Adaptation of the hepatitis B virus core protein to CD8+ T cell selection pressure Helenie Kefalakes 1,3, Bettina Budeus 2, Andreas Walker 1, Christoph Jochum 3, Gudrun Hilgard 3, Andreas Heinold 4, Falko M. Heinemann 4, Guido Gerken 3, Daniel Hoffmann 2, Joerg Timm 5
1
Institute of Virology, University of Duisburg-Essen, University Hospital, Essen, Germany Research Group Bioinformatics, Centre for Medical Biotechnology, University of Duisburg-Essen, Essen, Germany 3 Department of Gastroenterology and Hepatology, University of Duisburg-Essen, University Hospital, Essen, Germany 4 Institute for Transfusion Medicine, University of Duisburg-Essen, University Hospital, Essen, Germany 5 Institute for Virology, Heinrich-Heine-University, University Hospital, Duesseldorf, Germany 2
Kefalakes, Helenie Budeus, Budeus Walker, Andreas Jochum, Christoph Hilgard, Gudrun Heinold, Andreas Heinemann, Falko M. Gerken, Guido Hoffmann, Daniel Timm, Joerg
Key words: cytotoxic T lymphocyte; viral escape; immune selection; HBcAg, immune escape
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/hep.27771 This article is protected by copyright. All rights reserved.
Hepatology
Footnote Page Corresponding author: Prof. Dr. med Jörg Timm Heinrich-Heine-Universität Düsseldorf Universitätsklinikum Düsseldorf Institut für Virologie Universitätsstraße 1 40225 Düsseldorf Email: [email protected] Telefon: +49 211-81 12225 Telefax: +49 211-81 10792 Abbreviations: cccDNA - covalently closed circular DNA HBV – hepatitis B virus HLA – human leukocyte antigen HIV – human immune deficiency virus HCV – hepatitis C virus HDV – hepatitis D virus PBMC – peripheral blood mononuclear cells FDR – flase discovery rate IL-2 – interleukin 2 ICS – intracellular cytokine staining
Grant support: This work was supported by the Deutsche Forschungsgemeinschaft (DFG, TRR60 project B1) and by internal funds of the Medical Faculty of the University of Duisburg-Essen.
2 Hepatology This article is protected by copyright. All rights reserved.
Page 2 of 28
Page 3 of 28
Hepatology
Abstract Activation of HBV-specific CD8 T cells by therapeutic vaccination may promote sustained control of viral replication by clearance of cccDNA from infected hepatocytes. However, little is known about the exact targets of the CD8 T cell response and whether HBV reproducibly evades CD8 T cell immune pressure by mutation. The aim of this study was to address if HBV reproducibly selects substitutions in CD8 T cell epitopes that functionally act as immune escape mutations. The HBV core gene was amplified and sequenced from 148 patients with chronic HBV infection and the HLA class I genotype (A and B locus) was determined. Residues under selection pressure in the presence of particular HLA class I alleles were identified by a statistical approach utilizing the novel analysis package ‘SeqFeatR’. With this approach we identified 9 residues in HBV core under selection pressure in the presence of 10 different HLA class I alleles. Additional immunological experiments confirmed that seven of the residues were located inside epitopes targeted by patients with chronic HBV infection carrying the relevant HLA class I-allele. Consistent with viral escape, the selected substitutions reproducibly impaired recognition by HBV-specific CD8 T cells. Conclusion: Viral sequence analysis allows identification of HLA class I-restricted epitopes under reproducible selection pressure in HBV core. The possibility of viral escape from CD8 T cell immune pressure needs attention in the context of therapeutic vaccination against HBV.
3 Hepatology This article is protected by copyright. All rights reserved.
Hepatology
Introduction Despite an effective vaccine against hepatitis B virus (HBV) being available, chronic infections associated with the risk of progressive liver disease are an enormous public health problem affecting more than 240 million people worldwide. Effective suppression of viral replication by inhibition of the viral polymerase with nucleos(t)ide analogues is possible, but in most cases long term or even lifelong treatment is necessary to avoid recurrence of viral replication. In this setting the concept of therapeutic vaccination potentially is an attractive strategy to achieve viral clearance (1). It is widely accepted that CD8 T cell responses are important for clearance of HBV infection (2). Specifically, CD8 T cell responses directed against the HBV core protein are associated with recovery from HBV infection (3). Accordingly, the core protein is considered an attractive target for T cell based immune therapies. Notably, most studies so far utilized a set of HLA-A*02-restricted CD8 T cell epitopes for analyses of HBV immunology. As a consequence, there is only little knowledge about CD8 T cell epitopes presented by other HLA class I-alleles and about the antiviral function of the corresponding CD8 T cell responses. In the immune epitope database (http://www.iedb.org) a total of 17 distinct HLA class I-restricted epitopes have been reported for HBV core including only eight epitopes that are restricted by HLA types other than HLA-A*02.
In HIV and HCV there is broad evidence that CD8 T cells exert selection pressure on viral proteins ultimately leading to selection of escape mutations inside or in the flanking region of targeted epitopes (4, 5). In HBV infection it is largely unclear if immune escape mutations are selected in CD8 T cell epitopes. Early studies suggested that naturally occurring sequence variants of a dominant HLAA*02-restricted CD8 T cell epitope are not recognized by CD8 T cells (6) and could even act as antagonists of prototype-specific CD8 T cells (7). However, it was unclear if such substitutions are truly selected over the course of infection. In turn, Rehermann et al. studied the immune response and viral sequence diversity in set of frequently targeted HLA-A*02-restricted CD8 T cell epitopes and concluded that mutational immune escape is uncommon in HBV (8). More recently, statistical 4 Hepatology This article is protected by copyright. All rights reserved.
Page 4 of 28
Page 5 of 28
Hepatology
evidence for CD8 T cell escape of HBV was provided in viral sequence studies showing that specific sequence variants of viral epitopes were significantly more frequent in isolates from patients carrying the relevant HLA class I-allele (9, 10). Although this is an accepted approach for identification of residues under HLA class I-associated selection pressure, no evidence was provided that these residues were indeed located in regions targeted by CD8 T cells.
Here, we test the hypothesis, if sequence analysis of the HBV core protein allows identification of CD8 T cell epitopes under reproducible selection pressure by HLA class I-alleles. We analyzed 148 HBV pre-core/core sequences in concert with the host HLA class I genotype (A- and B-locus) and were able to identify nine residues with strong statistical support for selection of substitutions in the presence of different HLA class I-alleles. Importantly, in additional immunological assays we were able to confirm that seven of these nine residues were located inside epitopes targeted by CD8 T cells in patients with chronic HBV infection.
Methods: Patients and samples Patients with chronic HBV infection were recruited at the in- and outpatient clinic of the Department of Gastroenterology and Hepatology of the University Hospital Essen. Only genotype A and D patients without current treatment were included in the analysis. All patients were tested negative for serological markers of HCV, HDV and HIV. Informed consent was obtained from each patient and the study protocol was approved by the local ethics committee of the University Hospital Essen in accordance with the guidelines of the Declaration of Helsinki. Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll density gradient centrifugation (Biocoll; Biochrom, Germany). From PBMCs DNA was extracted using spin columns (Qiagen, Hilden, Germany). HLA-A and HLA-B typing was performed by use of sequence-specific oligonucleotides (LABType methodology) both provided by One Lambda Inc. (Canoga Park, CA) (11). 5 Hepatology This article is protected by copyright. All rights reserved.
Hepatology
Amplification and sequence analysis of the HBV core gene The genomic region covering the core gene was amplified and sequenced from 148 chronically infected subjects from Essen. Viral DNA was extracted with the QIAgen DNA Blood Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. A 700 bp Core-fragment was amplified in a two-step nested PCR using HBV specific primers HBVcore-F (tgtcaacgaccgaccttgagg), HBVcore-R (tgtagctcttgttcccaa)and HBVcore internal-F (aggctgtaggcataaattggt), HBVcore internal-R (ttcccaccttatgagtccaag). The PCR products were directly sequenced and the obtained sequences were aligned with the software Geneious 7.1.5 (Biomatters, Auckland, New Zealand). All sequences were submitted to GenBank (accession numbers KJ130249-130317).
Identification of HLA class I-associated sequence polymorphisms Amino acid sequences were first aligned with the software Geneious 7.1.5 (Biomatters, Auckland, New Zealand). Then, associations of amino acids at each alignment position with HLA-type were tested with the R-package “SeqFeatR”, freely available from http://cran.rproject.org/web/packages/SeqFeatR/index.html. For each alignment position and HLA-type, the program constructs 2x2 contingency tables with counts of the occurring combinations of HLA-types and amino acids, and then applies Fisher's exact tests to pin down significant associations. Since tests were carried for all alignment positions, we corrected test results for multiple comparisons by controlling the false discovery rate (FDR), i.e. the proportion of false positives among rejected null-hypotheses. For the computation, p-values from the individual Fisher's exact tests were transformed to q-values according to the implementation of the algorithm of Benjamini and Hochberg (12) in the p.adjust program of R-package stats. Those tests were considered significant that led to q-values below a preset FDR threshold of 0.2. The corresponding alignment positions were then considered candidates for further evaluation by immunological assays.
6 Hepatology This article is protected by copyright. All rights reserved.
Page 6 of 28
Page 7 of 28
Hepatology
Analysis of HBV-specific CD8 T cells HBV-specific CD8+ T cells were detected after antigen-specific expansion as previously described (13). Briefly, PBMCs were resuspended in complete medium (RPMI 1640 containing 10% fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin and 100 μM HEPES buffer) and stimulated with individual peptides (1 μg/ml), anti-CD28/CD49d (0.5 μg/ml; BD Biosciences), and recombinant interleukin-2 (IL-2) (20 U/ml; Hoffmann La Roche). Fresh medium containing IL-2 was added twice weekly. On day 10 the cells were tested for secretion of interferon gamma after restimulation with individual peptides (prototype or variant) by intracellular cytokine staining (ICS) and subsequent analysis on a FACS Canto (Becton Dickinson). A Friedman test followed by a Dunns post-test was used to compare the frequency of IFNg+CD8+ T cells upon restimulation with the prototype or variant epitope peptide.
Results Statistical evidence for viral escape from a dominant HLA-A*02-restricted CD8 T cell response in HBV core Earlier studies suggested that naturally occurring sequence variants in a dominant HLA-A*02restricted CD8 T cell epitope in HBV core were functionally associated with immune escape (6, 7), however, it was unclear if these substitutions were the product of selection pressure in HLA-A*02positive individuals. We addressed this question by analysis of 148 HBV core sequences including 62 sequences from HLA-A*02-positive patients and 86 sequences from HLA-A*02-negative patients. Indeed, consistent with HLA-A*02-associated selection pressure on three residues within the epitope, substitutions were significantly more frequent in patients carrying HLA-A*02 (figure 1). Although position 4 of the epitope is highly variable and sequence polymorphisms are present also in 26% of HLA-A*02 negative patients, substitutions preferably to alanine or threonine were significantly enriched in HLA-A*02 positive patients (p=0.0499). Moreover, substitutions in position 7 and 10 of the epitope were almost exclusively present in HLA-A*02 positive patients (p=0.0008 and 7 Hepatology This article is protected by copyright. All rights reserved.
Hepatology
p=0.0019). Collectively, variations from the prototype epitope sequence were present in 36 of 62 (58.1%) of HLA-A*02-positive patients and in 25 of 86 (29.1%) of HLA-A*02-negative patients (p=0.0007). Prior analysis of the functional relevance of sequence polymorphisms in position 4 and 10 of the epitope was consistent with immune escape from CD8 T cells (6, 7). Notably, the consensus sequence of genotype A and D was identical in the epitope, however, the frequencies of sequence polymorphisms differed between both genotypes. In all three positions with statistical support for immune selection, the frequency of polymorphisms was higher in genotype A isolates. This correlated with a higher frequency of HLA-A*02 in patients infected with genotype A (55.9% compared to patients infected with genotype D (36.1% ; supplementary table S1). In an effort to address the impact of HLA subtypes on the results of our analysis we obtained 4-digit HLA typing results for HLA-A*02. The non-HLA-A*02:01 subtypes present in our cohort were A*02:02 (n=1), A*02:03 (n=1), A*02:05 (n=4), A*02:06 (n=5), A*02:17 (n=3). Notably, the frequency of non-HLAA*02:01 subtypes was enriched in patients infected with genotype D. Thirteen of 43 HLA-A*02 positive patients (30.2%) infected with genotype D carried non-HLA-A*02:01 subtypes compared to one of 19 HLA-A*02 positive patients (5.3%) infected with genotype A. The frequencies of substitutions were similarly enriched in positions 7 and 10 of the epitope in patients carrying HLAA*02:01 as well as in patients carrying other subtypes (figure 1C). However, the substitution frequency in position 4 of the epitope was higher in HLA-A*02:01 positive patients compared to patients carrying other subtypes, however, the low numbers in the subgroups did not allow any statistical evaluation.
Identification of residues under HLA class I-restricted CD8 T cell selection pressure We next addressed if sequence analysis allows identification of HLA class I-associated selection pressure on the HBV pre-core/core protein. Therefore, an alignment of all HBV pre-core/core sequences was generated and for each position of the alignment it was determined if any amino acids were significantly more frequent in the presence of particular HLA class I-alleles. The analysis 8 Hepatology This article is protected by copyright. All rights reserved.
Page 8 of 28
Page 9 of 28
Hepatology
was performed for all HLA class I-alleles carried by at least 5 individuals from the cohort utilizing the package ‘SeqFeatR’. Consistent with prior studies of HIV (14), associations with a p-value
1
Institute of Virology, University of Duisburg-Essen, University Hospital, Essen, Germany Research Group Bioinformatics, Centre for Medical Biotechnology, University of Duisburg-Essen, Essen, Germany 3 Department of Gastroenterology and Hepatology, University of Duisburg-Essen, University Hospital, Essen, Germany 4 Institute for Transfusion Medicine, University of Duisburg-Essen, University Hospital, Essen, Germany 5 Institute for Virology, Heinrich-Heine-University, University Hospital, Duesseldorf, Germany 2
Kefalakes, Helenie Budeus, Budeus Walker, Andreas Jochum, Christoph Hilgard, Gudrun Heinold, Andreas Heinemann, Falko M. Gerken, Guido Hoffmann, Daniel Timm, Joerg
Key words: cytotoxic T lymphocyte; viral escape; immune selection; HBcAg, immune escape
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/hep.27771 This article is protected by copyright. All rights reserved.
Hepatology
Footnote Page Corresponding author: Prof. Dr. med Jörg Timm Heinrich-Heine-Universität Düsseldorf Universitätsklinikum Düsseldorf Institut für Virologie Universitätsstraße 1 40225 Düsseldorf Email: [email protected] Telefon: +49 211-81 12225 Telefax: +49 211-81 10792 Abbreviations: cccDNA - covalently closed circular DNA HBV – hepatitis B virus HLA – human leukocyte antigen HIV – human immune deficiency virus HCV – hepatitis C virus HDV – hepatitis D virus PBMC – peripheral blood mononuclear cells FDR – flase discovery rate IL-2 – interleukin 2 ICS – intracellular cytokine staining
Grant support: This work was supported by the Deutsche Forschungsgemeinschaft (DFG, TRR60 project B1) and by internal funds of the Medical Faculty of the University of Duisburg-Essen.
2 Hepatology This article is protected by copyright. All rights reserved.
Page 2 of 28
Page 3 of 28
Hepatology
Abstract Activation of HBV-specific CD8 T cells by therapeutic vaccination may promote sustained control of viral replication by clearance of cccDNA from infected hepatocytes. However, little is known about the exact targets of the CD8 T cell response and whether HBV reproducibly evades CD8 T cell immune pressure by mutation. The aim of this study was to address if HBV reproducibly selects substitutions in CD8 T cell epitopes that functionally act as immune escape mutations. The HBV core gene was amplified and sequenced from 148 patients with chronic HBV infection and the HLA class I genotype (A and B locus) was determined. Residues under selection pressure in the presence of particular HLA class I alleles were identified by a statistical approach utilizing the novel analysis package ‘SeqFeatR’. With this approach we identified 9 residues in HBV core under selection pressure in the presence of 10 different HLA class I alleles. Additional immunological experiments confirmed that seven of the residues were located inside epitopes targeted by patients with chronic HBV infection carrying the relevant HLA class I-allele. Consistent with viral escape, the selected substitutions reproducibly impaired recognition by HBV-specific CD8 T cells. Conclusion: Viral sequence analysis allows identification of HLA class I-restricted epitopes under reproducible selection pressure in HBV core. The possibility of viral escape from CD8 T cell immune pressure needs attention in the context of therapeutic vaccination against HBV.
3 Hepatology This article is protected by copyright. All rights reserved.
Hepatology
Introduction Despite an effective vaccine against hepatitis B virus (HBV) being available, chronic infections associated with the risk of progressive liver disease are an enormous public health problem affecting more than 240 million people worldwide. Effective suppression of viral replication by inhibition of the viral polymerase with nucleos(t)ide analogues is possible, but in most cases long term or even lifelong treatment is necessary to avoid recurrence of viral replication. In this setting the concept of therapeutic vaccination potentially is an attractive strategy to achieve viral clearance (1). It is widely accepted that CD8 T cell responses are important for clearance of HBV infection (2). Specifically, CD8 T cell responses directed against the HBV core protein are associated with recovery from HBV infection (3). Accordingly, the core protein is considered an attractive target for T cell based immune therapies. Notably, most studies so far utilized a set of HLA-A*02-restricted CD8 T cell epitopes for analyses of HBV immunology. As a consequence, there is only little knowledge about CD8 T cell epitopes presented by other HLA class I-alleles and about the antiviral function of the corresponding CD8 T cell responses. In the immune epitope database (http://www.iedb.org) a total of 17 distinct HLA class I-restricted epitopes have been reported for HBV core including only eight epitopes that are restricted by HLA types other than HLA-A*02.
In HIV and HCV there is broad evidence that CD8 T cells exert selection pressure on viral proteins ultimately leading to selection of escape mutations inside or in the flanking region of targeted epitopes (4, 5). In HBV infection it is largely unclear if immune escape mutations are selected in CD8 T cell epitopes. Early studies suggested that naturally occurring sequence variants of a dominant HLAA*02-restricted CD8 T cell epitope are not recognized by CD8 T cells (6) and could even act as antagonists of prototype-specific CD8 T cells (7). However, it was unclear if such substitutions are truly selected over the course of infection. In turn, Rehermann et al. studied the immune response and viral sequence diversity in set of frequently targeted HLA-A*02-restricted CD8 T cell epitopes and concluded that mutational immune escape is uncommon in HBV (8). More recently, statistical 4 Hepatology This article is protected by copyright. All rights reserved.
Page 4 of 28
Page 5 of 28
Hepatology
evidence for CD8 T cell escape of HBV was provided in viral sequence studies showing that specific sequence variants of viral epitopes were significantly more frequent in isolates from patients carrying the relevant HLA class I-allele (9, 10). Although this is an accepted approach for identification of residues under HLA class I-associated selection pressure, no evidence was provided that these residues were indeed located in regions targeted by CD8 T cells.
Here, we test the hypothesis, if sequence analysis of the HBV core protein allows identification of CD8 T cell epitopes under reproducible selection pressure by HLA class I-alleles. We analyzed 148 HBV pre-core/core sequences in concert with the host HLA class I genotype (A- and B-locus) and were able to identify nine residues with strong statistical support for selection of substitutions in the presence of different HLA class I-alleles. Importantly, in additional immunological assays we were able to confirm that seven of these nine residues were located inside epitopes targeted by CD8 T cells in patients with chronic HBV infection.
Methods: Patients and samples Patients with chronic HBV infection were recruited at the in- and outpatient clinic of the Department of Gastroenterology and Hepatology of the University Hospital Essen. Only genotype A and D patients without current treatment were included in the analysis. All patients were tested negative for serological markers of HCV, HDV and HIV. Informed consent was obtained from each patient and the study protocol was approved by the local ethics committee of the University Hospital Essen in accordance with the guidelines of the Declaration of Helsinki. Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll density gradient centrifugation (Biocoll; Biochrom, Germany). From PBMCs DNA was extracted using spin columns (Qiagen, Hilden, Germany). HLA-A and HLA-B typing was performed by use of sequence-specific oligonucleotides (LABType methodology) both provided by One Lambda Inc. (Canoga Park, CA) (11). 5 Hepatology This article is protected by copyright. All rights reserved.
Hepatology
Amplification and sequence analysis of the HBV core gene The genomic region covering the core gene was amplified and sequenced from 148 chronically infected subjects from Essen. Viral DNA was extracted with the QIAgen DNA Blood Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. A 700 bp Core-fragment was amplified in a two-step nested PCR using HBV specific primers HBVcore-F (tgtcaacgaccgaccttgagg), HBVcore-R (tgtagctcttgttcccaa)and HBVcore internal-F (aggctgtaggcataaattggt), HBVcore internal-R (ttcccaccttatgagtccaag). The PCR products were directly sequenced and the obtained sequences were aligned with the software Geneious 7.1.5 (Biomatters, Auckland, New Zealand). All sequences were submitted to GenBank (accession numbers KJ130249-130317).
Identification of HLA class I-associated sequence polymorphisms Amino acid sequences were first aligned with the software Geneious 7.1.5 (Biomatters, Auckland, New Zealand). Then, associations of amino acids at each alignment position with HLA-type were tested with the R-package “SeqFeatR”, freely available from http://cran.rproject.org/web/packages/SeqFeatR/index.html. For each alignment position and HLA-type, the program constructs 2x2 contingency tables with counts of the occurring combinations of HLA-types and amino acids, and then applies Fisher's exact tests to pin down significant associations. Since tests were carried for all alignment positions, we corrected test results for multiple comparisons by controlling the false discovery rate (FDR), i.e. the proportion of false positives among rejected null-hypotheses. For the computation, p-values from the individual Fisher's exact tests were transformed to q-values according to the implementation of the algorithm of Benjamini and Hochberg (12) in the p.adjust program of R-package stats. Those tests were considered significant that led to q-values below a preset FDR threshold of 0.2. The corresponding alignment positions were then considered candidates for further evaluation by immunological assays.
6 Hepatology This article is protected by copyright. All rights reserved.
Page 6 of 28
Page 7 of 28
Hepatology
Analysis of HBV-specific CD8 T cells HBV-specific CD8+ T cells were detected after antigen-specific expansion as previously described (13). Briefly, PBMCs were resuspended in complete medium (RPMI 1640 containing 10% fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin and 100 μM HEPES buffer) and stimulated with individual peptides (1 μg/ml), anti-CD28/CD49d (0.5 μg/ml; BD Biosciences), and recombinant interleukin-2 (IL-2) (20 U/ml; Hoffmann La Roche). Fresh medium containing IL-2 was added twice weekly. On day 10 the cells were tested for secretion of interferon gamma after restimulation with individual peptides (prototype or variant) by intracellular cytokine staining (ICS) and subsequent analysis on a FACS Canto (Becton Dickinson). A Friedman test followed by a Dunns post-test was used to compare the frequency of IFNg+CD8+ T cells upon restimulation with the prototype or variant epitope peptide.
Results Statistical evidence for viral escape from a dominant HLA-A*02-restricted CD8 T cell response in HBV core Earlier studies suggested that naturally occurring sequence variants in a dominant HLA-A*02restricted CD8 T cell epitope in HBV core were functionally associated with immune escape (6, 7), however, it was unclear if these substitutions were the product of selection pressure in HLA-A*02positive individuals. We addressed this question by analysis of 148 HBV core sequences including 62 sequences from HLA-A*02-positive patients and 86 sequences from HLA-A*02-negative patients. Indeed, consistent with HLA-A*02-associated selection pressure on three residues within the epitope, substitutions were significantly more frequent in patients carrying HLA-A*02 (figure 1). Although position 4 of the epitope is highly variable and sequence polymorphisms are present also in 26% of HLA-A*02 negative patients, substitutions preferably to alanine or threonine were significantly enriched in HLA-A*02 positive patients (p=0.0499). Moreover, substitutions in position 7 and 10 of the epitope were almost exclusively present in HLA-A*02 positive patients (p=0.0008 and 7 Hepatology This article is protected by copyright. All rights reserved.
Hepatology
p=0.0019). Collectively, variations from the prototype epitope sequence were present in 36 of 62 (58.1%) of HLA-A*02-positive patients and in 25 of 86 (29.1%) of HLA-A*02-negative patients (p=0.0007). Prior analysis of the functional relevance of sequence polymorphisms in position 4 and 10 of the epitope was consistent with immune escape from CD8 T cells (6, 7). Notably, the consensus sequence of genotype A and D was identical in the epitope, however, the frequencies of sequence polymorphisms differed between both genotypes. In all three positions with statistical support for immune selection, the frequency of polymorphisms was higher in genotype A isolates. This correlated with a higher frequency of HLA-A*02 in patients infected with genotype A (55.9% compared to patients infected with genotype D (36.1% ; supplementary table S1). In an effort to address the impact of HLA subtypes on the results of our analysis we obtained 4-digit HLA typing results for HLA-A*02. The non-HLA-A*02:01 subtypes present in our cohort were A*02:02 (n=1), A*02:03 (n=1), A*02:05 (n=4), A*02:06 (n=5), A*02:17 (n=3). Notably, the frequency of non-HLAA*02:01 subtypes was enriched in patients infected with genotype D. Thirteen of 43 HLA-A*02 positive patients (30.2%) infected with genotype D carried non-HLA-A*02:01 subtypes compared to one of 19 HLA-A*02 positive patients (5.3%) infected with genotype A. The frequencies of substitutions were similarly enriched in positions 7 and 10 of the epitope in patients carrying HLAA*02:01 as well as in patients carrying other subtypes (figure 1C). However, the substitution frequency in position 4 of the epitope was higher in HLA-A*02:01 positive patients compared to patients carrying other subtypes, however, the low numbers in the subgroups did not allow any statistical evaluation.
Identification of residues under HLA class I-restricted CD8 T cell selection pressure We next addressed if sequence analysis allows identification of HLA class I-associated selection pressure on the HBV pre-core/core protein. Therefore, an alignment of all HBV pre-core/core sequences was generated and for each position of the alignment it was determined if any amino acids were significantly more frequent in the presence of particular HLA class I-alleles. The analysis 8 Hepatology This article is protected by copyright. All rights reserved.
Page 8 of 28
Page 9 of 28
Hepatology
was performed for all HLA class I-alleles carried by at least 5 individuals from the cohort utilizing the package ‘SeqFeatR’. Consistent with prior studies of HIV (14), associations with a p-value
