High resolution analysis of cellular immune responses in resolved and persistent hepatitis C virus infection

GASTROENTEROLOGY 2004;127:924 –936

High Resolution Analysis of Cellular Immune Responses in Resolved and Persistent Hepatitis C Virus Infection GEORG M. LAUER,*,‡ ELEANOR BARNES,储,¶ MICHAELA LUCAS,储 JOERG TIMM,*,‡ KEI OUCHI,*,‡ ARTHUR Y. KIM,*,‡ CHERYL L. DAY,*,‡ GREGORY K. ROBBINS,*,‡ DEBORAH R. CASSON,# MARKUS REISER,** GEOFFREY DUSHEIKO,¶ TODD M. ALLEN,*,‡ RAYMOND T. CHUNG,# BRUCE D. WALKER,*,‡,§ and PAUL KLENERMAN储 *Partners AIDS Research Center, ‡Infectious Disease Division, and §Howard Hughes Medical Institute, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; 储Nuffield Department of Clinical Medicine, University of Oxford, Oxford, United Kingdom; ¶ Center for Hepatology, Royal Free Hospital, London, United Kingdom; #Gastrointestinal Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; and **Medizinische Universitaetsklinik, Knappschaftskrankenhaus, Bochum, Germany

Background & Aims: Cellular immune responses are thought to play a key role in the resolution of primary HCV infection. Although it has been consistently shown that CD4ⴙ T-cell responses are maintained in those with spontaneous resolution but lost in those with persistent infection, the role of CD8ⴙ T-cell responses remains controversial. Previous studies have largely focused on limited HLA alleles and predefined CD8ⴙ T-cell epitopes, and, thus, comprehensive studies remain to be performed. Methods: To understand the composition of the immune response associated with spontaneous resolution, we comprehensively mapped CD8ⴙ T-cell responses in 20 HLA-diverse persons with resolved HCV infection, using HCV peptides spanning the entire genome. We analyzed the magnitude, breadth, function, and phenotype using ELISpot, class-I tetramers, intracellular cytokine staining, and cytolytic assays. We studied in parallel HCV-specific responses and viral sequence variation in persistent infection. Results: Responses in individuals with resolved infection were strong and broad with robust proliferation in response to antigen. Responses in those persistently infected were rarely detected ex vivo and, when present, were narrowly directed and weak. However, they also proliferated in vitro. Dominant target epitopes differed among individuals in both cohorts, despite frequently shared HLAalleles. Conclusions: These data indicate that persisting, strong CD8ⴙ T-cell responses are observed in the majority of persons with resolved HCV infection and provide support for strategies to boost CD8ⴙ T-cell responses for the prevention or treatment of HCV infection but also highlight the diversity of responses that may need to be elicited to provide protection.

ulations of multispecific CD4⫹ and CD8⫹ T-cell responses are generated during primary HCV infection in individuals who both resolve infection and in those who go on to develop persistent infection.2–5 It has been hypothesized that maintenance of a functional T-cell response is crucial to long-term control of HCV viremia. Although it has been consistently shown that this is the case for CD4⫹ T-cell responses,2,3,6 –9 the role of maintaining a functional CD8⫹ T-cell response long term is controversial, with recent studies providing conflicting results. Three studies4,6,9 have shown that HCV-specific CD8⫹ T-cell responses are maintained many years or even decades following the spontaneous resolution of HCV, whereas this response is typically lost in individuals with persistent infection. Others10,11 have suggested quite the reverse, implying that ongoing viremia is required to maintain a CD8⫹ T-cell response, which is otherwise lost, and showing that CD8⫹ T-cell responses are detected more frequently in individuals with persistent infection but that these responses are functionally impaired in their proliferative capacity.11 To add further to the complexity, a separate study has shown that the magnitude of the CD8⫹ T-cell response critically determines outcome in the first 6 months following infection but that, after 6 months, CD8⫹ T-cell responses are undetectable in the majority of patients, irrespective of outcome.12 Furthermore, although multiple studies have shown genetic associations with HLA class II gene expression and protection against persistent infection, suggesting an

significant proportion of individuals infected with hepatitis C virus (HCV) spontaneously control infection long-term,1 and cellular immune responses are thought to play an important role. However, large pop-

Abbreviations used in this paper: HCV, hepatitis C virus; ICS, intracellular cytokine staining; SFC, spot-forming cells. © 2004 by the American Gastroenterological Association 0016-5085/04/$30.00 doi:10.1053/j.gastro.2004.06.015

A

September 2004

HIGH-RESOLUTION STUDY OF HCV IMMUNE RESPONSES

925

Table 1. CD8⫹ T Cell Responses in Subjects With Resolved Infection (R1-R20) Patient ID

HLA

R1 R2 R3

A3, 66 B7, 49 Cw7 A2, 3 B7, 60 Cw7 A2, 25 B37, 44 Cw7, 12

R4 R5

A2, 29 B44 Cw16 A24, 33 B35,65 Cw4,8

R6 R7 R8 R9

A11, 29 B35,40 Cw3,4 A1 B8,49 Cw7 A2,24 B15,44 Cw3,7 A1,68 B53,57 Cw4,6

R10 R11 R12 R13 R14

A1,2 B57,63 A1, 11 B8,44 Cw7,16 A1, 11 B35,57 Cw4,6 A1, 11 B8,65(14) Cw7,8 A2,11 B65(14),57 Cw6,8

R15 R16 R17 R18 R19 R20

A1,11 B37,62(15) Cw4,6 A2,23 B7,44 Cw4,7 A2,3 B41,44 Cw5,17 A24,33 B7,65(14) Cw7,8 A3,24 B7,37 Cw6,7 A2 B44,57 Cw5,6

Peptides recognized by CD8⫹ T cells E2-610 Cw7 YRLWHYPCTI (145) Core-41 B7 GPRLGVRAT (110) NS2-831 A25 LSPYYKRYIS (475) NS3-1073 A2 CINGVCWTV (750) NS4-1966 B37 SECCTPCSGSW (850) NS5-2594 A2 ALYDVVTKL (750) Core-88 B44 NEGCGWMGW (510) E1-207 B35* CPNSSIVY (500) NS4-1695 B35* IPDREVLY (160) NS5-2162 B35* EPEPDVAVL (25) neg neg NS4-1987 A2 VLDSFKTWL (65) E2-541 B57* NTRPPLGNWFG (90) NS3-1435 A1 ATDALMTGY (225) NS5-2912 B57* LGVPPLRAWR (350) NS3-1406 A2 KLVALGINAV (60) NS3-1395 B8 HSKKKCDEL (120) NS3-1435 A1 ATDALMTGY (75) NS3-1435 A1 ATDALMTGY (45) E2-541 B57* NTRPPLGNWFG (60) NS4-1751 VFTGLTHIDAHFLSQTKQSG (75) NS4-1801 B57* LTTSQTLLF (65) neg Core-41 B7 GPRLGVRAT (130) NS3-1406 A2 KLVALGINAV (90) NS3-1610 CLIRLKPTLHGPTPLLYR (100) Core-111 B7 DPRRRSRNL (155) NS3-1406 A2 KLVALGINAV (60)

NS3-1070 ATCINGVCWTVYHGAGTRTI (860) E2-610 Cw7 YRLWHYPCTI (860) NS2-957 B37 RDWAHNGL (900) NS4-1758 A25 ETFWAKHMW (200) NS5-2225 A25 ELIEANLLW (210) NS5-2819 A25 TARHTPVNSW (750) NS3-1359 B35 HPNIEEVAL (195) NS4-1745 A24* VIAPAVQTNW (705) NS5-2912 LGVPPLRAWR (150)

NS5-2461 TSRSACQRQKKVTFDRLQVL (45) NS3-1175 A68* HAVGLFRAA (290) NS5-2629 B57 KSKKTPMGF (125) NS4-1801 B57* LTTSQTLLF (50)

NS5-2912 LGVPPLRAWR (65) NS2-941 LGALTGTYVYNHLTPLRDWA (60) NS4-1771 GIQYLAGLSTLPGNPAIASL (45)

NS3-1406 A2 KLVALGINAV (30)

E1-322 MMMNWSPTT (70) NS5-2594 A2 ALYDVVTKL (65)

NOTE. The peptides to which responses were detected are described by the HCV-protein together with the aa position in the HCV-H77 sequence, followed by the HLA restriction of the peptide if known and the peptide sequence. The numbers in parentheses indicate the strength of the respective response in the ELISpot assay (SFC/106 PBMC). Fully defined novel minimal epitope peptides are marked with an asterisk.

important role for CD4⫹ T cells in determining outcome,13–15 no comparable association has been demonstrated for HLA class I gene expression and CD8⫹ T cells. The controversy surrounding the role of CD8⫹ T cells in mediating viral control of HCV infection may have arisen because of limitations in the methodologies used for previous studies, in which the analysis is often limited to a single HLA allele, often HLA A2, and to testing a small number of predicted HCV-specific epitopes. This approach underestimates the breadth and magnitude of the response and appears to give an incomplete assessment of the relative strength of the HCV-specific CD8⫹ T-cell responses among different individuals.16 To test the hypothesis that spontaneous control of HCV infection is associated with persistent HCV-specific CD8⫹ T-cell responses and that antigen persistence is not required for a multispecific response, we have comprehensively analyzed CD8⫹ T-cell responses in 20 individuals with diverse HLA backgrounds following spontaneous resolution of HCV infection and compared these to the responses observed in 20 individuals with persistent infection. In contrast to previous studies, we

have assessed CD8⫹ T-cell responses targeting the entire HCV polypeptide, using ex vivo ELISpot analysis for interferon-␥ production. This has allowed us to determine the HCV-specific T-cell response directed against multiple HCV proteins, irrespective of HLA type. Accurately defining the role of CD8⫹ T cells in protective immunity is important for the development of future therapeutic strategies.

Materials and Methods Study Subjects Twenty subjects (subjects R1 to R20) with spontaneously resolved HCV infection, defined as the absence of detectable HCV RNA by PCR (HCV Roche Amplicor assay, detection limit of 300 HCV RNA copies/mL of plasma) in the presence of HCV antibodies (third generation EIA) on at least 2 occasions were studied. Additionally, 20 treatment naïve patients with chronic HCV infection characterized by elevated liver function tests (LFTs) and persistent plasma viremia were consecutively enrolled from the outpatient clinic. All subjects were studied at least 2 years after acute HCV infection. Further subject characteristics are listed in Tables 1 and 2. In addition to these 40 subjects, we studied PBMC from 1 additional individual with chronic infec-

926

LAUER ET AL

GASTROENTEROLOGY Vol. 127, No. 3

Table 2. CD8⫹ T Cell Responses in Subjects With Chronic Infection (C1–C20) Patient ID

HLA

HCV genotype

HCV RNA (IU/mL)

ALT (IU/mL)

C1 C2

A1,68 B14,55 Cw7 A1,2 B37,44 Cw5,6

1b 2a/2c

300800 38600

216 21

C3 C4

A3,30 B13,51 Cw6,15 A1,2 B8,55 Cw3,7

1a 1a

530000 18400

32 43

C5 C6 C7 C8 C9 C10 C11 C12 C13

2b 2b 1a 2b 1b 3a 1b 1b 1a

456490 220970 342980 7840000 428540 ⬎1000000 12480000 203000 9760000

338 274 29 205 58 92 106 47 152

C14 C15 C16 C17

A2,29 B18,40 Cw3,5 A1,31 B8 Cw7 A1 B15 Cw9,10 A2,3 B39,44 Cw7,16 A1 B8 Cw7 A1,26 B44,51 Cw4,14 A11,23 B35,44 Cw4 A2,3 B7,44 Cw7 A2,26 B35,38 Cw4, 12 A2,36 B52,65 Cw8,15 A2,11 B35,44 Cw7,9 A23,30 B7,52 Cw7,16 A3,32 B7,14 Cw7,8

1a 2a/c 1a 1a

5340000 5240000 ⬎1000000 152000

48 68 120 96

C18 C19 C20

A2,29 B44,50 Cw4,16 A1,2 B8,44 Cw1 A3,24 B7,35 Cw4,7

3a 1a 1b

1213300 796000 ⬎850000

100 86 100

Peptides recognized by CD8⫹ T cells NS5-2568 B55 QPEKGGRKPA (65) NS2-957 B37 RDWAHNGL (90) NS3-1406 A2 KLVALGINAV (50) NEG NS3-1435 A1 ATDALMTGY (85) NS5-2898 B55 SPGEINRVAA (45) NEG NEG NEG P7-790 Cw7* FYGMWPLL (70) NS2-831 LSPYYKRYISWCLWWLQYFL (135) neg neg neg NS2-871 DAVILLMCAVHPTLVFDITK (130) neg neg neg E2-610 Cw7 YRLWHYPCTI (75) NS4-1941 AARVTAIL (255) P7-790 A29* FYGMWPLLL(280) NS5-2197 SVASSSASQLSA (120) neg

NS5-2898 B55 SPGEINRVAA (65) NS3-1073 A2 CINGVCWTV (120) NS5-2594 A2 ALYDVVTKL (70) NS5-2568 B55 QPEKGGRKPA (40)

NS3-1171 HCPAGHAVGIFRAAVCTRGVA (60)

NS4-1744 EVIAPAVQTNW (70) NS5-2594 A2 ALYDVVTKL (70)

NOTE. The peptides to which responses were detected are described by the HCV protein together with the aa position in the HCV-H77 sequence, followed by the HLA restriction of the peptide if known and the peptide sequence. The numbers in parentheses indicate the strength of the respective response in the ELISpot assay (SFC/106 PBMC). Fully defined novel minimal epitope peptides are marked with an asterisk.

tion (A1) in the phenotypic analysis using class I tetramers. The study was approved by the Institutional Review Boards, and all subjects gave written informed consent.

HLA Typing HLA typing was performed using standard molecular techniques.17

Synthesis of HCV-Derived Peptides Peptides corresponding to the amino acid sequence of the HCV-1a strain, spanning the entire HCV polyprotein, were synthesized as free acids using the 9-fluorenylmethoxy carbonyl method. The 301 peptides used in the initial ELISpot screening assay were 20 amino acids (aa) in length, overlapping adjacent peptides by 10 aa. In addition, a panel of previously defined HCV-specific CD8⫹ T-cell epitopes (n ⫽ 83), ranging in size from 8 to 11 aa, were tested.18 To define the optimal peptide sequence for novel HCV epitopes identified in the screening ELISpot assay, we synthesized N- and C-terminal truncated peptides.

Matrix ELISpot Assay The 301 HCV peptides were grouped into pools of 10 peptides each in a matrix array, such that each individual peptide could be found in 2 pools only. A positive response in 2 pools containing the same peptide could then be used to identify a putative response, which was confirmed using individual peptides in a repeat ELISpot assay. Interferon-␥ ELISpot assays were performed exactly as previously described. Specif-

ically, fresh or previously frozen PBMC were added at a concentration of 200,000 cells/well. Peptides were added directly to the wells at a final concentration of 10 ␮g/mL. The plates were incubated for 18 hours at 37°C, 5% CO2. Spots were counted on an ELISpot reader (AID, Strassberg, Germany). For quantification of ex vivo responses, the assay was performed at least in duplicate. Responses were considered positive if the number of spots per well minus the background was at least 25 spot-forming cells (SFC)/106 PBMC with a background of less than 15 SFC/106 PBMC. Phytohemagglutinin (PHA) served as a positive control.

Strategy for Further Defining CD8ⴙ T-Cell Responses Peptides giving a positive response in the screening ELISpot assay were truncated and used in further ELISpot assays to define the optimal epitope sequence. Strong CD4⫹ or CD8⫹ T-cell responses were differentiated by direct ex vivo intracellular cytokine staining (ICS) (see below). In addition, cell lines were generated by bulk peptide stimulation of PBMC. These lines were used to differentiate weaker CD4⫹/ CD8⫹ responses by ICS and to define the HLA restriction of responses (Figure 1). Tetramer staining ex vivo was used to identify CD8⫹ T-cell responses if a tetramer was available that corresponded to an epitope identified in the ELISpot assay. There were no epitopes identified in the screening ELISpot that did not proliferate in the bulk stimulation assays, thus ruling out the possibility that we specifically selected responses with proliferative capacity.

September 2004

HIGH-RESOLUTION STUDY OF HCV IMMUNE RESPONSES

927

Figure 1. Example of screening for HCV-specific CD8⫹ T-cell responses. The strength and specificity of the HCV specific CD8⫹ T-cell response was determined in each individual using an interferon-␥ ELISpot matrix spanning the entire HCV genome. Responses in a representative patient (R5) are shown here. The epitopes targeted are plotted beneath the corresponding position in the HCV genome map, and the magnitude of the response (SFC, spot-forming cells) is given. Subject R5 targeted 6 different epitopes, 1 located in E1, 1 in NS3, 2 in NS4, and 2 in NS5 (A). Strong responses, such as the one directed against the peptide NS4-1745, could be confirmed in a direct ex vivo ICS (B). Responses were confirmed as CD8 positive using peptide-specific T-cell lines (following a single round of peptide stimulation; C), and HLA restriction of the epitope was defined using partially HLA-matched and -mismatched heterologous B cell lines (C).

HLA Class I Peptide Tetramer Staining Phenotypic analysis and quantification of CD8⫹ T-cell responses was determined using HLA class I peptide tetramers incorporating CD8⫹ T-cell epitopes that were synthesized specifically for this study following the identification of CD8⫹ T-cell epitopes in the ELISpot assay. Tetramers were synthesized as previously described.4 The decision to generate a tetramer for a specific response depended on the availability of the heavy chain for tetramer synthesis rather than on the magnitude of the response in the ELISpot assay. Tetramers were synthesized for 4 epitopes restricted by HLA-A2 (HLA A2: NS3 peptide 1073–1081,19 CINGVWCTV; NS4 peptide 1406 –1415, KLVALGINAV20; NS4 peptide 1987–1995, VLDSFKTWL21; NS5B peptide 2594 –2603, ALYDVVTKL4) and 3 further epitopes restricted by HLA A1 (HLA A1: NS3 peptide 1435–1443, ATDALMTGY16), HLA A24 (HLA A24: NS4 1745–1754, VIAPAVQTNW), and HLA B35 (HLA B35: NS3 1359 –1367, HPNIEEVAL22); 0.5–1 million PBMC were stained as described. Briefly, tetramer staining was performed for 20 minutes at 37°C. After washing for 5

minutes with PBS containing 1% FCS at room temperature (RT), cells were pelleted and directly stained with combinations of the following antibodies: CD8-PerCP, CD27-FITC, CD28-APC, and CD45RA-FITC (all from Becton Dickinson, Franklin Lakes, NJ). For indirect antibody staining using CCR7, cells were washed twice, pelleted, and stained with anti–CCR7-Ab (Becton Dickinson) for 30 minutes at room temperature. After 2 further washes, cells were pelleted again, and a secondary anti-mouse-IgM-APC-conjugated Ab (Caltag, Burlingame, CA) was added for 30 minutes at RT. Cells were washed twice, and directly conjugated CD8-PerCP Ab was added for 20 minutes at 4°C to the cell pellets. All staining was performed in a volume of 100 ␮L PBS in the presence of 10 ␮L goat immunoglobulin to prevent nonspecific Ab binding. Flow cytometric analysis was performed with a Becton Dickinson fluorescence-activated cell sorter (FACS Calibur), and data analysis was performed using the CellQuest software. Staining was considered positive if, as previously described,4 tetramer-positive cells formed a cluster distinct from the tetramer negative CD8⫹ T-cell population, and the frequency of

928

LAUER ET AL

tetramer positive cells was greater than 0.02% of the total CD8⫹ population.

Intracellular IFN-␥ Staining ICS for IFN-␥ was performed as described previously.16 Briefly, 1 ⫻ 106 PBMC were incubated with 4 ␮mol/L peptide and anti-CD28 and anti-CD49d MAbs (1 ␮g/mL each; Becton Dickinson) at 37°C and 5% CO2 for 1 hour before the addition of Brefeldin A (1 ␮L/mL; SigmaAldrich). The cells were incubated for an additional 5 hours at 37°C and 5% CO2. PBMC were then washed and stained with surface antibodies, anti-CD3-APC, and anti-CD8-PE (Becton Dickinson) at room temperature for 20 minutes Following the washing, the PBMC were fixed and permeabilized (Caltag), and the fluorescein isothiocyanate-conjugated anti-IFN-␥ MAb (Becton Dickinson) was added. Cells were then washed and analyzed on a FACS-Calibur flow cytometer using CELLQuest software (Becton Dickinson).

Bulk Stimulation of Peripheral Blood Mononuclear Cells To establish CD8⫹ T-cell lines, cryopreserved or fresh PBMC (4 to 10 ⫻ 106) were stimulated with 1 ␮g/mL of synthetic HCV peptide and 0.5 ␮g/mL of the costimulatory antibodies anti-CD28 and anti-CD49d (Becton Dickinson) in R10. Irradiated feeder cells (20 ⫻ 106 allogeneic PBMC) were added to the culture in a 25-cm2 culture flask (Costar, Cambridge, MA). Recombinant interleukin-2 (IL-2; 25 IU/mL) was added on day 2 and twice a week thereafter. Lines were harvested for further analysis (ICS/HLA restriction and cytotoxicity assays) after day 9.

Cytotoxicity Assay Autologous B-lymphocyte cell lines (B-LCL) were pulsed with 10 ␮g of peptide and [51Cr]O4 (New England Nuclear, Boston, MA) and incubated for 1 hour at 37°C in 5% CO2. The B-LCL target cells were washed 3 times with cold R10 medium and incubated with effector cells at 37°C for 4 hours in 3 replicate wells. Cellular release of [51Cr]O4 into the supernatant was measured using a Top Count Microplate scintillation counter (Packard Instrument Company, Meriden, CT), and the percentage specific cytotoxicity was calculated by the formula percentage lysis ⫽ [(experimental release ⫺ spontaneous release)/(maximum release ⫺ spontaneous release)] ⫻ 100. Results are reported as the mean of triplicate values. Only experiments with a spontaneous release of ⬍20% were evaluated.

HLA Restriction To define HLA restriction, partially HLA-matched and -mismatched heterologous BCL were pulsed with 10 ␮g of peptide for an hour, washed 3 times with R10, and 2 ⫻ 105 of the BCL were added to the T-cell lines. Cytotoxicity assays or ICS were then performed as described previously.

GASTROENTEROLOGY Vol. 127, No. 3

Sequence Analysis Viral RNA was extracted from plasma or serum samples using a Qiagen vRNA extraction kit. When necessary, samples were pretreated with heparinase I (1 U/␮L; SigmaAldrich) prior to RNA isolation. Primer sets were designed for different genotypes based on alignments of all available sequences from the public HCV database (http://hcvpub.ibcp.fr). Initially, a set of external primer pairs was used to amplify fragments of the region of interest of the HCV genome in a combined reverse transcription and PCR. When necessary, additional internal primer pairs were used in nested PCR reactions to yield the desired fragments of ⬃0.4 kb for sequencing. Using the Qiagen One-Step RT-PCR kit, RT-PCR cycling conditions were as follows: 50°C for 60 minutes, 95°C for 15 minutes, followed by 35 cycles of 30 seconds at 94°C, 30 seconds at 54°C, 1.5 minutes at 72°C, and a final extension of 68°C for 20 minutes. Nested PCR conditions using a high-fidelity Taq polymerase (Titanium Taq DNA-Polymerase, BD Clontech) were 35 cycles of 30 seconds at 94°C, 30 seconds at 62°C, 1 minute at 72°C, and a final extension of 68°C for 20 minutes. PCR fragments were then gel or PCR purified (Qiagen Kit) and population sequenced bidirectionally on an ABI 3100 PRISM automated sequencer. Sequencher (Gene Codes Corp., Ann Arbor, MI) and MacVector 4.1 (Oxford Molecular) software programs were used to edit and align sequences.

Statistical Analysis Statistical analysis (Mann–Whitney rank sum test and correlation coefficient) was performed using GraphPad Prism 3.0a for Macintosh.

Results Circulating HCV-specific CD8ⴙ T-lymphocytes Restricted by Diverse HLA Alleles Can Be Detected Directly Ex Vivo Years After Spontaneous Resolution of HCV Infection First, we tested the hypothesis that spontaneous resolution of HCV viremia is associated with persistence of a broad and vigorous immune response by CD8⫹ HCV-specific T cells. For this, we assessed an HLA diverse cohort of individuals who had cleared HCV spontaneously at least 2 years prior to study entry, as indicated by a positive HCV antibody test and consistently undetectable HCV RNA. Previous studies using direct ex vivo techniques limited to selected HCV epitopes have reported relatively weak responses in persons who had spontaneously cleared plasma viremia in HCV infection.4,9 –11 In this study, we comprehensively screened the entire HCV genome in the matrix ELISpot for putative CD8⫹ T-cell epitopes, followed by the differentiation of CD4⫹ and CD8⫹ T cells

September 2004

HIGH-RESOLUTION STUDY OF HCV IMMUNE RESPONSES

929

Figure 2. Comparison of the strength and breadth of the HCV-specific CD8⫹ T-cell response. In each individual, the summation of spot-forming cells (SFC) for all epitopes targeted by that individual (A) and the number of epitopes targeted by each individual (B) are plotted, and the 2 cohorts are compared using the Mann–Whitney nonparametric test. Open squares and closed triangles represent individuals with chronic and resolved infection, respectively.

(the strategy for which is outlined in Figure 1), based on ex vivo tetramer or ex vivo ICS analysis, in addition to the generation of cell lines followed by ICS and definition of HLA restriction. Using this comprehensive approach, we could detect HCV-specific CD8⫹ T-cell responses in the peripheral blood of 17 of 20 (85%) persons with resolved infection (Table 1). In this cohort, a mean of 2.3 epitopes (range, 0 – 8) was targeted with a mean total spot magnitude of 584 (range, 0 – 4885) SFC/106 PBMC. Up to 8 epitopes were targeted in a single individual (R3), and the strength of some individual epitope-specific responses reached 900 SFC/106 PBMC in the ELISpot assay (Table 1). HCV-Specific CD8ⴙ T-Lymphocytes Are Detected Less Commonly and With Lower Frequencies in Individuals With Chronic HCV Infection Because previous studies have generated conflicting results when comparing immune responses in individuals with resolved and chronic infection, we applied this comprehensive assay to a similar cohort of persons with chronic untreated HCV infection. This cohort did not significantly differ in its age and gender distribution (data not shown). For this cohort with detectable viremia, we were also able to determine the HCV genotype (Table 2). As expected, there was a dominance of HCV genotype I in this cohort, as previously described for the United States and United Kingdom populations from which these subjects were derived. The distribution of major HLA alleles was similar in the resolved and chronic cohorts (HLA A1 7 vs. 8, HLA A2 9 vs. 10, HLA A3 4 vs. 5, HLA A11 5 vs. 2, HLA B7 5 vs. 4, HLA B8 3 vs. 4, HLA B35 3 vs. 4, and HLA B44 7 vs. 8, respectively). Therefore, a similar number of optimal

epitopes matching the individual’s HLA type were tested in both groups. In contrast to subjects with resolved infection, a significantly smaller number of individuals with chronic disease had HCV-specific CD8⫹ T-cell responses (Table 2; 9 of 20 vs. 17 of 20; P ⫽ 0.019). In addition, their responses were weaker, with a lower total magnitude (mean total spots, 95; range, 0 – 400) SFC/106 PBMC in the chronic cohort vs. mean total spots 584 (range, 0 – 4885) SFC/106 PBMC in the resolved cohort (P ⫽ 0.027, Figure 2A) and narrower, i.e., directed against a smaller number of epitopes (mean, 1; range, 0 – 4), in the chronic cohort vs. mean 2.3 (range, 0 – 8) in the resolved cohort (P ⫽ 0.039, Figure 2B). However, despite these significant differences, there was a substantial overlap between the 2 cohorts in terms of the breadth and the overall strength of the individual responses. The specific T-cell epitopes targeted in individuals with both resolved and chronic infection were derived from a wide range of HCV proteins, were restricted by a variety of HLA alleles, and were highly heterogeneous among individuals (Tables 1 and 2, Figure 3A). Although there was a lack of responses directed against core and E1 in the chronic cohort, this difference was not statistically significant compared with the group with resolved infection (P ⫽ 0.165, Fisher exact test). Although some epitopes were targeted by several individuals, none of the HCV epitopes were predictably targeted by all, or even a majority of persons of the same HLA type in either cohort (Figure 3B). Analysis of the Role of Sequence Variation in Subjects With Persistent HCV Infection One potential confounding factor in this analysis is the instability of the HCV genome, which leads to

930

LAUER ET AL

GASTROENTEROLOGY Vol. 127, No. 3

Figure 3. Distribution of HCV epitopes and frequency of recognition. Figure 3A shows the distribution of HCV-specific responses over the HCV genome in the entire cohort (n ⫽ 40). Each bar represents a distinctive HCV epitope, with the length of the bar indicating the number of subjects targeting this epitope. Bars above the line correspond to responses in the cohort with resolved infection, and those below the line represent responses detected in persons with chronic infection. Figure 3B shows recognition of specific epitopes known to be restricted by HLA alleles expressed by at least 3 persons in each cohort. Epitopes included were selected based on recognition in this study or because they had been described as being frequently recognized in previous studies. The bars indicate the percentage of subjects bearing the respective HLA allele who recognized the epitope in the ELISpot assay.

variability between different HCV strains, even within the same genotype. To address the role of sequence variation in our finding of limited T-cell responses in chronic HCV infection, we sequenced the autologous viral sequence for commonly recognized epitopes restricted by common HLA alleles (NS3 A2 1073, NS3 A2 1406, NS3 A1 1435) in all subjects carrying the respective HLA allele (Table 3). In cases in which the autologous sequence differed from our genotype 1a prototype sequence, we synthesized the respective variant peptide and compared its recognition to the prototype peptide using T-cell lines in an ELISpot using serial peptide dilutions. We found that the autologous epitope sequence was often determined by genotype, especially in the case of the 2 HLA A2-restricted epitopes. However, in the majority of individuals, the autologous sequence was

accurately represented by the prototype peptide, or a cross-reactive variant was present (Table 3). In cases in which genotype 2 or 3 was present, a distinct, nonreactive peptide sequence typically was detected for the HLA A2-restricted epitopes. Thus, the failure to detect responses against these 3 well-defined epitopes was not, in most cases tested, because of a lack of the appropriate antigen in the circulating virus. We next focused on the sequences of those 5 epitopes for which strong T-cell responses were detected in persons with chronic infection using both interferon-␥ ELISpot and tetramer analysis (Table 4). This analysis revealed an interesting phenomenon. Responses to genotype 1a peptides were detectable in subjects C2 and C18 with a total of 4 epitopes targeted, but, in these cases, the circulating virus was nongenotype 1 and contained peptide sequences that were not recognized by T-cell lines

September 2004

HIGH-RESOLUTION STUDY OF HCV IMMUNE RESPONSES

931

Table 3. Autologous Viral Sequences Analyzed for 3 Commonly Recognized Epitopes in the Individuals Positive for the Respective HLA Allele Patient ID

HLA

HCV genotype

C1 C2 C4 C5 C6 C7 C8 C9 C10 C12 C13 C14 C15 C18 C19

A1 A1,2 A1,2 A2 A1 A1 A2 A1 A1 A2 A2 A2 A2 A2 A1,A2

1b 2a/2c 1a 2b 2b 1a 2b 1b 3a 1b 1a 1a 2a/c 3a 1a

a

NS3 A2 1406 KLVALGINAV

S–S––L–––

A–RGM–L––– a V R A–RGM–V–––

NS3 A11435 ATDALMTGY Fa Fa Fa

S a

L

a

Fa Fa S–––L–––

A–RSM–V––– Fa n.d. a

a

a

a

a

S–––L––– n.d. a

a

A–RGM–L––– ––RGM–L––– a R

Fa

Autologous sequences that were either homologous or cross-reactive with our prototype peptide sequence are bold.

specific for the prototype peptide sequence. Only the peptide sequence of the HLA A1-restricted epitope in subject C4 was reactive with the respective prototypespecific T-cell line. Thus, some of the strongest T-cell responses in subjects with chronic infection did not recognize the autologous infecting virus. Analysis of Phenotype of HCV-Specific CD8ⴙ T Lymphocytes To characterize the phenotype of HCV-specific responses, we synthesized 7 HLA class I tetramers. The quality of the staining that we obtained with each of these tetramers is demonstrated in Figure 4. A comparative analysis could then be performed on CD8⫹ T-cell responses targeting 12 HCV epitopes: 7 from individuals with resolved infection and 5 from individuals with chronic infection, for whom there were sufficient PBMC available. Because some of the responses analyzed in chronic individuals did not cross-react with the circulating infecting virus, we also included 2 responses from an Table 4. Sequences of the Autologous Infecting Virus for Responses Analyzed by Class I Tetramers Patient ID C2

C4 C18 a

NS3 A2 1073 CINGVCWTV

NS3-1073 A2 CINGVCWTV NS3-1406 A2 KLVALGINAV S–S––L––– A–RGM–L––– NS5-2594 A2 ALYDVVTKL –––––TQR– NS3-1435 A1 ATDALMTGY Fa NS5-2594 A2 ALYDVVTKL –––––IQ––

Autologous sequences that were either homologous or cross-reactive with our prototype peptide sequence are bold.

additional chronically infected individual (A1), not part of the cohort described previously. This person recognized 2 epitopes (NS3 A2 1073 and NS5 A2 2594), with the circulating autologous virus perfectly matching the prototype peptide sequence. Cells were tested for expression of markers associated with T-cell differentiation,23,24 including CD45RA, CD27, CD28, and CCR7 (Figure 5A). As shown in Figure 5A and Figure 5B, HCVspecific CD8⫹ T-cells, with the exception of 1 epitope (NS5 2594) in an individual (R3) with resolved infection (Figure 5B, lower panels), displayed an “early memory” phenotype, with most T cells being double positive for CD27 and CD28 and low in CD45RA expression.23 CCR7 surface expression was relatively high, consistent with an “early memory” phenotype. In addition, we correlated the results from the functional IFN-␥ ELISpot with the results of direct quantification by tetramer analysis. Only a minority (10%– 20%) of tetramer-positive HCV-specific cells secreted IFN-␥, whether from individuals with resolved or chronic infection—a phenomenon that we have previously described (Figure 6).25 Moreover, by establishing peptide-specific T-cell lines for multiple epitopes in which peptide specificity was confirmed by ICS (Figure 7A), we demonstrated that, following stimulation, the HCV-specific T cells were capable of proliferation (in the presence of IL-2) in 12 of 12 epitopes tested in individuals with chronic infection and 29 of 29 epitopes tested in individuals with resolved infection. Furthermore, we were able to demonstrate that HCV-specific CD8⫹ T-cell lines from individuals with both resolved and persistent infection possess cytolytic function (Figure 7B). This analysis included responses in

932

LAUER ET AL

GASTROENTEROLOGY Vol. 127, No. 3

Figure 4. Ex vivo HCV tetramer response. Tetramers were custom synthesized for 7 different HCV epitopes and restricted by 4 different HLA alleles, and cells were stained ex vivo in both resolved and unresolved infection. The percentage of tetramer⫹ cells/total CD8⫹ cells is given in the upper right panel of each dot plot.

persistently infected individuals where the autologous sequence was intact or cross-reactive. Identification of Novel Epitopes This comprehensive approach led to the identification of 10 novel HCV-specific epitopes along with their restricting HLA alleles (Table 5). We confirmed these novel responses to be mediated by CD8⫹ T cells, to have cytolytic activity, and to have proliferative capacity as shown by expansion of specific cells after peptide stimulation.

Discussion The maintenance of HCV-specific CD8⫹ T cells following HCV resolution is currently controversial, with several studies in persons of specific HLA types, particularly HLA A2, reaching entirely opposing conclusions.9 –12 To address this issue, we used peptides spanning the entire HCV viral genome to map comprehensively and compare CD8⫹ T-cell responses in HLAdiverse individuals with persistent and resolved HCV infection. This has allowed us to define further the breadth and the magnitude of HCV-specific CD8⫹ T cells in resolved vs. persistent HCV infection in persons expressing diverse HLA alleles. Because the immunologic methods used necessarily relied on a single prototype strain of HCV, as has been standard in the field, the results likely underestimate the responses to autologous virus. However, exactly the same methods were used for persons with resolved and persistent infection, the data provide clear evidence that responses are enhanced in the absence of persistent viremia. Using this comprehensive approach, which is not limited to persons of preselected HLA types, we detected

HCV-specific CD8⫹ T-cell responses directly ex vivo in 85% of individuals with resolved HCV infection. HCVspecific CD8⫹ T cells targeting up to 8 different epitopes were detected in a single subject, with individual responses composing up to 2.8% of CD8⫹ T cells in the setting of undetectable viremia. In contrast, CD8⫹ T cells were detected in only 45% of individuals with chronic infection, were significantly weaker, and were directed against fewer epitopes. These percentages in chronic infection are similar to reported results examining liver-specific immune responses.19 Examination of intrahepatic T-cell responses in persons with resolved infection is not ethically possible because of lack of indication for liver biopsy. In contrast to some studies of highly variable RNA viruses in which specific immunodominant CTL responses are associated with different disease courses,26,27 we detected no relationship between specific epitopes targeted and disease outcome (Figure 3B). Moreover, even in persons with the same HLA types and same disease status, the epitopes targeted varied substantially. This lack of consistent immunodominance was similar to our previous more limited analysis16 focusing on HLA A2-restricted responses in which we likewise found no evidence that targeting of specific epitopes or gene products was associated with viral control or persistence in HCV infection. In persistently infected individuals, our sequence analysis for selected epitopes restricted by the frequent class I alleles HLA A1 and A2 demonstrated that the lack of recognition of these epitopes cannot be explained by sequence variation alone. The majority of subjects with chronic infection harbored HCV sequences identical or cross-reactive with the prototype sequence but still did not display the corresponding CD8⫹ T-cell

September 2004

HIGH-RESOLUTION STUDY OF HCV IMMUNE RESPONSES

933

Figure 5. Phenotype of HCVspecific CD8⫹ T cells. (A) Analysis of phenotypic marker expression across all tetramer responses analyzable. Triangles represent individuals with resolved infection, and squares represent individuals with chronic infection. For the chronic subjects, responses marked by solid squares represent those reactive with the autologous infecting virus of the individual. Open squares mark responses not recognizing the circulating autologous virus. (B) The surface expression pattern of HCV-specific CD8⫹ T cells was analyzed using antibodies for CD45RA, CD27, CD28, and CCR7 and is shown for 3 epitopes in 2 representative patients (C2 and R3). The panels on the left show the expression of CD27 and CD28 within the total CD8⫹ T-cell population and the tetramer-positive cells alone. The panels on the right show CD45RA and CCR7 expression in the tetramer-positive and -negative populations. The 2 upper rows are representative examples for the phenotype seen in both chronic and resolved subjects. In contrast, the lower row shows the staining for a unique response in a single patient with a distinct phenotype. Note that the middle and lower rows are different epitopes recognized by the same individual with resolved infection.

response. Therefore, this failure of the immune system to recognize consistently even the conserved regions of HCV is consistent with the hypothesis that HCV has the ability to interfere with the induction of primary immune responses by antigen-presenting cells.28 –31 Alternatively, primary responses may be intact, but clonal exhaustion through continuous high viral loads might play a role.32 At least for the specificities for which we had tetramers available, it is unlikely that cells were present but simply nonfunctional in vivo because tetramer analysis did not reveal such a phenotype in the persons studied. In examining the role of sequence diversity in persistent infection, we observed that some strong responses were detectable to genotype 1 peptide sequences in in-

dividuals infected with a different genotype. The epitopes targeted in these cases showed multiple substitutions compared with the infecting genotype and lacked cross-reactivity. The presence of T-cell responses in these cases may represent a residue from a previous resolved infection with a genotype 1 strain, as has been reported previously for CD4⫹ T-cell responses.14 If these responses were removed from consideration in the analysis of the differences between the 2 groups, the significance of the difference in breadth and vigor of the response would further increase. We found no correlation between the measured CD8⫹ T-cell response and viral load or ALT values in our cohort with chronic infection (data not shown). This is in agreement with some, but not all, of the previous

934

LAUER ET AL

GASTROENTEROLOGY Vol. 127, No. 3

Table 5. Novel CD8⫹ T Cell Epitopes HCV protein

aa Position

Sequence

HLA restriction

E1 E2 P7 P7 NS3 NS4 NS4 NS4 NS5 NS5

207–214 541–550 790–798 790–797 1175–1184 1695–1712 1745–1754 1801–1809 2162–2170 2912–2921

CPNSSIVY NTRPPLGNWF FYGMWPLLL FYGMWPLL HAVGLFRAA IPDREVLY VIAPAVQTNW LTTSQTLLF EPEPDVAVL LGVPPLRAWR

B35 B57 A29 Cw7 A68 B35 A24 B57 B35 B57

NOTE. Novel fully defined minimal HCV epitope peptides and their restricting HLA allele. Figure 6. Correlation between tetramer-positive T cells and IFN-␥ secretion. The number of tetramer-positive cells was correlated with the number of IFN-␥ producing specific cells as detected in the ELISpot assay ex vivo (SFC, spot forming cells). The number of IFN-␥-secreting cells was universally lower, representing 10%–20% of tetramer-positive cells in individuals with both resolved and chronic infection.

studies.19,33,34 Given the sequence variation between autologous infecting virus and the prototype peptides used in all these studies, some responses might have been missed, and some detected responses might not have recognized the autologous infecting virus. Only comprehensive screening with peptides exactly matching the autologous infecting virus will finally resolve this question in the future. Thus, a critical question remains as to whether the responses that do persist and target the virus present in vivo do correlate with relative levels of control in chronic infection. Recent studies have suggested that CD8⫹ T cells may exist in a spectrum ranging from early differentiated (or central memory) cells high in CD27 and CD28, through to fully differentiated (or effector memory) T cells low in

CD27 and CD28. Expression of CCR7, a secondary lymph node homing marker, is also associated with the early differentiated phenotype. Our analysis of the phenotype of HCV-specific CD8⫹ T cells employed 7 different tetramers based on 4 different HLA alleles. We found that the phenotype of HCV-specific CD8⫹ T cells were CD27High, CD28High, CD45RALow, and CCR7High in each epitope in all individuals with both chronic and resolved infection, with the exception of a single response in 1 patient with resolved infection (Figure 5B, patient R3, epitope NS5-2594), in whom responses to 1 of 2 phenotyped epitopes were CD27High, CD28High, CD45RAHigh, and CCR7Low. Although it is clear that distinct phenotypes may arise in different viral infections, our data suggest that the maturation phenotype in individuals with both resolved and chronic infection is relatively consistent. The validity of this is further underlined by the fact that, in subjects with chronic infection, the phenotype was identical for responses that clearly recognized the autologous viral sequence. There are a number of possible mechaFigure 7. HCV-specific CD8⫹ T cells can proliferate and lyse target cells after a single round of peptide stimulation in vitro in individuals with both resolved and chronic infection. HCV-specific CD8⫹ T cells from subjects with resolved, as well as chronic infection could be expanded after 10 –14 days, following a single round of peptide stimulation (A) in individuals with both resolved and chronic infection. Peptide specificity was confirmed by ICS for IFN-␥ The cell lines specifically lysed peptide pulsed targets in a standard 4-hour chromium release assay at an effector/target ratio 30:1 (B).

September 2004

nisms to account for both the observed early memory phenotype and loss of antigen-responsive cells in persistent infection. Effector memory populations may be compartmentalized to the liver in which they actively contribute to the control of viremia36 and/or are deleted following interaction with antigen presented in the liver environment.37 Alternatively, effector memory cells may not be generated through mechanisms such as the infection and malfunction of dendritic cells in HCV-infected individuals.38 It has recently been suggested that HCV-specific CD8⫹ T cells with impaired functional properties can be found in chronic, but not resolved, infection in that CD8⫹ T cells from patients with chronic infection proliferate poorly in response to peptide stimulation both in the presence and in the absence of IL-2.11 This contrasts with early work in subjects with chronic infection, which relied on T-cell expansion to identify HCV-specific CD8⫹ T-cell responses20,39 – 42 and with reports demonstrating a superior proliferative capacity for the central memory subset of CD8⫹ T cells,35 which we found to be predominant in chronic HCV infection. In support of this, we have readily generated CD8⫹ T-cell lines from individuals with both resolved and persistent infection following a single round of peptide stimulation and have recently shown that even weak responses, which are not detectable directly ex vivo, can be readily expanded in a short-term in vitro culture.16 The findings in this current study strongly support the idea that, although IFN-␥ production is relatively weak as previously described,4 in vitro proliferative capacity is maintained, at least under conditions of IL-2 supplementation. It remains to be seen whether quantitative differences in T-cell function will be unmasked under different culture conditions, as has been recently shown for HIV-1-specific CD8⫹ T-cell responses.43 In conclusion, we have performed a comprehensive analysis of CD8⫹ T-cell responses in a spontaneously resolving vs. chronic viral infection in an HLA diverse cohort. We demonstrate that diverse HCV-specific T-cell responses can be preserved at relatively high levels in the absence of detectable virus. In contrast, responses in chronic infection are weaker and narrower, despite continuous high-level antigenic stimulation. Whether broadening or magnifying the CD8⫹ T-cell response in those with chronic infection is possible and whether this would contribute to disease resolution are important questions that arise from this study and that demand further prospective analysis. This study also highlights the diversity of responses that are associated with resolution of infection and that may need to be elicited to provide protection

HIGH-RESOLUTION STUDY OF HCV IMMUNE RESPONSES

935

References 1. Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001;345:41–52. 2. Gerlach J, Diepolder H, Jung M-C, Gruener N, Schraut W, Zachoval R, Hoffman R, Schirren C, Santantonio T, Pape G. Recurrence of HCV after loss of virus specific CD4⫹ T cell response in acute Hepatitis C. Gastroenterology 1999;117:933–941. 3. Thimme R, Oldach D, Chang KM, Steiger C, Ray SC, Chisari FV. Determinants of viral clearance and persistence during acute hepatitis C virus infection. J Exp Med 2001;194:1395–1406. 4. Lechner F, Wong DK, Dunbar PR, Chapman R, Chung RT, Dohrenwend P, Robbins G, Phillips R, Klenerman P, Walker BD. Analysis of successful immune responses in persons infected with hepatitis C virus. J Exp Med 2000;191:1499 –1512. 5. Lechner F, Gruener NH, Urbani S, Uggeri J, Santantonio T, Kammer AR, Cerny A, Phillips R, Ferrari C, Pape GR, Klenerman P. CD8⫹ T lymphocyte responses are induced during acute hepatitis C virus infection but are not sustained. Eur J Immunol 2000; 30:2479 –2487. 6. Wertheimer AM, Miner C, Lewinsohn DM, Sasaki AW, Kaufman E, Rosen HR. Novel CD4⫹ and CD8⫹ T-cell determinants within the NS3 protein in subjects with spontaneously resolved HCV infection. Hepatology 2003;37:577–589. 7. Ferrari C, Valli A, Galati L, Penna A, Scaccaglia P, Giuberti T, Schianchi C, Missale G, Marin MG, Fiaccadori F. T-cell response to structural and nonstructural hepatitis C virus antigens in persistent and self-limited hepatitis C virus infections. Hepatology 1994;19:286 –295. 8. Day CL, Lauer GM, Robbins GK, McGovern B, Wurcel AG, Gandhi RT, Chung RT, Walker BD. Broad specificity of virus-specific CD4⫹ T-helper-cell responses in resolved hepatitis C virus infection. J Virol 2002;76:12584 –12595. 9. Takaki A, Wiese M, Maertens G, Depla E, Seifert U, Liebetrau A, Miller JL, Manns MP, Rehermann B. Cellular immune responses persist and humoral responses decrease two decades after recovery from a single-source outbreak of hepatitis C. Nat Med 2000;6:578 –582. 10. Chang KM, Thimme R, Melpolder JJ, Oldach D, Pemberton J, Moorhead-Loudis J, McHutchison JG, Alter HJ, Chisari FV. Differential CD4(⫹) and CD8(⫹) T-cell responsiveness in hepatitis C virus infection. Hepatology 2001;33:267–276. 11. Wedemeyer H, He XS, Nascimbeni M, Davis AR, Greenberg HB, Hoofnagle JH, Liang TJ, Alter H, Rehermann B. Impaired effector function of hepatitis C virus-specific CD8⫹ T cells in chronic hepatitis C virus infection. J Immunol 2002;169:3447–3458. 12. Gruner NH, Gerlach TJ, Jung MC, Diepolder HM, Schirren CA, Schraut WW, Hoffmann R, Zachoval R, Santantonio T, Cucchiarini M, Cerny A, Pape GR. Association of hepatitis C virus-specific CD8⫹ T cells with viral clearance in acute hepatitis C. J Infect Dis 2000;181:1528 –1536. 13. Cramp ME, Carucci P, Underhill J, Naoumov NV, Williams R, Donaldson PT. Association between HLA class II genotype and spontaneous clearance of hepatitis C viraemia. J Hepatol 1998; 29:207–213. 14. Harcourt G, Hellier S, Bunce M, Satsangi J, Collier J, Chapman R, Phillips R, Klenerman P. Effect of HLA class II genotype on T helper lymphocyte responses and viral control in hepatitis C virus infection. J Viral Hepat 2001;8:174 –179. 15. Zavaglia C, Martinetti M, Silini E, Bottelli R, Daielli C, Asti M, Airoldi A, Salvaneschi L, Mondelli MU, Ideo G. Association between HLA class II alleles and protection from or susceptibility to chronic hepatitis C. J Hepatol 1998;28:1–7. 16. Lauer GM, Ouchi K, Chung RT, Nguyen TN, Day CL, Purkis DR, Reiser M, Kim AY, Lucas M, Klenerman P, Walker BD. Comprehensive analysis of CD8(⫹) T-cell responses against hepatitis C

936

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

LAUER ET AL

virus reveals multiple unpredicted specificities. J Virol 2002;76: 6104 – 6113. Bunce M, O’Neill CM, Barnardo MC, Krausa P, Browning MJ, Morris PJ, Welsh KI. Phototyping: comprehensive DNA typing for HLA-A, B, C, DRB1, DRB3, DRB4, DRB5 and DQB1 by PCR with 144 primer mixes utilizing sequence-specific primers. Tissue Antigens 1995;46:355–367. Ward S, Lauer G, Isba R, Walker B, Klenerman P. Cellular immune responses against hepatitis C virus: the evidence base 2002. Clin Exp Immunol 2002;128:195–203. Wong DK, Dudley DD, Afdhal NH, Dienstag J, Rice CM, Wang L, Houghton M, Walker BD, Koziel MJ. Liver-derived CTL in hepatitis C virus infection: breadth and specificity of responses in a cohort of persons with chronic infection. J Immunol 1998;160:1479–1488. Cerny A, McHutchison JG, Pasquinelli C, Brown ME, Brothers MA, Grabscheid B, Fowler P, Houghton M, Chisari FV. Cytotoxic T lymphocyte response to hepatitis C virus-derived peptides containing the HLA A2.1 binding motif. J Clin Invest 1995;95:521–530. Urbani S, Uggeri J, Matsuura Y, Miyamura T, Penna A, Boni C, Ferrari C. Identification of immunodominant hepatitis C virus (HCV)-specific cytotoxic T-cell epitopes by stimulation with endogenously synthesized HCV antigens. Hepatology 2001;33:1533– 1543. Ibe M, Sakaguchi T, Tanaka K, Saito S, Yokota S, Tanaka T, Shimotohno K, Chujoh Y, Shiratori Y, Omata M, Miwa K, Takiguchi M. Identification and characterization of a cytotoxic T cell epitope of hepatitis C virus presented by HLA-B*3501 in acute hepatitis. J Gen Virol 1998;79:1735–1744. Appay V, Dunbar PR, Callan M, Klenerman P, Gillespie GM, Papagno L, Ogg GS, King A, Lechner F, Spina CA, Little S, Havlir DV, Richman DD, Gruener N, Pape G, Waters A, Easterbrook P, Salio M, Cerundolo V, McMichael AJ, Rowland-Jones SL. Memory CD8⫹ T cells vary in differentiation phenotype in different persistent virus infections. Nat Med 2002;8:379 –385. Champagne P, Ogg GS, King AS, Knabenhans C, Ellefsen K, Nobile M, Appay V, Rizzardi GP, Fleury S, Lipp M, Forster R, Rowland-Jones S, Sekaly RP, McMichael AJ, Pantaleo G. Skewed maturation of memory HIV-specific CD8 T lymphocytes. Nature 2001;410:106 –111. Gruener NH, Lechner F, Jung MC, Diepolder H, Gerlach T, Lauer G, Walker B, Sullivan J, Phillips R, Pape GR, Klenerman P. Sustained dysfunction of antiviral CD8(⫹) T lymphocytes after infection with hepatitis C virus. J Virol 2001;75:5550 –5558. Goulder PJ, Altfeld MA, Rosenberg ES, Nguyen T, Tang Y, Eldridge RL, Addo MM, He S, Mukherjee JS, Phillips MN, Bunce M, Kalams SA, Sekaly RP, Walker BD, Brander C. Substantial differences in specificity of HIV-specific cytotoxic T cells in acute and chronic HIV infection. J Exp Med 2001;193:181–194. Wherry EJ, Blattman JN, Murali-Krishna K, van der Most R, Ahmed R. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J Virol 2003;77:4911– 4927. Bain C, Fatmi A, Zoulim F, Zarski JP, Trepo C, Inchauspe G. Impaired allostimulatory function of dendritic cells in chronic hepatitis C infection. Gastroenterology 2001;120:512–524. Auffermann-Gretzinger S, Keeffe EB, Levy S. Impaired dendritic cell maturation in patients with chronic, but not resolved, hepatitis C virus infection. Blood 2001;97:3171–3176. Bertolino P, Trescol-Biemont MC, Thomas J, de St Groth BF, Pihlgren M, Marvel J, Rabourdin-Combe C. Death by neglect as a deletional mechanism of peripheral tolerance. Int Immunol 1999; 11:1225–1238. Bertolino P, Trescol-Biemont MC, Rabourdin-Combe C. Hepatocytes induce functional activation of naive CD8⫹ T lymphocytes but fail to promote survival. Eur J Immunol 1998;28:221–236. Moskophidis D, Lechner F, Pircher H, Zinkernagel RM. Virus persistence in acutely infected immunocompetent mice by ex-

GASTROENTEROLOGY Vol. 127, No. 3

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

haustion of antiviral cytotoxic effector T cells. Nature 1993;362:758 –761. Sugimoto K, Ikeda F, Stadanlick J, Nunes FA, Alter HJ, Chang KM. Suppression of HCV-specific T cells without differential hierarchy demonstrated ex vivo in persistent HCV infection. Hepatology 2003;38:1437–1448. Nelson DR, Marousis CG, Davis GL, Rice CM, Wong J, Houghton M, Lau JY. The role of hepatitis C virus-specific cytotoxic T lymphocytes in chronic hepatitis C. J Immunol 1997;158:1473– 1481. Wherry EJ, Teichgraber V, Becker TC, Masopust D, Kaech SM, Antia R, von Andrian UH, Ahmed R. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat Immunol 2003;4:225–234. Minutello MA, Pileri P, Unutmaz D, Censini S, Kuo G, Houghton M, Brunetto MR, Bonino F, Abrignani S. Compartmentalization of T lymphocytes to the site of disease: intrahepatic CD4⫹ T cells specific for the protein NS4 of hepatitis C virus in patients with chronic hepatitis C. J Exp Med 1993;178:17–25. Limmer A, Ohl J, Kurts C, Ljunggren HG, Reiss Y, Groettrup M, Momburg F, Arnold B, Knolle PA. Efficient presentation of exogenous antigen by liver endothelial cells to CD8⫹ T cells results in antigen-specific T-cell tolerance (In Process Citation). Nat Med 2000;6:1348 –1354. Kanto T, Hayashi N, Takehara T, Tatsumi T, Kuzushita N, Ito A, Sasaki Y, Kasahara A, Hori M. Impaired allostimulatory capacity of peripheral blood dendritic cells recovered from hepatitis C virus-infected individuals. J Immunol 1999;162:5584 –5591. Cerny A, Ferrari C, Chisari FV. The class I-restricted cytotoxic T lymphocyte response to predetermined epitopes in the hepatitis B and C viruses. Curr Top Microbiol Immunol 1994;189:169 – 186. Chang KM, Gruener NH, Southwood S, Sidney J, Pape GR, Chisari FV, Sette A. Identification of HLA-A3 and -B7-restricted CTL response to hepatitis C virus in patients with acute and chronic hepatitis C. J Immunol 1999;162:1156 –1164. Rehermann B, Chang KM, McHutchinson J, Kokka R, Houghton M, Rice CM, Chisari FV. Differential cytotoxic T-lymphocyte responsiveness to the hepatitis B and C viruses in chronically infected patients. J Virol 1996;70:7092–7102. Rehermann B, Chang KM, McHutchison JG, Kokka R, Houghton M, Chisari FV. Quantitative analysis of the peripheral blood cytotoxic T lymphocyte response in patients with chronic hepatitis C virus infection. J Clin Invest 1996;98:1432–1440. Migueles SA, Laborico AC, Shupert WL, Sabbaghian MS, Rabin R, Hallahan CW, Van Baarle D, Kostense S, Miedema F, McLaughlin M, Ehler L, Metcalf J, Liu S, Connors M. HIV-specific CD8⫹ T cell proliferation is coupled to perforin expression and is maintained in nonprogressors. Nat Immunol 2002;3:1061–1068.

Received September 8, 2003. Accepted June 10, 2004. Address requests for reprints to: Paul Klenerman, Peter Medawar Building for Pathogen Research, Nuffield Department of Clinical Medicine, University of Oxford, South Parks Road, Oxford OX1 3SY, United Kingdom. e-mail: [email protected]; fax: (44) 1865-281-236. Supported by the Deutsche Forschungsgemeinschaft (DFG LA 1241/1-1, to G.M.L.), the European Union (grant code QLK2-CT-199900356 and QLK2-CT-2002-01329, to M.L.), the National Institutes of Health (AI31563), the Doris Duke Charitable Foundation, and the Howard Hughes Medical Institute. G.M.L. and E.B. contributed equally to this work. E.B. is a Medical Research Council clinical training fellow. P.K. is a Wellcome Trust Senior Clinical Fellow. B.D.W. is the recipient of a Doris Duke Distinguished Clinical Scientist Award.