of mineral solution [45] was sterilely added prior to pouring the plates.
Escherichia coli W mutant precultures were grown in 5 mL LB medium with the antibiotics (50 μg/mL kanamycin or carbenicillin) required for maintenance and selection of the plasmids. The cultures were grown for 16 h at 37 °C and 200 rpm and used for the 2 % inoculation of 100 mL minimal sucrose medium in 500 mL shake flasks. For the production of hyperoside and quercitrin, quercetin was added to the minimal medium at a concentration of 0.15 or 1.5 g/L. Growth conditions were the same as previously described [45] . Samples were taken at regular intervals from the broth and, after centrifugation, the supernatant was used for the analysis and quantification of sugars. For the analysis of quercetin and its glycosides, 200 µL of the culture was collected and extracted with 800 µL ethyl acetate. The organic layer was collected, evaporated in a SpeedVac ™ vacuum concentrator (Thermo Fisher, USA) and dissolved in 200 µL of DMSO for HPLC quantification.
The bioreactor set-up and fermentation conditions used are the same as previously described [45] . Production experiments were performed on minimal sucrose medium without MOPS buffer and with the addition of quercetin as acceptor.
Culture samples were primarily analyzed by TLC on Silica gel 60 F 254 precoated plates (Merck, Germany). All plates were run in a closed TLC chamber and developed using standard visualization techniques and agents: UV fluorescence (254 nm) or by staining with 10 % (v/v) H 2 SO 4 and subsequent charring. The mobile phase for detecting the various flavonols and corresponding glycosides consisted of an ethyl acetate:acetic acid:formic acid:water (100:11:11:27 v/v) mixture [63] . Product spot intensities of other flavonol glycosides were processed and quantified using ImageJ [64] . HPLC quantification of sucrose, fructose and glucose was performed using an X-bridge Amide column (35 μm, Waters, USA) as described previously [45] . Quercetin, hyperoside, quercitrin and isoquercitrin were detected with the method described by Pandey et al. [41] using a Varian HPLC system (Agilent technologies, California). Mass spectrometry for determination of the various flavonol glycosides was performed with a Micromass Quattro LC (McKinley Scientific, USA). Detection was performed in negative mode ESI-224 MS with a capillary voltage of 2.53 kV, a cone voltage of 20 V, cone and desolvation gas flows of 93 and 420 L/h, and source and cone temperatures of 150 and 350 °C, respectively.
Quercetin glycosides were extracted from the broth with an equal volume of ethyl acetate after which the organic layer was evaporated to dryness. The remaining product was dissolved in the solvent system described above and run on a preparative TLC plate. The band containing hyperoside (R f 0.53) or quercitrin (R f 0.75) was scraped off, extracted with ethyl acetate and evaporated to yield a bright yellow powder. Products were confirmed by NMR. Spectra were reported elsewhere [47, 65] .
|
What characteristics does glycosylation have on flavonoids?
| 5,300 |
solubility, stability and bio-activity
| 3,403 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the relationship between urbanization and risk of emergence of flu-like diseases?
| 585 |
findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk.
| 1,274 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What factors and characteristics of semi-urban landscapes promote viral transmission?
| 586 |
higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission
| 1,505 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the relationship between HIN1 viral transmission and poultry production.
| 587 |
landscapes where intensive and extensive forms of poultry production overlap were found at greater risk
| 1,877 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the principle behind infection Convergence Model ?
| 588 |
The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
| 4,614 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the Boosted Regression Tree method?
| 590 |
BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model
| 25,944 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the advantage of Boosted Regression Tree method?
| 591 |
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance.
| 26,125 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the relationship between land use and emergence of HPAI H5N1?
| 592 |
high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks
| 34,261 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
Where is the highest risk of HPAI H5N1 like disease emergence?
| 589 |
Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely.
| 5,073 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
How does land use fragmentation increase the risk of flu-like diseases?
| 594 |
Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts
| 11,013 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the relationship between the outbreak of HPAI H5N1 like diseases and rice cultivation?
| 596 |
extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers
| 16,923 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the relationship between aquaculture and spread of H5N1 like diseases?
| 597 |
extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus
| 17,269 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the relationship between proximity ofwater bodies to agricultural lands and spread of H5N1 like diseases?
| 598 |
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry.
| 17,475 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is the effect of diversity of chicken flock on H5N1 disease?
| 599 |
diversity of chicken flock-size had a strong association with HPAI H5N1
| 31,106 |
1,628 |
Evidence for the Convergence Model: The Emergence of Highly Pathogenic Avian Influenza (H5N1) in Viet Nam
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580613/
SHA: ee5b43d20a640664510cb7a540caaae4a8e19933
Authors: Saksena, Sumeet; Fox, Jefferson; Epprecht, Michael; Tran, Chinh C.; Nong, Duong H.; Spencer, James H.; Nguyen, Lam; Finucane, Melissa L.; Tran, Vien D.; Wilcox, Bruce A.
Date: 2015-09-23
DOI: 10.1371/journal.pone.0138138
License: cc-by
Abstract: Building on a series of ground breaking reviews that first defined and drew attention to emerging infectious diseases (EID), the ‘convergence model’ was proposed to explain the multifactorial causality of disease emergence. The model broadly hypothesizes disease emergence is driven by the co-incidence of genetic, physical environmental, ecological, and social factors. We developed and tested a model of the emergence of highly pathogenic avian influenza (HPAI) H5N1 based on suspected convergence factors that are mainly associated with land-use change. Building on previous geospatial statistical studies that identified natural and human risk factors associated with urbanization, we added new factors to test whether causal mechanisms and pathogenic landscapes could be more specifically identified. Our findings suggest that urbanization spatially combines risk factors to produce particular types of peri-urban landscapes with significantly higher HPAI H5N1 emergence risk. The work highlights that peri-urban areas of Viet Nam have higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture than rural and urban areas. We also found that land-use diversity, a surrogate measure for potential mixing of host populations and other factors that likely influence viral transmission, significantly improves the model’s predictability. Similarly, landscapes where intensive and extensive forms of poultry production overlap were found at greater risk. These results support the convergence hypothesis in general and demonstrate the potential to improve EID prevention and control by combing geospatial monitoring of these factors along with pathogen surveillance programs.
Text: Two decades after the Institute of Medicine's seminal report [1] recognized novel and reemerging diseases as a new category of microbial threats, the perpetual and unexpected nature of the emergence of infectious diseases remains a challenge in spite of significant clinical and biomedical research advances [2] . Highly Pathogenic Avian Influenza (HPAI) (subtype H5N1) is the most significant newly emerging pandemic disease since HIV/AIDS. Its eruption in Southeast Asia in 2003-4 and subsequent spread globally to more than 60 countries fits the complex systems definition of "surprise" [3] . In this same year that IOM had published its final report on microbial threats which highlighted H5N1's successful containment in Hong Kong in 1997 [4] , massive outbreaks occurred in Southeast Asia where it remains endemic, along with Egypt's Nile Delta. Since 2003, HPAI H5N1 has killed millions of poultry in countries throughout Asia, Europe, and Africa, and 402 humans have died from it in sixteen countries according to WHO data as of January 2015. The threat of a pandemic resulting in millions of human cases worldwide remains a possibility [5] .
Lederberg et al. [1] first pointed to the multiplicity of factors driving disease emergence, which later were elaborated and described in terms of 'the convergence model' [6] . The model proposes emergence events are precipitated by the intensifying of biological, environmental, ecological, and socioeconomic drivers. Microbial "adaptation and change," along with "changing ecosystems" and "economic development and land use" form major themes. Joshua Lederberg, the major intellectual force behind the studies summed-up saying "Ecological instabilities arise from the ways we alter the physical and biological environment, the microbial and animal tenants (humans included) of these environments, and our interactions (including hygienic and therapeutic interventions) with the parasites" [6] .
Combining such disparate factors and associated concepts from biomedicine, ecology, and social sciences in a single framework remains elusive. One approach suggested has been to employ social-ecological systems theory that attempts to capture the behavior of so-called 'coupled natural-human systems', including the inevitable unexpected appearance of new diseases, themselves one of the "emerging properties" of complex adaptive systems (CAS) [7, 8] . The convergence model can be so adapted by incorporating the dynamics of urban, agricultural, and natural ecosystem transformations proposed with this framework. These associated multifaceted interactions including feedbacks that affect ecological communities, hosts and pathogen populations, are the proximate drivers of disease emergence.
The initial HPAI H5N1 outbreaks in Vietnam represent an ideal opportunity to adapt and test a CAS-convergence model. Emergence risk should be highest in the most rapidly transforming urban areas, peri-urban zones where mixes of urban-rural, modern-traditional land uses and poultry husbandry coincide most intensely. Specifically we hypothesized a positive association between the presence of HPAI outbreaks in poultry at the commune level and: 1) peri-urban areas, as defined by Saksena et al. [9] , 2) land-use diversity, and 3) co-location of intensive and extensive systems of poultry.
We used the presence or absence at the commune level of HPAI H5N1 outbreaks in poultry as the dependent variable. Vietnam experienced its first HPAI H5N1 outbreak in late 2003, since then, there have been five waves and sporadic outbreaks recorded over the years [10, 11] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. We used data from the Viet Nam 2006 Agricultural Census to develop an urbanicity classification that used data collected at a single point in time (2006) but across space (10,820 communes) to infer processes of change (urbanization, land-use diversification, and poultry intensification) [9] . The 58 provinces in Vietnam (not counting the 5 urban provinces that are governed centrally) are divided into rural districts, provincial towns, and provincial cities. Rural districts are further divided into communes (rural areas) and towns, and provincial towns and cities are divided into wards (urban subdistricts) and communes. A commune in Viet Nam is thus the third level administrative subdivision, consisting of villages/hamlets. For the purpose of simplicity we will henceforth use the term "commune" to refer to the smallest administrative unit whether it is a commune, town, or ward. We included risk factors documented in previous work. We also aimed to understand the differences, if any, in risk dynamics at different scales; comparing risks at the national scale to those at two sub-national agro-ecological zones. For this purpose we chose to study the Red River and Mekong River deltas, well known hot spots of the disease. Hence we conducted two sets of analyses (waves 1 and 2) for three places (nation, Red River Delta, and Mekong Delta) producing a total of 6 wave-place analyses. Data on outbreaks were obtained from the publicly available database of Viet Nam's Department of Animal Health. Given the highly complex dynamics of the epidemics and in keeping with recent methodological trends, we used multiple modeling approaches-parametric and non-parametric-with a focus on spatial analysis. We used both 'place' oriented models that can take into account variations in factors such as policies and administration as well as 'space' oriented models that recognize the importance of physical proximity in natural phenomenon [12] .
Very few empirical studies have attempted to determine whether urbanization is related to EID outbreaks or whether urbanization is associated primarily with other factors related to EID outbreaks. One immediate problem researchers face is defining what is rural, urban, and transitional (i.e., peri-urban). Some studies have used official administrative definitions of urban and rural areas, but this approach is limited in its bluntness [13] . Other studies prioritized human population density as a satisfactory surrogate [11, [14] [15] [16] [17] [18] [19] [20] , but this approach ignores the important fact that density is not a risk factor if it is accompanied by sufficient infrastructure to handle the population. Spencer [21] examined urbanization as a non-linear characteristic, using household-level variables such as water and sanitation services. He found evidence that increased diversity in water supply sources and sanitation infrastructure were associated with higher incidences of HPAI. These studies employed a limited definition of urbanization that lacked a well-defined characterization of peri-urbanization.
Still other studies have mapped the relative urban nature of a place, a broad concept that is often referred to as 'urbanicity' [22] [23] [24] [25] . While these studies show differences in the rural/ urban nature of communities across space and time, they have been limited to small-to medium-scale observational studies; and they have failed to distinguish between different levels of "ruralness". Perhaps the best known model of peri-urbanization is McGee's concept of desakota (Indonesian for "village-town") [26] . McGee identified six characteristics of desakota regions: 1) a large population of smallholder cultivators; 2) an increase in non-agricultural activities; 3) extreme fluidity and mobility of population; 4) a mixture of land uses, agriculture, cottage industries, suburban development; 5) increased participation of the female labor force; and 6) "grey-zones", where informal and illegal activities group [26] . Saksena et al. [9] built on McGee's desakota concepts and data from the 2006 Viet Nam Agricultural Census to establish an urbanicity classification. That study identified and mapped the 10,820 communes, the smallest administrative unit for which data are collected, as being rural, peri-urban, urban, or urban core. This project used the Saksena classification to assess associations between urbanicity classes, other risks factors, and HPAI outbreaks.
Researchers have estimated that almost 75% of zoonotic diseases are associated with landcover and land-use changes (LCLUC) [27, 28] . LCLUC such as peri-urbanization and agricultural diversification frequently result in more diverse and fragmented landscapes (number of land covers or land uses per unit of land). The importance of landscape pattern, including diversity and associated processes, which equate to host species' habitat size and distribution, and thus pathogen transmission dynamics is axiomatic though the specific mechanisms depend on the disease [29, 30] . Landscape fragmentation produces ecotones, defined as abrupt edges or transitions zones between different ecological systems, thought to facilitate disease emergence by increasing the intensity and frequency of contact between host species [31] Furthermore, fragmentation of natural habitat tends to interrupt and degrade natural processes, including interspecies interactions that regulate densities of otherwise opportunistic species that may serve as competent hosts [32] , although it is not clear if reduced species diversity necessarily increases pathogen transmission [33] . Rarely has research connected land-use diversification to final health endpoints in humans or livestock; this study attempts to link land-use diversity with HPAI H5N1 outbreaks.
Human populations in the rapidly urbanizing cities of the developing world require access to vegetables, fruits, meat, etc. typically produced elsewhere. As theorized by von Thünen in 1826 [34] , much of this demand is met by farms near cities [35] , many in areas undergoing processes of peri-urbanization [26] . Due to the globalization of poultry trade, large-scale chicken farms raising thousands of birds have expanded rapidly in Southeast Asia and compete with existing small backyard farmers [36] . Large, enterprise-scale (15,000-100,000 birds) operations are still rare in Viet Nam (only 33 communes have such a facility). On the other hand, domestic and multinational companies frequently contract farmers to raise between 2,000 and 15,000 birds.
Recent studies have examined the relative role of extensive (backyard) systems and intensive systems [15, [17] [18] [19] 37] . In much of Asia there is often a mix of commercial and backyard farming at any one location [36] . Experts have suggested that from a biosecurity perspective the co-location of extensive and intensive systems is a potential risk factor [38] . Intensive systems allow for virus evolution (e.g. Low Pathogenic Avian Influenza to HPAI) and transformation, while extensive systems allow for environmental persistence and circulation [39] . Previous studies of chicken populations as a risk factor have distinguished between production systems-native chickens, backyard chickens; flock density; commercial chickens, broilers and layers density, etc. [15, [17] [18] [19] 37] . In isolation, however, none of these number and/or density based poultry metrics adequately measures the extent of co-location of intensive and extensive systems in any given place. Intensive and extensive systems in Viet Nam have their own fairly well defined flock sizes. A diversity index of the relative number of intensive and extensive systems of poultry-raising can better estimate the effect of such co-location; this study attempts to link a livestock diversity index with the presence or absence of HPAI H5N1 outbreaks at the commune level.
This study investigated for the 10,820 communes of Viet Nam a wide suite of socio-economic, agricultural, climatic and ecological variables relevant to poultry management and the transmission and persistence of the HPAI virus. Many of these variables were identified based on earlier studies of HPAI (as reviewed in Gilbert and Pfeiffer [40] ). Three novel variables were included based on hypotheses generated by this project. All variables were measured or aggregated to the commune level. The novel variables were:
• Degree of urbanization: We used the urbanicity classification developed by Saksena et al. [9] to define the urban character of each commune. The classification framework is based on four characteristics: 1) percentage of households whose main income is from agriculture, aquaculture and forestry, 2) percentage of households with modern forms of toilets, 3) percentage of land under agriculture, aquaculture and forestry and 4) the Normalized Differentiated Vegetation Index (NDVI). The three-way classification enabled testing for non-linear and non-monotonous responses.
• Land-use diversity: We measured land-use diversity using the Gini-Simpson Diversity Index [41] . The Gini-Simpson Diversity Index is given by 1-λ, where λ equals the probability that two entities taken at random from the dataset of interest represent the same type. In situations with only one class (complete homogeneity) the Gini-Simpson index would have a value equal to zero. Such diversity indices have been used to measure land-use diversity [42] . We used the following five land-use classes: annual crops, perennial crops, forests, aquaculture and built-up land (including miscellaneous uses) for which data were collected in the 2006 Agricultural Census. The area under the last class was calculated as the difference between the total area and the sum of the first four classes.
The following variables are listed according to their role in disease introduction, transmission and persistence, though some of these factors may have multiple roles.
• Human population related transmission.
Human population density [11, 14-16, 18, 19, 44, 45] .
• Poultry trade and market.
Towns and cities were assumed to be active trading places [10, 18, 37, 44, 46] . So, the distance to the nearest town/city was used as indicator of poultry trade.
Trade is facilitated by access to transportation infrastructure [37, 47, 48] . So, the distance to the nearest a) national highway and b) provincial highway was used as indicator of transportation infrastructure.
• Disease introduction and amplification.
The densities of chicken were calculated based on commune area [15, 19, 37, 49] .
• Intermediate hosts.
Duck and geese densities were calculated using total commune area [11, 19, 49] .
As previous studies have shown a link between scavenging in rice fields by ducks and outbreaks, we also calculated duck density using only the area under rice.
• Agro-ecological and environmental risk factors.
Previous studies have shown that the extent of rice cultivation is a risk factor, mainly due its association with free ranging ducks acting as scavengers [10] . We used percentage of land under rice cultivation as a measure of extent.
Rice cropping intensity is also a known risk factor [11, 17, 37] . We used the mean number of rice crops per year as a measure of intensity.
The extent of aquaculture is a known risk factor [10] , possibly because water bodies offer routes for transmission and persistence of the virus. The percentage of land under aquaculture was used as a metric.
Proximity to water bodies increases the risk of outbreaks [47, [50] [51] [52] , possibly by increasing the chance of contact between wild water birds and domestic poultry. We measured the distance between the commune and the nearest: a) lake and b) river.
Climatic variables-annual mean temperature and annual precipitation-have been associated with significant changes in risk [48, 53] .
Elevation, which is associated with types of land cover and agriculture, has been shown to be a significant risk factor in Vietnam [10] .
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. In the absence of reliable and inexpensive data on the extent of surface water we used CTI as a proxy. CTI has been used in Ecological Niche Models (ENM) of HPAI H5N1 [55, 56] . However, given the nature of ENM studies, the effect of CTI as a risk factor has been unknown so far. CTI has been used as a risk factor in the study of other infectious and non-infectious diseases [57] . Some studies have shown that at local scales, the slope of the terrain (a component of CTI) was significantly correlated with reservoir species dominance [58] . CTI is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. CTI is computed as follows: CTI = ln (A s / (tan (β)) where; A s = Area Value calculated as ((flow accumulation + 1) Ã (pixel area in m 2 )) and β is the slope expressed in radians [59] .
Though previous studies have indicated that Normalized Difference Vegetation Index (NDVI) is a risk factor [10, 20, 55, 60, 61], we did not include it explicitly in our models, as the urban classification index we used included NDVI [9] .
We obtained commune level data on HPAI H5N1 outbreaks from the publicly available database of the Department of Animal Health [10] . Viet Nam experienced its first major epidemic waves between December 2003 and February 2006 [10] . We chose to study the first wave (Wave 1) that ended in February 2004 and the second wave (Wave 2) that occurred between December 2004 and April 2005. In Wave 1, 21% of the communes and in Wave 2, 6% of the communes experienced outbreaks. We used data from the 1999 Population Census of Viet Nam to estimate human population per commune. We relied on data from two Agriculture Censuses of Viet Nam. This survey is conducted every five years covering all rural households and those peri-urban households that own farms. Thus about three-fourths of all of the country's households are included. The contents of the survey include number of households in major production activities, population, labor classified by sex, age, qualification, employment and major income source; agriculture, forestry and aquaculture land used by households classified by source, type, cultivation area for by crop type; and farming equipment by purpose. Commune level surveys include information on rural infrastructure, namely electricity, transportation, medical stations, schools; fresh water source, communication, markets, etc. Detailed economic data are collected for large farms. We used the 2006 Agriculture Census for most variables because the first three epidemic waves occurred between the Agricultural Censuses of 2001 and 2006 but were closer in time to the 2006 census [10] . However, for data on poultry numbers we used the 2001 Agriculture Census data set because between 1991 and 2003 the poultry population grew at an average rate of 7% annually. However, in 2004, after the first wave of the H5N1 epidemic, the poultry population fell 15%. Only by mid-2008 did the poultry population return close to pre-epidemic levels. Thus, we considered the poultry population data from the 2001 census to be more representative. We aggregated census household data to the commune level. A three-way classification of the rural-to-urban transition was based on a related study [9] .
Raster data on annual mean temperature and precipitation were obtained from the World-Clim database and converted to commune level data. The bioclimatic variables were compiled from the monthly temperature and precipitation values and interpolated to surfaces at 90m spatial resolution [62] . This public database provides data on the average climatic conditions of the period 1950-2000.
Elevation was generated from SRTM 90 meter Digital Elevation Models (DEM) acquired from the Consortium for Spatial Information (CGIAR-CSI). Compound Topographical Index (CTI) data were generated using the Geomorphometry and Gradient Metrics Toolbox for Arc-GIS 10.1.
Prior to risk factor analysis we cleaned the data by identifying illogical values for all variables and then either assigning a missing value to them or adjusting the values. Illogical values occurred mainly (less than 1% of the cases) for land-related variables such as percentage of commune land under a particular type of land use. Next we tested each variable for normality using the BestFit software (Palisade Corporation). Most of the variables were found to follow a log-normal distribution and a log-transform was used on them. We then examined the bi-variate correlations between all the risk factors (or their log-transform, as the case may be). Correlations were analyzed separately for each place. Certain risk factors were then eliminated from consideration when |r| ! 0.5 (r is the Pearson correlation coefficient). When two risk factors were highly correlated, we chose to include the one which had not been adequately studied explicitly in previously published risk models. Notably, we excluded a) elevation (correlated with human population density, chicken density, duck density, percentage land under paddy, annual temperature and compound topographical index), b) human population density (correlated with elevation and CTI), c) chicken density (only at national level, correlated with CTI), d) duck and goose density (correlated with elevation, chicken density, percentage land under paddy, land use diversity index and CTI), e) annual temperature (correlated with elevation and CTI) and f) cropping intensity (correlated with percentage land under paddy).
Considering the importance of spatial autocorrelation in such epidemics, we used two modeling approaches: 1) multi-level Generalized Linear Mixed Model (GLMM) and 2) Boosted Regression trees (BRT) [63, 64] with an autoregressive term [65] . GLMM is a 'place' oriented approach that is well suited to analyzing the effect of administrative groupings, while BRT is a 'space' oriented approach that accounts for the effects of physical proximity. We began by deriving an autoregressive term by averaging the presence/absence among a set of neighbors defined by the limit of autocorrelation, weighted by the inverse of the Euclidean distance [65] .
The limit of the autocorrelation of the response variable was obtained from the range of the spatial correlogram ρ (h) [66] . To determine which predictor variables to include in the two models, we conducted logistic regression modeling separately for each of them one by one but included the autoregressive term each time. We finally included only those variables whose coefficient had a significance value p 0.2 (in at least one wave-place combination) and we noted the sign of the coefficient. This choice of p value for screening risk factors is common in similar studies [15, 18, 45, 67] . We used a two-level GLMM (communes nested under districts) to take account of random effects for an area influenced by its neighbors, and thus, we studied the effect of spatial autocorrelation. We used robust standard errors for tests of fixed effects. Boosted regression trees, also known as stochastic gradient boosting, was performed to predict the probability of HPAI H5N1 occurrence and determine the relative influence of each risk factor to the HPAI H5N1 occurrence. This method was developed recently and applied widely for distribution prediction in various fields of ecology [63, 64] . It is widely used for species distribution modeling where only the sites of occurrence of the species are known [68] . The method has been applied in numerous studies for predicting the distribution of HPAI H5N1 disease [16, 51, [69] [70] [71] . BRT utilizes regression trees and boosting algorithms to fit several models and combines them for improving prediction by performing iterative loop throughout the model [63, 64] .
The advantage of BRT is that it applies stochastic processes that include probabilistic components to improve predictive performance. We used regression trees to select relevant predictor variables and boosting to improve accuracy in a single tree. The sequential process allows trees to be fitted iteratively through a forward stage-wise procedure in the boosting model. Two important parameters specified in the BRT model are learning rate (lr) and tree complexity (tc) to determine the number of trees for optimal prediction [63, 64] . In our model we used 10 sets of training and test points for cross-validation, a tree complexity of 5, a learning rate of 0.01, and a bag fraction of 0.5. Other advantages of BRT include its insensitivity to co-linearity and non-linear responses. However, for the sake of consistency with the GLMM method, we chose to eliminate predictors that were highly correlated with other predictors and to make log-transforms where needed. In the GLMM models we used p 0.05 to identify significant risk factors.
The predictive performances of the models were assessed by the area under the curve (AUC) of the receiver operation characteristic (ROC) curve. AUC is a measure of the overall fit of the model that varies from 0.5 (chance event) to 1.0 (perfect fit) [72] . A comparison of AUC with other accuracy metrics concluded that it is the most robust measure of model performance because it remained constant over a wide range of prevalence rates [73] . We used the corrected Akaike Information Criteria (AICc) to compare each GLMM model with and without its respective suite of fixed predictors.
We used SPSS version 21 (IBM Corp., New York, 2012) for GLMM and R version 3.1.0 (The R Foundation for Statistical Computing, 2014) for the BRT. For calculating the spatial correlogram we used the spdep package of R.
The fourteen predictor variables we modeled (see tables) were all found to be significantly associated with HPAI H5N1 outbreaks (p 0.2) in at least one wave-place combination based on univariate analysis (but including the autoregressive term) ( Table 1) . Land-use diversity, chicken density, poultry flock size diversity and distance to national highway were found to have significant associations across five of the six wave-place combinations.
power of the GLMM models, as measured by the AUC, is very good with AUC values ranging from 0.802 to 0.952 (Tables 2-7 ). The predictive power of the national models was higher than that of the delta models. The predictive power of the BRT models is good, with AUCs ranging from 0.737 to 0.914. The BRT models also had a better predictive power at the national level than at the delta level. These values are higher than those reported for Wave 1 (AUC = 0.69) and Wave 2 (AUC = 0.77) by Gilbert et al. [11] . Both Gilbert et al. [11] and this study found that at the national level the predictive performance for Wave 2 was higher than that for Wave 1. Wave 2 mainly affected the Mekong River Delta. Previous studies indicated the duck density was an important predictor [11] ; our results, however, indicated that the diversity of duck flock size was a more important predictor than duck density.
Both the GLMM and BRT models found annual precipitation to be a significant factor. The GLMM model indicated a negative association; similar to what was found by studies in China [51] and in the Red River Delta [53] . A global study of human cases also found occurrence to be higher under drier conditions [74] . Generally, the role of precipitation was found to be far more significant in the deltas than for the country as a whole.
The unadjusted Relative Risk (RR) of peri-urban areas in comparison with non-peri-urban areas was 1.41 and 1.60 for Waves 1 and 2, respectively. In terms of urbanicity, we found that chicken density, percentage of land under rice, percentage of land under aquaculture, flock size diversity for duck and geese, and the Compound Topographical Index (CTI) to be highest in peri-urban areas (Fig 1a-1e) . We also found that land-use diversity was higher in rural areas, but peri-urban areas had diversity levels only marginally lower (Fig 1f) . The urbanicity variable alone, however, was not found to be significantly associated with HPAI H5N1 in any place according to the GLMM model except for the urban level in Red River Delta for Wave 2 and in the Mekong River Delta for Wave 1. The BRT model ranked urbanicity as one of the least influential variables. Land-use diversity was found to be significantly associated with HPAI H5N1 in both waves for Viet Nam according to the GLMM model, but at the delta level the association was significant only for Wave 2 in the Mekong River Delta. The BRT model indicated that land-use diversity highly influenced HPAI H5N1 at the national level in Wave 2. For the remaining waveplace combinations land-use diversity had middle to below-middle rank of influence.
Both the GLMM and BRT models indicated that the diversity of chicken flock-size had a strong association with HPAI H5N1 for both waves at the national level. This was generally found to be true at the delta levels with some exceptions. The diversity of duck and goose flock size was also significantly associated with HPAI H5N1 in all places, but the associations were much stronger in Wave 2 than in Wave 1.
The GLMM model indicated that the CTI had a very strong association with HPAI H5N1 at the national level in both waves although this was not true in the two deltas. The CTI is a steady state wetness index commonly used to quantify topographic control on hydrological processes. Accumulation numbers in flat areas, like deltas, are very large; hence the CTI was not a relevant variable in the GLMM model in these areas. The BRT model however indicated that CTI had middle to low influence in all waves and places. We found very high spatial clustering effects as indicated by the fact that in all waves and places the BRT model found the spatial autocorrelation term to have the highest rank of influence. As expected, the relative influence of the autocorrelation term at the national level was higher (60-78%) than at the delta levels (14-35%). In the GLMM models we found the Akaike Information Criterion (AIC) using the entire set of 14 variables to be much lower than the AICs of a GLMM model without fixed effects. This indicated that though clustering effects were significant, our theory driven predictor variables improved model performance.
A limitation of using surveillance methods for the dependent variable (poultry outbreaks) is that the data may have reporting/detection biases [11] . Under-reporting/detection in rural areas as compared to peri-urban areas is possible. We believe that the urbanicity and the shortest distance to nearest town risk factors serve as rough proxies for reporting/detection efficiency. Previous studies have tended to use human population density as a proxy for this purpose. In our study we found a strong association between human population density and urbanicity. But we acknowledge that a categorical variable such as urbanicity may provide less sensitivity than a continuous variable such as human population density in this specific context.
This study explored the validity of a general model for disease emergence that combined the IOM 'convergence model' [6] and the social-ecological systems model [7, 8] , for investigating the specific case of HPAI in Vietnam. We sought to test the hypotheses that measures of urbanization, land-use diversification, and poultry intensification are correlated with outbreaks in poultry. Our results generally support the hypothesis that social-ecological system transformations are associated with H5NI outbreaks in poultry.
The results presented here highlight three main findings: 1) when relevant risk factors are taken into account, urbanization is generally not a significant independent risk factor; but in peri-urban landscapes emergence factors converge, including higher levels of chicken densities, duck and geese flock size diversities, and fraction of land under rice or aquaculture; 2) high land-use diversity landscapes, a variable not previously considered in spatial studies of HPAI H5N1, are at significantly greater risk for HPAI H5N1 outbreaks; as are 3) landscapes where intensive and extensive forms of poultry production are co-located.
Only one other study has explicitly examined urbanicity in the context of HPAI H5N1. Loth et al. [17] found peri-urban areas in Indonesia were significantly associated with HPAI H5N1 cases, even based on multivariate models. Our study, however, attempted both to associate HPAI H5N1 with degree of urbanicity and to determine the features of peri-urban areas that place them at risk. When those features (i.e., chicken densities, duck and geese flock size diversities, and the fraction of land under rice or aquaculture) are included in multivariate models, the role of the urbanization variable per se diminishes. We found in the main river deltas in Viet Nam (Red River and Mekong), urbanization had no significant association with HPAI H5N1. This may be due to the fact that the deltas are more homogenous, in terms of urbanization, than the country as a whole. This is the first study to examine land-use diversity as a risk factor for HPAI H5N1. Measured by the Gini-Simpson Diversity Index of the five land-use classes on which data were collected in the 2006 Viet Nam Agricultural Census, and the presence or absence of HPAI outbreaks at the commune level, our results indicate a strong association between land-use diversity and HPAI H5N1 at the national level and in the Mekong River Delta. This metric captures both the variety of habitats and of the complexity of geospatial patterning likely associated with transmission intensity. Our results are similar to what has been observed by studies of other EIDs using fragmentation metrics (e.g. [75] [76] [77] . This is one of the few studies, however, to link landscape fragmentation to an EID disease in poultry and not just to the vector and/or hosts of the EID.
Previous studies have focused on poultry production factors such as type of species, size of flocks, and extent of commercialization (e.g. [15, [17] [18] [19] . This study expands on those findings by providing evidence that when intensive and extensive systems of chicken and/or duck and geese production co-exist in the same commune, the commune experiences higher risk of disease outbreak. Future studies need to examine the biological causal mechanisms in this context.
We suggest that national census data (particularly agricultural censuses) compiled at local levels of administration provide valuable information that are not available from remotely sensed data (such as poultry densities) or require a large amount of labor to map at national to larger scales (land-use diversity). Mapping land-use classes at the national scale for local administrative units (i.e., the 10,820 communes in Viet Nam) is not an insignificant task. Future studies, however, could examine the correlation between a census-based metric with metrics derived from remote sensing used to measure proportional abundance of each landcover type within a landscape [78] . Vietnam is relatively advanced in making digital national population and agricultural census data available in a format that can be linked to administrative boundaries. While other nations are beginning to develop similar capacities, in the short term the application of this method to other countries may be limited. Ultimately, both census and remotely sensed data can be used independently to map the urban transition and diversity of land use; these tools, however, may provide their greatest insights when used together.
Another important contribution of this study was the discovery of the importance of CTI. So far CTI had been used only in ecological niche modeling studies of HPAI H5N1; the specific role and direction of influence of CTI had has so far been unknown. Our study, the first to use CTI as a risk factor, found it had a large positive influence on HPAI H5N1 risk at the national level. Previous studies have highlighted the role of surface water extent in the persistence and transmission of the HPAI H5N1 virus. These studies measured surface water extent as area covered by water, magnitude of seasonal flooding, distance to the nearest body of water, or other variables that are often difficult to map using remotely sensed data, especially for large area studies. CTI on the other hand has the potential to serve as an excellent surrogate which can easily be measured in a GIS database. The national and regional (delta) models differed quite considerably, both in terms of performance and significant risk factors. In the deltas we commonly found only chicken density, duck flock size diversity and annual precipitation to be significant. This suggests dynamics of risk at the commune level are strongly dependent on the spatial range of analysis, consistent with another study in the Mekong Delta [61] . Though that study's model initially included three dozen commonly known risk factors, the significant risk factors were limited to poultry flock density, proportion households with electricity, re-scaled NDVI median May-October, buffalo density and sweet potato yield. Another study in the Red River Delta [79] found that in addition to the typical poultry density metrics, only the presence of poultry traders was significant. We speculate that for smaller regions, especially for known hot-spots, the relevant risk factors are those that reflect short-range, short-term driving forces such as poultry trading, presence of live bird markets and wet markets etc. Improving model performance for smaller regions would require highly refined and nuanced metrics for poultry trading, road infrastructure, water bodies, etc.-data that are typically not available through census surveys. The differences between the national and regional models suggest that our results can inform planners making decisions at different hierarchical levels of jurisdiction: national, region and local.
Our study has the potential to inform the design of future research related to the epidemiology of other EIDs in Viet Nam and elsewhere. For example, we speculate that in Southeast Asia, Japanese encephalitis, the transmission of which is associated with rice cultivation and flood irrigation [80] , may also show a strong association with peri-urbanization. In some areas of Asia these ecological conditions occur near, or occasionally within, urban centers. Likewise, Hantaan virus, the cause of Korean hemorrhagic fever, is associated with the field mouse Apodemus agrarius and rice harvesting in fields where the rodents are present [80] . Our work has demonstrated that the percentage of land under rice in peri-urban areas and rural areas is similar. Hence diseases associated with rice production are likely to peak in peri-urban areas given other risk factors such as land-use diversity, CTI, and distance to infrastructure. Our poultry flock-size diversity findings may also be relevant to understanding the dynamics of other poultry related infections such as Newcastle disease. Finally, these results suggest the validity of a general model of zoonotic disease emergence that integrates IOM's convergence model with the subsequently proposed social-ecological systems and EID framework. Thus, convergence represents the coalescence in time and space of processes associated with land-cover and land-use changes. Project results question whether the urban/rural land-use dichotomy is useful when large areas and parts of the population are caught between the two. Planners need better tools for mapping the rural-urban transition, and for understanding how the specific nature of peri-urban environments creates elevated health risk that require adaptation of existing planning, land use, and development practices.
|
What is Compound Topological Index and how is it related to the risk of disease transmission?
| 600 |
Compound Topographical Index (CTI, also known as Topographical Wetness Index) is a measure of the tendency for water to pool. Studies in Thailand and elsewhere [54] have shown that the extent of surface water is a strong risk factor, possibly due to the role of water in long-range transmission and persistence of the virus. I
| 18,005 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.