Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Strong Vaccine-Induced CD8 T-Cell Responses Have Cytolytic Function in a Chimpanzee Clearing HCV Infection

  • Babs E. Verstrepen,

    Affiliation Department of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Ernst J. Verschoor,

    Affiliation Department of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Zahra C. Fagrouch,

    Affiliation Department of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Petra Mooij,

    Affiliation Department of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Natasja G. de Groot,

    Affiliation Department of Comparative Genetics and Refinement, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Ronald E. Bontrop,

    Affiliation Department of Comparative Genetics and Refinement, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Willy M. Bogers,

    Affiliation Department of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

  • Jonathan L. Heeney,

    Affiliation Department of Veterinary Medicine, University of Cambridge, Cambridge, United Kingdom

  • Gerrit Koopman

    Affiliation Department of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

Strong Vaccine-Induced CD8 T-Cell Responses Have Cytolytic Function in a Chimpanzee Clearing HCV Infection

  • Babs E. Verstrepen, 
  • Ernst J. Verschoor, 
  • Zahra C. Fagrouch, 
  • Petra Mooij, 
  • Natasja G. de Groot, 
  • Ronald E. Bontrop, 
  • Willy M. Bogers, 
  • Jonathan L. Heeney, 
  • Gerrit Koopman


A single correlate of effective vaccine protection against chronic HCV infection has yet to be defined. In this study, we analyzed T-cell responses in four chimpanzees, immunized with core-E1-E2-NS3 and subsequently infected with HCV1b. Viral clearance was observed in one animal, while the other three became chronically infected. In the animal that cleared infection, NS3-specific CD8 T-cell responses were observed to be more potent in terms of frequency and polyfunctionality of cytokine producing cells. Unique to this animal was the presence of killing-competent CD8 T-cells, specific for NS31258–1272, being presented by the chimpanzee MHC class I molecule Patr-A*03∶01, and a high affinity recognition of this epitope. In the animals that became chronically infected, T-cells were able to produce cytokines against the same peptide but no cytolysis could be detected. In conclusion, in the animal that was able to clear HCV infection not only cytokine production was observed but also cytolytic potential against specific MHC class I/peptide-combinations.


Hepatitis C virus (HCV) infection is characterized by a high propensity for development of chronic infection, which typically manifests as asymptomatic for a long period of time. However, over decades the virus causes subtle, but cumulatively irreversible, hepatic damage. A vaccine that could prevent development of persistent HCV infection would therefore be of great clinical benefit.

The chimpanzee is the only validated animal model of HCV infection [1], [2]. The dilemma that biomedical research in non-human primates, and chimpanzees in particular, is inevitably associated with low animal numbers and limiting statistical analysis, has been discussed elsewhere [3]. Nevertheless, we believe that results from in depth immune-profiling of immunized chimpanzees may provide new insights into immune mechanisms operating in the early phase after infection and as such is important for optimal vaccine development in the future.

Spontaneous HCV clearance has been associated with the presence of broad and strong T-cell responses in both humans and chimpanzees [4][7]. Moreover, specific memory T-cell responses often correlate with early clearance after HCV reinfection [8][10]. These data imply that T-cell-based vaccines may facilitate HCV clearance, protect from viral persistence and thus from HCV-related disease progression.

For this reason, several HCV T-cell vaccine-candidates have been evaluated. Some of these vaccine-strategies induced strong T-cell responses however, no correlation was observed between either the magnitude or the breadth of vaccine-induced T-cell responses, and viral clearance. The lytic capacity of vaccine-induced CD8 T-cells, in the context of MHC class-I presentation was not previously evaluated in relation to viral clearance.

Previously, we reported partial control of an HCV 1b challenge after a DNA prime-MVA-boost vaccine strategy, targeting HCV core, E1, E2 and NS3 in chimpanzees [11]. In these animals, neither (neutralizing) antibody responses nor ex vivo cytokine production or proliferative responses were uniquely associated with control of HCV infection. Therefore, the current study aimed to identify potential mechanisms of protection via detailed functional characterization of the T-cell responses against the most dominantly recognized antigen, NS3 [11].

We report that the animal that cleared infection had a high percentage of polyfunctional cytokine producing CD4 and CD8 T-cells against NS3-peptides. Moreover, a strong cytolytic T-cell (CTL) response against epitope NS31258–1272 was identified, which was uniquely presented in the context of the chimpanzee MHC class I molecule Patr-A*03∶01.

Materials and Methods

Ethics statement

This retrospective study was performed on cryopreserved PBMCs, isolated from animals that were part of an HCV vaccine-efficacy study [11]. As previously described, the preceding study had been performed in six purpose bred, naive mature chimpanzees (Pan troglodytes) that were housed at the Biomedical Primate Research Centre (BPRC), Rijswijk, The Netherlands, according to international guidelines for non-human primate care and use (The European Council Directive 86/609/EEC, and Convention ETS 123, including the revised Appendix A). The Institutional Animals Care and Use Committee (DEC-BPRC) approved study protocols (#426) according to strict international ethical and scientific standards and guidelines. The qualification of the members of this committee, including their independence from a research institute, is stated in the “Wet op de Dierproeven” (1996). The experiment can only be performed after a positive recommendation from this ethical committee. Permission for projects on non-human primates is requested in an animal ethical committee form. This form is discussed in the ethical committee and, if approved, a written permission is sent to the project leader and a copy to the director of the BPRC, who holds licence to perform animal experimentation. The project was monitored by a qualified independent veterinarian, specifically regarding the ethical issues of the projects.

The animals were in good physical health with normal biochemical and hematological values. The animals were socially housed in a BSL3-facility with spacious cages, and were provided with commercial food pellets supplemented with appropriate treats. Drinking water was provided ad libitum. Environmental enrichment was provided daily and suffering of the animals was alleviated wherever possible. In 2002, the European Council banned the use of apes for biomedical research in Europe. Yet, this was decided after the initiation of the original vaccine-study described by Rollier et. al. [11]. For this reason, the European Council approved continuation of this particular study. The following years, all chimpanzees that were housed at the BPRC were outplaced to public zoos or animal sanctuaries. Before 2001, the use of chimpanzees for biomedical research was highly restricted and regulated, it was not allowed to perform terminal experiments on these animals [12].


For stimulation and expansion of HCV-specific cells, two different NS3-peptide-pools were compiled: NS3vaccine was a mixture of 15-mer peptides with 11 amino acids overlap, covering the NS3-sequence of the vaccine constructs; NS3challenge was a mixture of 15-mer peptides with 11 amino acids overlap covering the NS3-variants observed in the inoculum. Identification of NS3 epitopes within NS3 was first performed via the matrix setup [13] and confirmed in a second assay using individual peptides (peptide-sequences are listed in Table S1).

In vitro expansion of NS3-specific T-cell lines

Cells isolated before infection were expanded and tested for response against peptide-pool NS3vaccine, whereas cells isolated after challenge were tested against peptide-pool NS3challenge. PBMCs isolated at the day of challenge were stimulated in two separate cultures with either NS3vaccine or NS3challenge.

After thawing, PBMCs were washed and resuspended in culture medium, R20 (RPMI, Invitrogen, CA, USA) supplemented with 20% FCS (MP medicals, Solon, OH, USA), pen/strep (100 U/ml/100 µg/ml) (Invitrogen) and L-glutamine (2 mM) (Invitrogen)) and cultured at a final density of 10 million cells per ml in a 24 well culture plate (Greiner bio-one). The cultures were initiated in R20 containing 5 ng/ml IL-7 (Peprotech) and NS3-peptide-pool (5 µg/ml per peptide). After overnight incubation, 1 ml R20 containing IL-7 (10 ng/ml) and IL-2 (Proleukin Chiron, final concentration 10 IU/ml) was added per well. Cultures were checked daily for general condition and depending on cell density, fresh R20 IL-2/IL-7 was added or cultures were split. After 12 days, the expanded cells were collected and resuspended in fresh R20 for further analysis.

T-cell epitope mapping

The cytokine production profile of HCV-specific T-cell lines was determined by intracellular cytokine staining (ICS) after restimulation with NS3-peptides. For T-cell epitope mapping 0.2×106 expanded cells were either restimulated with NS3-peptide-pool (1 µg/ml/peptide) or without NS3-peptide, in medium containing co-stimulatory αCD28 and αCD49d molecules (2 µg each, BD-Biosciences) and 5% FCS (MP-Biomedicals). After 2 hrs, Brefeldin A (Golgiplug 1∶1000, BD-Biosciences) was added and 16 hours later the surface markers were stained with a panel of fluorochrome labeled antibodies, containing CD3-PacificBlue (clone SP-34-2, (BD-Pharmingen), CD14-PE-TexasRed (clone RMO52, BeckmanCoulter), CD20-PE-TexasRed (clone B9E9, BeckmanCoulter), CD4-PE-Cy7 (clone SK3, BD-Pharmingen) and CD8-APC-H7 (clone SK1, BD-Pharmingen). Stained cells were fixed and permeabilized (Cytofix/Cytoperm, BD Biosciences), followed by staining of accumulated intracellular cytokines with IFNγ-APC (clone B27, BD Pharmingen), IL-2-PE (clone MQ1-17H12, BD Pharmingen) and TNFα-FITC (clone MAb11, BD Pharmingen). Cell staining was analyzed on a FACSAria (BD Bioscience) and DIVA software Version 6.1.1.

To evaluate antigen sensitivity, T-cell lines were restimulated with increasing concentrations of individual peptide ranging from 0.05 to 10 µg/ml before intracellular cytokine production was measured.

MHC-class-I genotyping

Animals were genotyped for their MHC-class-I repertoire as previously described [14] (Table 1).

Non-radioactive cytotoxicity assay

NS3-peptide-specific T-cell lines were tested for their capacity to kill peptide-loaded target cells, using a new non-radioactive cytotoxicity assay. In this assay, a panel of matched Patr MHC-class-I expressing transfectants, either constructed in 721.221 cells (kindly provided by C. Walker, Center for Vaccines and Immunity, The Ohio State University) or K562 [15] were used as target cells. In brief, 3×106 MHC-matched transfected target cells were labeled with 1 µM CFSE (CFSElow) (Fluka) and another 3×106 MHC-matched transfected target cells, expressing the same class-I molecule were labeled with 8 µM CFSE (CFSEhigh). Subsequently, CFSElow cells were incubated with medium alone, whereas, CFSEhigh target cells were pulsed with HCV-peptide (8 µg/ml) for 2 hrs at 37°C. Excess peptide was removed by washing before mixing CFSEhigh and CFSElow with 1×106 NS3-specific expanded T-cells (ET ratio 10∶1). The mixed target and effector cells were cocultured for 16 hrs, after which CFSE staining was analyzed using a FACSAria (BD Bioscience) and DIVA software Version 6.1.1.

Specific lysis was calculated using the following formula;

and lysis was considered positive when exceeding 25%.

Peptide-binding assay

A cell-based peptide-binding-competition-assay was performed as previously described [15]. In this assay, peptides are tested for binding to transfected cell lines expressing chimpanzee class-I molecules, while competing with a biotin-labeled reference-peptide. In a 96 wells plate, 100.000 MHC class-I transfected K562 or 721.221 cells were incubated with 500nM of the biotinylated-indicator-peptide (in case of A*03∶01 ATALECVYK was used, and in case of B*01∶01 LSDMHLCSI, in which C indicates the position of the biotin-labeled cysteine) and 0.05 to 100 nM of the peptide of interest, for instance p59 (NS31258–1272), LGFGAYMSK or AATLGFGAY. Peptide KGGLRPRAG, predicted to have no or a very low binding affinity to MHC class I molecules based on the incorporation of glycines on the anchor positions, was used as control. After overnight incubation, excess peptide was removed by washing and indicator-peptide bound to cells was quantified using the DELFIA-system and the Victor3 1420 multilabel counter (Perkin Elmer). The binding affinity of the indicator peptide was verified using its unlabeled peptide as competitor, this is referred to as standardized competitor [15].


Characterization of NS3 specific CD4 and CD8 T-cell cytokine responses

We previously reported [11] an HCV-vaccine evaluation-study in chimpanzees, in which 4 out of 4 immunized animals became viremic shortly after challenge with HCV 1b J4. However, at a later stage, animal Vac1 cleared HCV infection while the three other animals, Vac2, Vac3 and Vac4 became chronically infected. As reported previously, IFNγ release, ex vivo lymphoproliferative responses, Th1/Th2 cytokine release patterns, and HCV-specific antibodies were comparable amongst the four animals [11]. NS3 was the immunodominant antigen in all four animals, while the responses against core, E1 and E2 were not only less intense, but also largely absent in Vac1, the animal which was protected against chronic infection. Therefore, functional analysis of the cellular responses was focused on the immunodominant NS3 antigen [11].

After 12 days of culturing in the presence of a pool of NS3 peptides, T-cell lines were restimulated with individual peptides and analyzed for intracellular expression of IFNγ, IL-2 and TNFα using the gating strategy shown in Figure S1. Cells within the lymphocyte gate that were CD3 positive, and CD14 and CD20 negative were selected and subsequently IL-2, IFNγ and TNFα single, dual or triple cytokine expression was analyzed in the CD4 and CD8 T-cell subsets within the T-cell lines. Cytokine expression profiles varied per peptide as illustrated in Figure 1. For instance, only moderate cytokine induction was observed after peptide specific restimulation of expanded CD4 cells in Vac1. In contract, strong dual IFNγ and IL-2 expression was seen in expanded CD8 T-cells restimulated with peptide 59 (NS31258–1272), while medium alone or peptide 64 (IRTGVRTITTGGPIT) did not induce cytokine production. In comparison, in Vac2 p59 only gave a marginal cytokine induction in expanded CD8 T-cells, while p64 induced moderate dual IFNγ/IL-2 and IFNγ/TNFa cytokine production (Figure 1). Instead, expanded CD4 T-cells in Vac2, showed stronger response towards p59.

Figure 1. Evaluation of peptide-specific IL-2, IFNγ and TNFα expression in CD4 and CD8 T-cells.

A representative example of the analysis performed in Vac1 (left panel) and Vac2 (right panel), two weeks after the final vaccine boost, is shown. PBMC were first expanded for a 12 day period using a pool of all NS3 peptides, restimulated with individual peptides (medium alone, p59 or p64) and analyzed for induction of IFNγ, IL-2 and TNFα cytokine expression by expanded CD4 and expanded CD8 cells, using the gating strategy described in Figure S1. The numbers in the quadrants, represent the percentage of positive cells calculated from the parent-population.

In this manner the responses against each individual NS3 peptide was investigated and the results are summarized in Figure 2 in the form of heat maps; showing the percentages of IFNγ producing CD4 (Figure 2(A)) and IFNγ/IL-2 dual cytokine producing CD8 (Figure 2(B)) T-cells in all four animals at all time points tested. In general, analysis of peptide responses by dual cytokine expression gives a clear-cut distinction between positive and negative responses, because of low background responses. However, for CD4 T-cells the number of dual cytokine expressing cells was often rather low, making it necessary to consider the total number of IFNγ cytokine producing cells. The peptide recognition profiles shown in Figures 2(A and B) indicate that some regions within the NS3 peptide sequence, indicated by red boxes, are broadly recognized amongst the animals. Other peptide specific responses were more restricted and only gave positive responses at some time points in one to two animals. No unique peptide response-pattern could be discerned for Vac1 that could explain vaccine induced viral clearance as compared to Vac2, 3, and 4. Moreover, no relation was observed between the number of peptides recognized and the course of infection (Table 2).

Figure 2. Heat map of NS3-peptide-specific T-cell responses in four immunized chimpanzees at different study time points.

(A) Percentage of expanded CD4 T-cells producing IFNγ upon restimulation with individual peptides, subdivided into three categories; i.e. low response (light green; 0.001%–0.1% specific IFNγ production), intermediate (green; 0.1–1% specific response), high (dark green;>1% specific response). (B) Percentage of expanded CD8 T-cells showing IL-2/IFNγ double cytokine production upon restimulation with individual peptide, subdivided into three categories; i.e. low response (light blue; 0.01%–0.1% specific IFNγproduction), intermediate (blue; 0.1–1% specific response), high (dark blue; > 1% specific response). White areas indicate where no responses were detected. The red boxes highlight areas within NS3 that are broadly recognized. Per animal 6 time points of analysis are shown lined up beneath each other; -12) Two weeks following the 1st MVA boost, cells stimulated with NS3vaccine, -4) Four weeks following 2nd MVA boost, cells stimulated with NS3vaccine, 0) Day of HCV infection, stimulation with NS3vaccine, 0) Day of HCV infection, stimulation with NS3challenge, and 4 and 36 weeks after challenge, stimulation with NS3challenge. The individual peptides tested are indicated at the top of the map by the numbers 1 to 156. The red boxes represent broadly recognized regions within NS3. Total frequency (all peptide responses combined) of (C) IFNγ production by expanded CD4 T-cells and (D) IFNγ/IL2 dual cytokine production by expanded CD8 T-cells per animal. (E) Frequency of p59 specific triple positive IL2/TNFα/IFNγ producing CD8 T-cells. The numbers on the X-axis represent the study week, relative to time of challenge. * Despite several attempts, frozen cells from Vac1 from 4 weeks following HCV infection did not respond to either NS3vaccine or NS3challenge peptide stimulation and no expansion could be achieved. @ Due to a high IFNγ background, peptide specific CD8 responses could not be detected

Table 2. Number of NS3-peptides that induce IFNγ production by CD4 T-cells or IL-2/IFNγ dual cytokine production by CD8 T-cells.

Next, the sum of the % of cytokine producing cells induced by all 156 individual peptides was calculated for expanded CD4 and CD8 lines. This total frequency of peptide-specific IFNγ-producing expanded CD4 T-cells was comparable between Vac1 and Vac2, but somewhat lower in Vac3 and Vac4 (Figure 2(C)). After the last immunization (week -4), total peptide specific CD8 T-cell responses were higher in Vac1 as compared to the other animals (Figure 2(D)). Triple cytokine production, i.e. simultaneous expression of IFNγ, IL-2 and TNFα by expanded CD8 cells, was also higher in Vac1 and for instance observed in 12.3% of the p59-specific expanded CD8 T-cells 4 weeks prior to HCV infection, whereas this was less than 7% in Vac2, Vac3 and Vac4 (Figure. 2(E)).

In conclusion, we observed increased number of NS3 specific cytokine producing expanded CD8 T-cells after the last immunization, and especially more triple IFNγ, IL-2 and TNFα cytokine producing cells in the animal that cleared the virus, relative to those animals that became chronically infected.

Cytolytic capacity of NS3 p59 specific CD8 T-cells in Vac1

Next, NS3-peptide-specific cells were evaluated for their capacity to kill peptide-loaded target cells. Similar to standard 51chromium release assays, T-cells were expanded prior to analysis [6], [16]. The expanded T-cells were tested for their lytic capacity for all individual peptides that were found to induce IL-2/IFNγ dual expression in expanded CD8 cells (Figure 2(B)) and was tested against a panel of MHC matched target cells [15] (Table 1).

Figure 3(A) illustrates specific killing of p59-pulsed CFSEhigh relative to unpulsed CFSElow Patr-A*03∶01 target cells, using expanded T-cells from Vac1, isolated 10 weeks after HCV challenge. As a control, no difference in ratio of the area under the curve was observed if both the CFSEhigh and CFSE low Patr-A*03∶01 targets were unpulsed (Figure 3(C)), precluding non-specific toxicity due to labeling with higher CFSE concentration. Killing of target cells was only observed for particular peptide-MHC class I combinations. As shown in Figures 3(B and D), neither unpulsed nor p59-pulsed Patr-B*02∶01 targets were lysed by expanded cells from Vac1.

Figure 3. Cytolytic killing of CFSE-loaded target cells.

(A) Specific lysis of NS31258–1272-pulsed Patr-A*03∶01 target cells (red peak) in comparison with unpulsed Patr-A*03∶01 target cells (yellow peak) by NS3-peptide-pool expanded PBMC from Vac1 isolated 10 weeks after HCV 1bJ4 challenge (blue peak). (B) No specific lysis of NS31258–1272-pulsed Patr-B*02∶01 target cells. (C) and (D) Control experiments showing no killing of unpulsed Patr-A*03∶01 or Patr-B*02∶01 target cells. (E,F,G) Percentage of lysis for all MHC class-I-peptide-combinations tested in Vac1, Vac2, Vac3 respectively.

Amongst all peptides tested, i.e. 24, 22 and 35 in Vac1, Vac2 and Vac3 respectively, highly efficient killing was only observed against p59 in Vac1 and only when this peptide was presented by Patr-A*03∶01 (Figure 3(E to G)). Only the peptides partly overlapping with p59 gave some lysis, while none of the other MHC class-I-peptide-combinations exceeded 25% of specific killing. Due to the limited number of cells after expansion, Vac4 could not be tested.

Collectively, the data show that although stimulation with p59 can induce some level of cytokine production in all vaccinees, a p59-mediated cytolytic capacity was only observed in Vac1.

NS3p59(aa1258–1272) binds with high affinity to Patr-A*03∶01

Peptide binding to MHC-class I is determined by the anchor-residues fitting into the B- and the F-pocket of the relevant allotype for most MHC class-I-molecules. Usually MHC class I molecules bind nonamer peptides, in which the p2-position of the peptide anchors into the B-pocket and the p9-position of the peptide anchors into the F-pocket. Peptide 59, which was observed to initiate cytolysis when presented in the context of Patr-A*03∶01, was scanned using the NetMHCpan version 2.8 prediction algorithm [17] against the MHC class I molecules present in the studied cohort (Table 1). It is known that peptides that bind, or are predicted to bind to MHC class I molecules with an affinity <500 nm are potential T-cell epitopes [18][20]. In p59 four potential nonamer peptides are present, AATLGFGAY, ATLGFGAYM, TLGFGAYMS, and LGFGAYMSK. The only combination that was predicted to bind with an affinity <500 nm was LGFGAYMSK in the context of Patr-A*03∶01. A cell-based peptide binding competition assay was performed to test the binding-affinity of p59, and the nonamer peptides AATLGFGAY and LGFGAYMSK to Patr-A*03∶01 and as control to Patr-B*01∶01. The peptide binding motif of Patr-A*03∶01 is defined as X(STA)XXXXXX(RK) and for Patr-B*01∶01 as (X(ST)XXXXXX(IL) [15], [21]. Figure 4 shows that all three NS3 peptides have a high binding affinity to Patr-A*03∶01, with IC50 values between 0.03 to 0.2 µM, which is higher compared to the control peptide (IC50 = 8.2 µM) and the standardized competitor peptide of Patr-A*03∶01 with IC50 = 1.5 µM [15]. Peptide LGFGAYMSK was found to have the highest binding affinity to Patr-A*03∶01. In contrast, only low binding affinity was observed for p59 in the context of Patr-B*01∶01 as compared to its standardized competitor (IC50 = 1.0 µM), whereas for the two nonamer peptides, no amenable regression curves could be drawn suggesting low or no binding affinity (not shown).

Figure 4. Dose-inhibition curves for selected NS3-peptides to Patr-A*03∶01.

Binding affinity of selected NS3-peptides to Patr-A*03∶01, established by peptide binding competition assay. Patr-A*0301 cells were incubated with increasing concentrations of NS3 p59 (top left), LGFGAYMSK (LGF, top right), AATLGFGAY (AAT, bottom left), or KGGLRPRAG (control, bottom right) in the presence of biotin-labeled indicator peptide. Binding of indicator peptide was quantified by incubation with europium labeled streptavidin. Indicated is reduction in percentage of europium positive cells relative to cells incubated with indicator-peptide only. The indicated IC50 values (µM) are derived from the regression curves of three independent experiments in the case of p59, LGF and AAT, and of two individual experiments in the case of KGG. The IC50 values were estimated using non-linear least-squares regression with the “R” platform for statistical computing.

T-cells of Vac1 respond at lower peptide concentration

Antigen specific responses not only depend on the presentation of peptides by MHC class I molecules but also on the binding of the peptide/MHC complex to the T-cell-receptor (TCR). Therefore, we tested whether the response-threshold of CD4 and CD8 T-cell lines from Vac1 differed from Vac2 and Vac3. To this end, NS3-specific expanded T-cells, isolated 4 weeks prior to HCV infection, were expanded from PBMCs, restimulated with different concentrations of NS31258–1272 ranging from 0.05 to 10 µg/ml, and stained for intracellular IL-2 and IFNγ. Expanded CD8 T-cells from Vac1 were found to respond at a peptide concentration as low as 0.05 µg/ml whereas, at least, 20 times more peptide was required for CD8 T-cells from Vac2 and Vac3 (Figure 5). For IFNγ production by expanded CD4 T-cells the difference is even more striking. Here, we could hardly detect p59-specific IFNγ producing CD4 T-cells in Vac2 and Vac3 while these cells are abundantly present in Vac1.

Figure 5. Antigen sensitivity of p59 specific cytokine induction.

NS3-expanded T-cells from Vac1, Vac2 and Vac3, from 4 weeks prior to HCV challenge, were restimulated with a concentration range of p59 and evaluated for induction of IL-2/IFNγ expressing CD8 T-cells (left graph) and IFNγ expressing CD4 T-cells (right graph).


The chimpanzee is the closest living relative of man and the only validated animal model for HCV infection. The evolutionary proximity of both species is reflected by close genetic and immunological similarity. For these reasons the chimpanzee provides an exquisite animal model, in which immune responses against HCV vaccines and the effect on deliberate infection can be investigated in an experimental setting [2]. Here, we report of an in depth profiling of CD4 and CD8 T-cell responses in four chimpanzees, which were immunized with an experimental vaccine for HCV and subsequently challenged with HCV 1b J4 [11].

While all four animals became infected, only the Patr-A*03∶01 positive animal was able to clear HCV infection. Detailed mapping of CD4 and CD8 T-cell cytokine responses, and the analysis of their cytolytic potential, revealed that expanded T-cells of the vaccinee which effectively cleared HCV infection displayed: a) a higher frequency of peptide-specific IL-2/IFNγ dual cytokine producing CD8 T-cells; b) peptide-specific TNFα production in CD8 T-cells; c) cytolytic capacity against peptide-loaded target cells; d) cytokine induction at a low antigen concentration for the cytolysis-inducing peptide. Remarkably, the cytolysis-inducing peptide p59 was found to bind strongly to Patr-A*03∶01, which was only present in Vac1. Using the NetMHCpan algorithm, it was predicted that peptide LGFGAYMSK, one of the nonamer peptides present in the 15-mer p59, would show the strongest binding to Patr-A*03∶01. Peptide LGFGAYMSK was also predicted to bind to the MHC class I molecules Patr-A*01∶01, A*04∶01, A*04∶04, and A*09∶01, present in all of the studied animals, but only with intermediate affinity between 0.9 and 5 µM. These data are in agreement with the differences in cytokine responses and cytolyses reported in expanded T-cells here and could indicate a particular affinity threshold is necessary for a peptide to induce adequate immune responses for combating HCV. The strong binding of the 15-mer p59 to Patr-A*03∶01 is somewhat unexpected, as generally nonamer peptides bind to MHC class I molecules. Nonetheless, in humans it has been shown that particular MHC class I molecules are capable of binding longer peptides with bulged structures [22], [23].

Stimulation assays would give information about the biological relevance of the binding experiments. However, the limited number of cells stored from these animals, preclude performing such assays.

Unlike ex vivo detection of CTL in humans, detection of these cells in chimpanzees can often only be achieved after amplification of antigen specific responses [6], [16]. Obviously, in vitro expansion of T-cells from frozen PBMC has its limitations. Antigen presenting cells may be affected by the freezing/thawing process resulting in –partly hampered- antigen presentation. Furthermore, the addition of cytokines and peptides may skew the immune system differently as compared to the in vivo situation.

To determine the degranulating capacity of T-cells, CD107a stainings were performed on T-cell lines. However, the results of these assays were unclear and no conclusion could be drawn from it, basically because of the lack of a proper negative control.

Unlike in the intracellular cytokine staining assays, where the difference between cells restimulated with peptide or medium alone was very clear, no clear difference could be observed when CD107a was used. Probably the expansion-protocol used here does agree with antigen specific CD107a staining. Unfortunately, the number of stored cells from these animals is limited and assays like this require too many cells to repeat the assays.

And although we cannot exclude that upon further amplification of antigen specific responses, CTL responses might also be detectable in the other animals, our data suggest that they are of lower magnitude.

To study the possibility of viral escape from immune pressure as a cause of development of chronicity in Vac2, Vac3 and Vac4, analysis of viral sequences was performed and compared with the inoculum sequence (data not shown). Only a limited number of amino acid substitutions were detected in the areas that overlap with cytokine-inducing peptides. There was however no evidence for mutational escape in the immunodominant p59 in any of the animals.

This study highlights a potential role for CD8 mediated cytolytic responses as a discriminatory factor between viral clearance and the development of chronic infection in HCV vaccinated subjects and underscores that this parameter is to be considered as an important immunological readout in future vaccine trials. Indeed, studying peptide binding and CD8 cytolytic responses could complement the other contributing immune mechanisms previously described [11], [24], [25].

It should be noted that the reported data were obtained from peripheral blood and not from the site of HCV-induced inflammation, the liver. For ethical reasons and animal welfare, the number of liver biopsies taken during the study, was limited and not sufficient for the comprehensive analysis as presented in this manuscript. Nevertheless, a direct comparison of CD8 T-cell function between liver and peripheral blood in chimpanzees during HCV infection, showed in essence that the cells in the circulation reflect the situation in the liver [26], [27].

The data presented here, are in line with the finding that self-limiting HCV infection in humans is associated with the presence of HLA-B*57 and HLA-B*27 restricted CD8 T-cell responses [28], [29]. Chimpanzee Patr-A*03∶01 and Patr-B*01∶01 have almost identical B-pocket residues as compared to HLA-B*57. Moreover, peptide binding studies have shown that both Patr-molecules may potentially bind HLA-B*57 or HLA-B*27 [15].

In conclusion, we observed that in contrast to peptide-specific cytokine production, the induction of an effective CD8 T-cell effector function elicited in the context of peptide presentation by appropriate class-I molecules appeared to be associated with clearance of infection.

Supporting Information

Figure S1.

Gating strategy for the evaluation of cytokine production by CD4 and CD8 cells. Cells from Vac1, 2 weeks after the last vaccine boost were stimulated with NS3vaccine and restimulated with either NS31258–1272 (p59) or medium alone. Cells within the lymphogate were selected, followed by selection of the CD3 positive, CD20/CD14 negative population, subsequently CD4 and CD8 positive cells were selected. Expression of IFNγ as a function of IL-2 or TNFα was plotted.


Table S1.

Peptide numbering of Vaccine- and Challenge-sequence relative to HCV reference strain H77.



The authors would like to thank Thea de Koning for editing the manuscript and Henk van Westbroek for preparing the graphics. The NS3-peptides were provided by BeiResources from ATCC.

Author Contributions

Conceived and designed the experiments: BV EV PM NdG RB WB JH GK. Performed the experiments: BV NdG ZF. Analyzed the data: BV NdG EV ZF PM GK. Contributed reagents/materials/analysis tools: BV NdG EV. Wrote the paper: BV GK NdG PM.


  1. 1. Bettauer RH (2010) Chimpanzees in hepatitis C virus research: 1998–2007. J Med Primatol 39: 9–23.
  2. 2. Bukh J (2012) Animal models for the study of hepatitis C virus infection and related liver disease. Gastroenterology 142: : 1279–1287 e1273.
  3. 3. Bacchetti P, Deeks SG, McCune JM (2011) Breaking free of sample size dogma to perform innovative translational research. Sci Transl Med 3: 87ps24.
  4. 4. Thimme R, Bukh J, Spangenberg HC, Wieland S, Pemberton J, et al. (2002) Viral and immunological determinants of hepatitis C virus clearance, persistence, and disease. Proc Natl Acad Sci U S A 99: 15661–15668.
  5. 5. Diepolder HM, Gruener NH, Heeney J (2003) Of chimps and men—finding the path to an HCV vaccine. Hepatology 37: 1472–1474.
  6. 6. Cooper S, Erickson AL, Adams EJ, Kansopon J, Weiner AJ, et al. (1999) Analysis of a successful immune response against hepatitis C virus. Immunity 10: 439–449.
  7. 7. Watanabe H, Wells F, Major ME (2010) Clearance of hepatitis C in chimpanzees is associated with intrahepatic T-cell perforin expression during the late acute phase. J Viral Hepat 17: 245–253.
  8. 8. Osburn WO, Fisher BE, Dowd KA, Urban G, Liu L, et al. (2010) Spontaneous control of primary hepatitis C virus infection and immunity against persistent reinfection. Gastroenterology 138: 315–324.
  9. 9. Prince AM, Brotman B, Lee DH, Pfahler W, Tricoche N, et al. (2005) Protection against chronic hepatitis C virus infection after rechallenge with homologous, but not heterologous, genotypes in a chimpanzee model. J Infect Dis 192: 1701–1709.
  10. 10. Nascimbeni M, Mizukoshi E, Bosmann M, Major ME, Mihalik K, et al. (2003) Kinetics of CD4+ and CD8+ memory T-cell responses during hepatitis C virus rechallenge of previously recovered chimpanzees. J Virol 77: 4781–4793.
  11. 11. Rollier CS, Paranhos-Baccala G, Verschoor EJ, Verstrepen BE, Drexhage JA, et al. (2007) Vaccine-induced early control of hepatitis C virus infection in chimpanzees fails to impact on hepatic PD-1 and chronicity. Hepatology 45: 602–613.
  12. 12. Haaijman JJ, Heeney JL (1995) European centralized chimpanzee facility 'preclinical studies for evaluation of effective vaccines and therapeutic strtegies for AIDS'. AIDS research at EC level. IOS press. pp. 139–148.
  13. 13. Hoffmeister B, Kiecker F, Tesfa L, Volk HD, Picker LJ, et al. (2003) Mapping T cell epitopes by flow cytometry. Methods 29: 270–281.
  14. 14. de Groot NG, Otting N, Arguello R, Watkins DI, Doxiadis GG, et al. (2000) Major histocompatibility complex class I diversity in a West African chimpanzee population: implications for HIV research. Immunogenetics 51: 398–409.
  15. 15. de Groot NG, Heijmans CM, Zoet YM, de Ru AH, Verreck FA, et al. (2010) AIDS-protective HLA-B*27/B*57 and chimpanzee MHC class I molecules target analogous conserved areas of HIV-1/SIVcpz. Proc Natl Acad Sci U S A 107: 15175–15180.
  16. 16. Balla-Jhagjhoorsingh SS, Koopman G, Mooij P, Haaksma TG, Teeuwsen VJ, et al. (1999) Conserved CTL epitopes shared between HIV-infected human long-term survivors and chimpanzees. J Immunol 162: 2308–2314.
  17. 17. Hoof I, Peters B, Sidney J, Pedersen LE, Sette A, et al. (2009) NetMHCpan, a method for MHC class I binding prediction beyond humans. Immunogenetics 61: 1–13.
  18. 18. Allen TM, Mothe BR, Sidney J, Jing P, Dzuris JL, et al. (2001) CD8(+) lymphocytes from simian immunodeficiency virus-infected rhesus macaques recognize 14 different epitopes bound by the major histocompatibility complex class I molecule mamu-A*01: implications for vaccine design and testing. J Virol 75: 738–749.
  19. 19. Loffredo JT, Sidney J, Wojewoda C, Dodds E, Reynolds MR, et al. (2004) Identification of seventeen new simian immunodeficiency virus-derived CD8+ T cell epitopes restricted by the high frequency molecule, Mamu-A*02, and potential escape from CTL recognition. J Immunol 173: 5064–5076.
  20. 20. Sette A, Vitiello A, Reherman B, Fowler P, Nayersina R, et al. (1994) The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes. J Immunol 153: 5586–5592.
  21. 21. Sidney J, Peters B, Moore C, Pencille TJ, Ngo S, et al. (2007) Characterization of the peptide-binding specificity of the chimpanzee class I alleles A 0301 and A 0401 using a combinatorial peptide library. Immunogenetics 59: 745–751.
  22. 22. Paul S, Weiskopf D, Angelo MA, Sidney J, Peters B, et al. (2013) HLA class I alleles are associated with peptide-binding repertoires of different size, affinity, and immunogenicity. J Immunol 191: 5831–5839.
  23. 23. Bell MJ, Burrows JM, Brennan R, Miles JJ, Tellam J, et al. (2009) The peptide length specificity of some HLA class I alleles is very broad and includes peptides of up to 25 amino acids in length. Mol Immunol 46: 1911–1917.
  24. 24. Fattori E, Zampaglione I, Arcuri M, Meola A, Ercole BB, et al. (2006) Efficient immunization of rhesus macaques with an HCV candidate vaccine by heterologous priming-boosting with novel adenoviral vectors based on different serotypes. Gene Ther 13: 1088–1096.
  25. 25. Zubkova I, Choi YH, Chang E, Pirollo K, Uren T, et al. (2009) T-cell vaccines that elicit effective immune responses against HCV in chimpanzees may create greater immune pressure for viral mutation. Vaccine 27: 2594–2602.
  26. 26. Shin EC, Capone S, Cortese R, Colloca S, Nicosia A, et al. (2008) The kinetics of hepatitis C virus-specific CD8 T-cell responses in the blood mirror those in the liver in acute hepatitis C virus infection. J Virol 82: 9782–9788.
  27. 27. Ramalingam RK, Meyer-Olson D, Shoukry NH, Bowen DG, Walker CM, et al. (2008) Kinetic analysis by real-time PCR of hepatitis C virus (HCV)-specific T cells in peripheral blood and liver after challenge with HCV. Journal of virology 82: 10487–10492.
  28. 28. Kim AY, Kuntzen T, Timm J, Nolan BE, Baca MA, et al.. (2011) Spontaneous control of HCV is associated with expression of HLA-B 57 and preservation of targeted epitopes. Gastroenterology 140: : 686–696 e681.
  29. 29. Neumann-Haefelin C, Timm J, Schmidt J, Kersting N, Fitzmaurice K, et al. (2010) Protective effect of human leukocyte antigen B27 in hepatitis C virus infection requires the presence of a genotype-specific immunodominant CD8+ T-cell epitope. Hepatology 51: 54–62.