The influence of T cell cross-reactivity on HCV-peptide specific human T cell response


  • This study was approved by the local Ethics committees at the Royal Free Hospital and University College London Hospitals, London.

  • Potential conflict of interest: Nothing to report.


Detection of hepatitis C virus (HCV)-specific T cell response after exposure to hepatitis C in anti-HCV–positive or anti-HCV–negative patients has been associated with an ability to successfully control the infection. However, cross-reactivity between common human pathogens and HCV sequences has been demonstrated. The aim of this study was to investigate the impact of T cell cross-reactivity on HCV-specific T cell responses and their detection in HCV infected and non-infected subjects. The magnitude, function, and cross-reactivity of HCV peptide reactive T cells were studied in non–HCV-infected newborns and adults using a broad array of HCV peptides (601 peptides) spanning the entire HCV sequence. Comparisons were made with responses present in recovered and in chronically HCV-infected patients. HCV peptide reactive T cells are detectable in adults irrespective of previous HCV exposure and cross-reactivity between HCV peptides, and sequences of common pathogens, such as human herpes virus 1, can be demonstrated. Furthermore, the comprehensive magnitude of HCV-peptide reactive T cells present in chronically HCV-infected patients is similar and in some cases even lower than that of HCV-peptide reactive T cell response found in HCV-negative adults. In conclusion, the presence of oligo-specific HCV-peptide reactive T cells in humans does not always reflect a demonstration of previous HCV contact, whereas cross-reactivity with other common pathogens can potentially influence the HCV-specific T cell profile. The conspicuous deficit of HCV-peptide–specific T cells found in chronically HCV-infected patients confirms the profound collapse of virus-specific T cell response caused by HCV persistence. (HEPATOLOGY 2006;43:602–611.)

Newer immunological tools have improved in vitro analyses of the prevalence and function of antigen-specific CD4 and CD8 T cell responses in persistent viral infections. The most significant of these advances have been the capability to directly visualize epitope-specific cells (MHC class I and class II tetrameric complexes)1, 2 and the ability to measure the secretion of T cell cytokines (intracellular cytokine staining, ELISPOT).3, 4 Furthermore, the understanding that the in vivo T cell stimulatory complex, formed by MHC molecules and processed viral peptide, can be mimicked in vitro by the addition of synthetic peptides5 to PBMC has evolved into their direct use in the analysis of specific T cell responses. Measurement of T cell reactivity using large pools of synthetic peptides spanning different pathogenic antigens is becoming the standard for the study of pathogen-specific T cell repertoire in humans. This method is particularly useful in the analysis of T cell responses against viruses that do not easily infect antigen presenting cells, such as hepatitis C virus (HCV).6–10

Type C hepatitis, caused by HCV, is believed to affect approximately 170 million people worldwide and is a leading cause of chronic liver disease. Hepatitis C is a preferentially hepatotropic non-cytopathic RNA virus. Different patterns of HCV-specific T cell response are demonstrable in recovered versus chronically infected HCV patients. Historically this differential response was analyzed using limited arrays of HCV proteins and peptides.6, 11–14 More recently, the use of comprehensive panels of synthetic peptides spanning the entire HCV polyprotein have confirmed and detailed further the broader and stronger HCV-specific T cell response profile present in resolved HCV infection.8, 10 The contrast between this response and the profound quantitative and functional defects of HCV-specific T cells observed in chronically infected subjects has been emphasized by this technique.

However, the use of such broad panels of peptides could increase the possibility of detection of HCV-peptide reactive T cells that have not been initially primed by HCV virions, but are cross-reactive T cells triggered by unrelated antigens. Indeed, T cells present an intrinsic degeneracy in their recognition of peptides. Single T cells can recognize different peptides that present variable degrees of amino acid (AA) homology15, 16 and this degeneracy of T cell recognition represents the structural basis of the phenomenon of cross-reactivity.17 CD8+ T cell cross-reactivity between HCV and influenza A virus determinants have recently been demonstrated18 and could have a potential impact on the profile of disease associated with HCV infection.19

Cross-reactivity could also account for the reported presence of HCV-peptide reactive T cells in subjects who lack all serological markers of HCV infection.20, 21 These results were interpreted in the past as indicative of previous HCV exposure, because patients acutely infected with HCV can maintain long-term HCV-specific cellular immunity still detectable after loss of humoral responses.22 However, T cell cross-reactivity among homologous peptides covering sequences of different pathogens cannot be excluded. This phenomenon has obvious implications in data analysis; however, it can also influence the outcome of HCV infection if HCV infects an individual with pre-existing memory responses, which can be rapidly and efficiently recalled by homologous HCV sequences. In this context, the protective efficacy of HCV-peptide reactive T cells found in HCV-exposed but ostensibly non-infected subjects may be more difficult to predict if these cells merely represent T cells specific for unrelated pathogens cross-reacting with HCV peptides but not primed by HCV.

In this study, we set out to analyze whether cross-reactive T cells activated by HCV peptides constitutes a common occurrence in humans. The magnitude, function, and cross-reactivity of HCV peptide reactive T cells were studied in non–HCV-infected newborns and adults. Comparisons were made with responses present in both resolved HCV infection and in chronically infected patients. We demonstrate that non–HCV-infected adults possess HCV-peptide reactive T cells, which can react with peptide sequences of common pathogens. The impact of cross-reactive T cells in the interpretation of the data found in HCV-infected subjects and their potential ability to alter the immunopathological profile of viral infection is discussed.


HCV, hepatitis C virus; NC, negative control; SP, sexual partners; HHV1, human herpes virus; IFN-γ, interferon gamma; PBMC, peripheral blood mono-nuclear cell; IL, interleukin; SFC, spot-forming cells; HBV, hepatitis B virus.

Material and Methods


Thirty-two subjects were included in the study. Blood was obtained from ten healthy volunteers, constituting the negative control group (NC, subjects 1-10); all were anti-HCV negative and without any HCV risk factors. A further six subjects, the sexual partners (SP, subjects 11-16) of six chronically infected HCV patients, were studied. Patients with resolved (subjects 17-22) and chronic (subjects 23-32) HCV infection were recruited from the viral hepatitis clinics at the Royal Free Hospital, London, and Divisione Malattie Infettive, Parma. All patients with chronic HCV had raised serum alanine aminotransferase, detectable anti-HCV, and HCV RNA for a period of more than 6 months. Patients with resolved HCV were anti-HCV positive (ELISA and RIBA) on two occasions over a 6-month period; however, HCV RNA was undetectable by polymerase chain reaction (PCR). The age, sex, disease profiles, serological status, risk factors, and alanine aminotransferase levels for all subjects are outlined in Table 1. Cord blood from anti-HCV–negative mothers (n = 5) was provided through our collaboration with the Department of Obstetrics and Gynaecology, UCL. The study was approved by the local Ethics Committee at the Royal Free and University College Hospitals.

Table 1. Characteristics of the Study Groups
Subject no.DiagnosisHCV RNAAnti-HCVRisk FactorsALTAgeSex
  1. NOTE. NC had no exposure to or risk factors for HCV. None of the individuals recruited were laboratory or hospital workers.

  2. Abbrevations: NC, negative control; SP, sexual partner; + denotes presence of known risk factors.


Synthetic Peptides and Antibodies.

Synthetic peptides representing a panel of 601 15-mer peptides overlapping by 10 residues and spanning the entire sequence of HCV-1 were purchased from Chiron Mimotopes (Victoria, Australia). Peptides homologous to HCV reactive peptides covering selected sequences of human herpes virus (HHV1) and vaccinia virus were purchased from Primm it (Milano, Italy). Anti-CD8 (conjugated with Quantum Red or fluorescein isothiocyanate; FITC) and anti-interferon gamma (IFN-γ) FITC were purchased from SIGMA Aldrich (St. Louis, MI). Anti-perforin (FITC) was purchased from BD Pharmingen.

Isolation of Peripheral Blood Mononuclear Cells and In VitroExpansion of Hepatitis B Virus and HCV-Specific Cytotoxic T Cells.

Peripheral blood mononuclear cells (PBMC) were isolated from fresh heparinized blood or cord blood by Ficoll-Hypaque density gradient centrifugation and suspended in RPMI 1640 supplemented with 25 mmol/L Hepes, 2 mmol/L L-glutamine, 50 μg/mL gentamycin, and 8% human serum (complete medium) and used immediately in ELISPOT assays or expanded in vitro. For in vitro expansion, PBMC were suspended in 96-well plates at a concentration 1.5 × 106/mL in complete medium and stimulated with peptides at 1 μmol/L final concentration. Recombinant IL-2 was added on day 4 of culture (50 UI/mL) and the immunological assays were performed on days 8 through 10.

IFN-γ Intracellular Staining.

Freshly separated PBMC or cells obtained after in vitro expansion were incubated in medium alone (control) or with viral peptides (1 μmol/L) for 1 hour; Brefeldin A (10 μg/mL) was added for an additional 4 hours of incubation. After washing the cells were stained with anti-CD8 or anti-CD4 quantum red monoclonal antibody for 20 minutes at 4°C, and then fixed and permeabilized as described. Cells were stained with anti–IFN-γ-FITC for 15 minutes at room temperature, washed again, and analyzed on a Becton Dickinson flow cytometry (FACScalibur) using the CELLQuest software.


Six hundred and one 15-mer peptides based on the sequence of HCV-1 (genotype 1a covering all structural (core, E1, E2) and non-structural (NS3, NS4, NS5) HCV regions and overlapping by 10 residues were grouped in pools of 12 in a matrix array8 such that each peptide is present in two pools only.

HCV-specific T cell responses ex vivo were analyzed after overnight stimulation with individual peptide mixtures. Briefly, 96-well plates (Multiscreen-IP-Millipore S.A.S., Malshelm, France) were coated overnight at 4°C as recommended by the manufacturer with 5 μg/mL capture mouse anti-human IFN-γ monoclonal antibody (1 DIK, Mabtech, Sweden). Plates were then washed 7 times with phosphate-buffered saline/0.05% Tween 20, blocked with RPMI/10% fetal calf serum for 2 hours at 37°C. 2 × 105 PBMC were seeded per well, and peptides were added at a concentration of 10 μg/mL. After overnight incubation at 37°C, 5% CO2, the plates were washed with phosphate-buffered saline/0.05% Tween 20, and then 50 μL of 1 μg/mL biotinylated secondary mouse anti-human IFN-γ monoclonal antibody (7B6-1, Mabtech, Sweden) was added. After 3 hours of incubation at room temperature, plates were washed 4 times, and 100 μL goat alkaline phosphatase anti-biotin Ab (Vector Laboratories Inc. Burlingame, CA) was added to wells, and the plates were incubated for a further 2 hours at room temperature. Plates were then washed 4 times, and 75 μL alkaline phosphatase conjugate substrate (5-bromo-4-chloro-3-indolyl phosphate, Biorad Laboratories, Hercules, CA) was added. After 4 to 7 minutes, the colorimetric reaction was stopped by washing with distilled water. Plates were air-dried and spots were counted using an automated ELISPOT reader (AID ELISPOT Reader System, Autoimmune Diagnostika Gmbh, Strassberg, Germany). The number of specific IFN-γ–secreting PBMC was calculated by subtracting the number of spots obtained in the non-stimulated control well from the stimulated sample. A positive control (PBMC stimulated with PHA) was included with each plate to validate the sensitivity of the assay.

Multiple Cytokine Secretion Assays.

The production of multiple cytokines [IFN-γ, tumor necrosis factor alpha, interleukin-2 (IL-2), IL-4, IL-5, and IL-6) in the supernatant of the HCV-peptide reactive T cell lines was tested using a BD Cytometric Bead Array (BD Bioscience). One hundred microliters of supernatants of peptide stimulated or unstimulated HCV-peptide reactive T cell lines generated aspreviously described was collected after 18 hours of peptide stimulation. Production of cytokines was tested simultaneously with the assay according to the manufacturers' instructions.

Statistical Analysis.

The Mann-Whitney U test was used where appropriate to determine the significance of differences between groups; a P value of less than .05 was taken as significant.


Direct Ex Vivo Quantitation of ELISPOT Response in Cord Blood of HCV-Negative Infants.

To exclude the interference by the repertoire of memory T cells specific for unrelated pathogens present in adults, overlapping peptides covering the entire HCV polyprotein were initially tested on lymphocytes derived from cord blood of infants born to anti-HCV–negative mothers. The 601 peptides covering the entire HCV genome were grouped in a matrix array into pools of 12 and used to directly stimulate IFN-γ production from T cells in a direct ex vivo ELISPOT assay. Figure 1 shows the results of all five tests. PHA-stimulated wells of cord blood detected at least 100 spot-forming cells (SFC)/well. In contrast, the HCV peptide-stimulated wells were predominantly negative. However, a number of individual mixtures did elicit a response higher than background, but this never exceeded 12 SFC/well in all the subjects analyzed.

Figure 1.

IFN-γ production by direct ex vivo ELISPOT analysis on cord blood: PBMC from cord blood from the 5 different indicated newborns were stimulated overnight with 106 pools of overlapping 15-mer peptides covering the whole HCV sequence and with PHA. IFN-γ, interferon gamma; PBMC, peripheral blood mononuclear cells; HCV, hepatitis C virus; SFC, spot forming cells.

We assume that cord blood represents the closest example of a truly naïve T cell repertoire; thus, we based our calculation of what constitutes a positive result in adults on these results. The mean of the maximum responses obtained in each of the five different cord blood specimens was calculated to be 10.2 SFC/well. HCV-peptide reactive T cell responses in adults were considered positive if the number of spots per well minus the background was at least 3 times the mean value of maximum responses found in cord blood (3 × 10 = 30 SFC/well, 150 SFC/106 PBMC).

Direct Ex Vivo Quantitation of ELISPOT Response in HCV-Infected and Non-Infected Subjects.

The pools of HCV peptides covering HCV proteins were then tested directly ex vivo in six patients with resolved HCV infection (anti-HCV positive, HCV-RNA negative on two different occasions 6 months after infection), 10 patients with chronic HCV infection and 16 subjects with no serological markers of HCV infection. Ten anti-HCV–negative subjects had no classical risk factors for HCV infection (NC 1-10; Table 1) and were considered the control group. A further six subjects were anti-HCV negative but all had reported exposure to HCV, (i.e., were sexual partners of HCV-RNA–positive patients with chronic hepatitis, (SP11-16; Table1)). The results of the direct ex vivo ELISPOT of two representative subjects from each group are shown in Fig. 2A.

Figure 2.

IFN-γ production by direct ex vivo ELISPOT analysis on adults: PBMC were stimulated overnight with 106 pools of overlapping 15-mer peptides covering the whole HCV sequence. (A) Results of 2 representative subjects from each group of HCV-infected and non-infected patients are shown. The line corresponds to a value of 30 SFC/2 × 105 PBMC. Responses superior to this value were considered positive. (B) Quantity of SFC stimulated by all peptide pools covering the whole HCV polyprotein. Each dot represents an individual patient for each group. (C) Number of positive peptide pools in each subject. Dots represent individual patients. IFN-γ, interferon gamma; PBMC, peripheral blood mononuclear cells; HCV, hepatitis C virus.

HCV-peptide reactive T cells were easily detectable ex vivo in resolved patients, (Fig. 2, Sub.17), in keeping with several previous studies. The overall responses to all different peptide pools present a median total magnitude of 850 (range, 710-2421) SFC/20 million cells. Some pools were able to elicit a response reaching a magnitude of 277 SFC/well, and further analysis with individual peptides shows the presence of at least three T cell epitopes recognized in each of the patients with resolved HCV infection.

The overall quantity of IFN-γ–producing T cells detected in PBMC from patients with chronic HCV infection and HCV-negative subjects was lower than that observed in resolved patients; median total magnitude of response to all peptide pools was 104 (range, 24-292) SFC/20million cells in chronic, 411 (range, 208-502) in NC, and 479.5 (range, 328-548) in SP; however, HCV-peptide reactive T cells could be detected in all tested subjects irrespective of their HCV infection status or risk factors. In keeping with previous studies, the T cell responses present in NC, SP, and chronically infected individuals lack the multispecificity found in patients with resolved HCV infection. The magnitudes of the individual responses were variable; however, in some cases (see Fig. 2, Sub. 2, where response to pool 14 was 106 SFC/2 × 105) HCV-peptide pools were able to elicit a magnitude almost identical to those found in resolved HCV infection. Interestingly, the direct ex vivo frequency of HCV-peptide reactive T cells found in patients with chronic HCV was lower than those present in HCV-negative patients (Fig. 2A-B). The difference in total SFC between these two groups (Fig. 2B) was found to be statistically significant (P < .0001).

To confirm that the IFN-γ production found by direct ex vivo ELISPOT analysis in HCV-negative patients was sustained by CD4 or CD8+ T cells, we generated HCV-peptide reactive T cell lines. CD4 or CD8+ T cells able to react to individual HCV peptides can be demonstrated in HCV-negative subjects and SP of HCV-infected patients (Fig. 3). Their profile of cytokine production was Th1/Tc1, with the ability to produce IFN-γ and tumor necrosis factor alpha but not IL-4, IL-5, or IL-2 after peptide-specific stimulation. (Fig. 3B).

Figure 3.

Characterization of HCV-peptide reactive T cells present in HCV-negative control subjects. (A) Dot plot analysis of IFN-γ–producing T cells specific for HCV-NS3 1436-50and HCV-NS5B 2811-25peptides. PBMC were stimulated with individual peptides and after 10 days, were re-stimulated with the stimulatory peptide. Frequency of IFN-γ–producing cells was tested with ICCS. (B) Multiple cytokine analysis of supernatants of the peptide-stimulated cells. Supernatants of T cell lines demonstrating presence of IFN-γ+ cells were analyzed for the indicated cytokines. Supernatants were collected after overnight stimulation with HCV peptide. IFN-γ, interferon gamma; PBMC, peripheral blood mononuclear cells; HCV, hepatitis C virus.

These results show that HCV-peptide reactive T cells can be detected in the circulation of HCV-negative subjects at frequencies and magnitudes comparable to those in SP of HCV-positive subjects and higher than those observed in chronically infected patients.

HCV-Peptide Reactive T Cells After In Vitro Expansion.

The low quantity of HCV-peptide reactive T cells observed in chronically HCV-infected patients is attributable to their low frequency in the blood and to the impaired IFN-γ production of T cells.8, 23–25In vitro expansion of peptide-stimulated PBMC is a strategy previously used7, 26–28 to detect peptide-specific T cells present at low frequencies in the circulation. Thus, to further evaluate the HCV-peptide reactive T cell repertoire present in HCV-infected (resolved and chronic) and in HCV-negative subjects, T cell responses against HCV peptide pools were analyzed after in vitro expansion. PBMC of HCV-infected patients and HCV-negative subjects (NC and SP) were stimulated with the different peptide mixtures, and after 10 days of in vitro expansion, the presence of HCV-peptide reactive T cells was tested.

In vitro expansion increased the number of peptide pools able to elicit a T cell response in all groups. The subsequent identification of the individual peptides able to elicit a response shows that HCV-negative (NC) subjects present T cells able to recognize a median of 4.5 (range, 3-6) HCV peptides (Fig. 4). Although broad HCV peptide-reactive T cell responses were detected in the blood of all 6 patients with resolved HCV infection, median of 21.5 (range, 15-25). In 6 patients with chronic infection, the HCV-specific repertoire was broader than that of HCV-negative subjects. The median number of individual peptides recognized by chronic HCV patients analyzed was 7 (range, 3-15), and comparison between this group and the negative controls (NC) was found to be statistically significant (P = .0055). This demonstrates a significant enhancement of the HCV-peptide specific T cell repertoire in the circulation of patients with chronic HCV after in vitro expansion.

Figure 4.

Quantification of HCV peptides recognized by T cells after in vitro expansion. Experimental plan: PBMC were stimulated in vitro with the 106 different peptide pools. After 10 days of in vitro expansion, cells were re-stimulated overnight with the initial stimulatory peptide pools in an IFN-γ ELISPOT assay. Lines were considered positive when SFC of stimulated wells were at least double the SFC in unstimulated wells. Positive lines were then tested with individual peptides present in the pools. (A) The number of positive peptides found in the different patient groups using the experimental plan outlined is shown. (B) Ex vivo and in vitro ELISPOT pattern of response against HCV mixtures is shown for one representative negative control and one chronic HCV patient. IFN-γ, interferon gamma; PBMC, peripheral blood mononuclear cells; HCV, hepatitis C virus; SFC, spot-forming cells.

Identification of HCV-Peptide Reactive T Cells Cross-reacting With Homologous Sequences of Common Pathogens.

Having established the presence HCV-peptide reactive T cells in the peripheral blood of HCV-negative subjects, we investigated whether T cells cross-reacting with homologous sequences of common pathogens can sustain these responses.

We first searched the National Centre for Biotechnology Information Gene Bank database using “Blast program” for peptides displaying a degree of homology with the 11 HCV peptides (Table 2) able to activate directly ex vivo PBMC of normal subjects without risk factors.

Table 2. HCV Peptide Responses Detected in the Negative Control Group
PeptideSequenceResponseShared Homology
  • *

    Indicates < 6 shared AA.

288NS3 1436-50 STDALMTGFTGDFDSCD4Measles*
73E1 361-375 YFSMVGNWAKVLVVLCD4Vaccinia virus
563NS5B 2811-25 PLARAAWETARHTPVCD8Human Herpes virus 1
15Core 71-85 PEGRTWAQPGYPWPLCD4Human Papillomavirus* Human Herpes virus 1*
205NS2/NS3 1021-35 KGWRLLAPITAYAQQCD4Human Enterovirus*
216NS3 1076-90 GVCWTVYHGAGTRTICD4Human Papillomavirus*
277NS3 1381-95 PLEVIKGGRHLIFCHCD4Rubella* Human Herpes virus 7*
327NS3 1631-45 VQNEVTLTHPITKYICD4Vaccinia virus
387NS4B 1931-45 PTHYVPESDAAARVTCD8Human Herpes virus 3
588NS5B 2936-50 GRAAICGKYLFNWAVCD8Hepatitis B virus

Variable degrees of sequence homology with different pathogens or self human proteins were found by the in silico analysis for each HCV peptide analyzed. We thus focused our functional analysis on the following 3 sequences: NS5B 2811-25 (563) PLARAAWETARHTPV, NS5B 2936-50 (588) GRAAICGKYLFNWAV, and E1361-375 (73) YFSMVGNWAKVLVVL, which were selected for the following criteria:

  • 1the HCV peptides elicit high direct ex vivo frequency (NS5B 2811-25 91 SFC/subject 3, NS5B 2936-50 70 SFC/subject 6, E1 361-375 48 SFC/subject 4);
  • 2peptide reactivity was confirmed at different time points (not shown),
  • 3high sequence homology with common pathogens or antigens was noted (Table 3).
Table 3. HCV Peptide Sequence Homology With Common Pathogens
PeptidesSequence*SourceGenBank accession no.
  1. NOTE. Boldface indicates shared amino acids.

HCV NS5B 2811-25PLARAAWETARHTPVHCV genotype 1B, NS5BP26624
HCV NS5B 2936-50GRAAICGKYLFNWAVHCV genotype 1B, NS5BP26624
HBV Env 333-40GKYLWEWAHBV-Major surface AgQ05496
HCV E1 361-375YFSMVGNWAKVLVVLHCV genotype 1B, E1P26624
VV 275-284LVGNWDKNDVVaccinia VirusBAD97827

The peptides HCV NS5B 2936-50and HCV E1 361-375showed clear homology respectively with the sequences of hepatitis B virus (HBV) envelope protein (GKYLWEWAS) and with vaccinia virus protein (LVGNWDKNDV)(Table 3). Peptides covering these homologous sequences were synthesized and tested both directly ex vivo in PBMC and then on HCV-peptide reactive T cell lines; however, they consistently failed to activate PBMC and HCV peptide reactive T cells (not shown).

In contrast, evidence of cross-reactivity between the HCV-NS5B2811-25 specific CD8+T cells with a human herpes virus (HHV1) sequence was found. The 15-mer peptides NS5B 2811-25 shared several amino acids with the UL55 protein of HHV1 (Table 3). However, the shared AA between the two sequences encompasses an 11–amino acid length, a length that is not considered compatible with a classical CD8+ epitope. For this reason, we first determined the HCV-epitope contained within the 15-mer peptides. Overlapping 10- and 9-mers covering the 15-mer peptide NS5B 2811-25 region were tested directly ex vivo (Fig. 5) to determine the minimal sequence able to activate NS5B 2811-25 specific CD8+ T cells. Figure 5 shows that the 10-mers NS5B 2816-25 AWETARHTPV peptide elicited a direct ex vivo IFN-γ ELISPOT response similar to the NS5B 2811-25 PLARAAWETARHTPV peptide. In contrast, no response was activated by NS5B 2811-20PLARAAWET and NS5B 2815-23AAWETARHT peptides. Having established that the NS5B 2811-25 peptide–specific T cells direct their recognition on the 10 mer NS5B 2816-25 AWETARHTPV, the corresponding homologous HHV1-peptide QAETAMHTSK was synthesized and tested. Figure 5B shows that the HHV1-peptide QAETAMHTSK can directly activate PBMC of the subject responding to HCV NS5B 2811-25 peptide. In addition, HCV NS5B 2811-25 peptide specific T lines can be stimulated with both NS5B 2816-25 AWETARHTPV and HHV1 QAETAMHTSK peptides.

Figure 5.

Recognition of NS5B 2816-25and HHV1 peptides by HCV-NS5B CD8+ T cells. (A) Ex vivo IFN-γ ELISPOT analysis. PBMC of subject 3 were stimulated overnight with the indicated peptides. The results are expressed in spot-forming cells (SFC)/2 × 105 PBMC. (B) Plots represent IFN-γ–producing CD8+ cells of NS5B 2816-25 specific T cell line stimulated with no peptide, NS5B 2816-25, and HHV1 UL55 29-39 peptides. Numbers in the right upper quadrant indicate the percentages of IFN-γ–producing CD8+ after 4 hours' stimulation. IFN-γ, interferon gamma; PBMC, peripheral blood mononuclear cells; HCV, hepatitis C virus.


This study was designed to analyze whether in vitro evaluation of HCV-specific T cell responses in humans using large arrays of synthetic peptides is potentially altered by the presence of cross-reactive T cells specific for different pathogens but with a degenerate recognition of HCV peptides. The results confirm that this is the case, because mono-oligospecific HCV-peptide reactive T cells can be found in the blood of HCV-negative adults without any classical risk factors of HCV infection (including past blood transfusion, sexual contact with HCV-infected subjects, and intravenous drug use). In all of these HCV-negative subjects, HCV peptide reactive CD4 and/or CD8+ T cells were detected when a large panel of synthetic peptides covering the whole HCV polyprotein (601 peptides) was employed. On the whole, T cell response against HCV peptides was clearly greater (both in magnitude and multispecificity) in patients who recovered from HCV infection than in HCV-negative adults. Nevertheless, for selected individual peptide mixtures, the reactive T cells found in HCV-negative adults present a direct ex vivo frequency comparable to HCV-specific T cells present in subjects who recovered from HCV infection. In addition, these cells had the potential to expand and produce Th1-like cytokines (IFN-γ, tumor necrosis factor alpha) but not IL-4 or IL-5. Thus, at least functionally, these responses were indistinguishable from classical virus-specific effector/memory T cells present in patients with resolved viral infection.

Whether the presence of these HCV-peptide reactive T cells in HCV-negative patients is indicative of previous exposure to HCV (thus, suggesting an under estimation of HCV infection in the general population) or whether it is the consequence of a degenerate recognition of HCV-peptides by cross-reactive T cells is difficult to ascertain.

We favor the latter interpretation, because HCV-specific T cell response present in subjects who recovered from HCV infection are often multispecific with HCV-specific T cells able to recognize multiple HCV epitopes located in the different HCV proteins, whereas those found in HCV-negative subjects show T cells reactive to 1 or 2 peptides out of the 601 used. Furthermore, this mono-oligospecific HCV-reactive T cell response does not broaden into a multispecific response after in vitro expansion, a method that increases substantially the number of HCV peptides recognized by T cells of recovered and chronic patients. More importantly, we were able to demonstrate that NS5B 2816-25 specific CD8+ T cells found in HCV-negative subjects are reactive to a homologous HHV1 sequence. Thus, in addition to the cross-reactivity between HCV and influenza A virus determinant already demonstrated by Wedemeyer et al.,18 we found that a further human pathogen might induce cross-reactive HCV-specific T cells, suggesting that T-cell cross-reactivity between different pathogens is a common phenomenon in adults.

We cannot rule out occult HCV in such subjects; however, in the absence of any indication of liver disease, occult HCV infection29, 30 is unlikely.

Although we failed to demonstrate the direct evidence of cross-reactivity of all the HCV peptide-reactive T cells found in the HCV-negative subjects studied, we should not forget that T cell cross-reactivity does not necessarily depend on identity of amino acids on the linear sequence of the peptide. T cell receptors of a given T cell clone can be activated by quite dissimilar peptides.31, 32 Thus, the strategy of identifying cross-reactive peptides based on sequence homology even though commonly used is bound to underestimate the extent of cross-reactivity. In addition, the search of homologous peptides is based on the search of known sequences, and thus unable to detect yet-unidentified homologous sequences.

What are the putative consequences and potential implications of the detection of HCV peptide-reactive T cells in HCV-negative adults? The pathological impact of cross-reactivity on virus infection has been analyzed in animal models, where it was shown that a previous encounter with different pathogens can influence virus-related immunopathology. We have recently shown that severe cases of HCV-induced hepatitis are characterized by a large expansion of HCV-specific CD8-T cells able to cross-react with an influenza neuraminidase sequence,9 suggesting that virus-related immuno-pathology can be influenced by the presence of potentially cross-reactive T cells in humans as well. The demonstration that all HCV-negative adults present at least one HCV-peptide reactive T cell response suggests that cross-reactivity can potentially influence the degree of immunopathology in HCV infection. Moreover, rapid expansion on HCV exposure of memory responses originally primed by unrelated pathogens may favor protection against HCV if cross-reactivity leads to expansion of T cell specificities that are normally involved in HCV control.

The identical quantitative detection of HCV-peptide reactive T cells in HCV-negative patients with or without risk factors suggests that HCV peptide reactive T cells from sexual partners of HCV chronic patients are not necessarily primed by HCV.

Because we cannot distinguish between HCV-primed or HCV cross-reactive T cells, the potential protective effect of isolated HCV-peptide reactive T cells found in sexual partners of patients with chronic HCV infection is questionable. Cross-reactive T cells detected in vitro may be primed in vivo by unrelated pathogens and simply recalled in vitro by HCV epitopes irrelevant to virus control (because they are not generated by HCV antigen processing during natural infection). However, our data do not contest the idea that low-level HCV inoculation can prime an HCV-specific cellular immune response. When HCV contact is more likely to occur (i.e., sexual partners of patients with acute HCV infection and high viremia),33 these subjects develop a multispecific CD4 and CD8+ response, a profile different from the oligo-reactivity found in our HCV-negative subjects.

A further interesting observation revealed by our comprehensive analysis of HCV-peptide reactive T cell response in different groups of HCV-infected and non-infected individuals is that the overall quantity and the individual ex vivo frequency of HCV-peptide reactive T cells is similar and frequently even lower in chronically HCV-infected patients than in HCV-negative adults. These data further support the evidence that persistent HCV infection qualitatively alters the circulating and intrahepatic pool of HCV-specific T cells.8, 34, 35 To what extent HCV–peptide reactive T cells not found in chronically infected patients are deleted, compartmentalized within the liver, or suppressed in their functional capacity needs further investigation.

In conclusion, this study illustrates that HCV peptide reactive T cells can be demonstrated in adults irrespective of previous HCV exposure. Therefore, the presence of such isolated HCV-peptide reactive T cells does not necessarily equate to previous HCV exposure because cross-reactivity between HCV peptides and sequences of common pathogens, such as human herpes virus 1, was revealed. However, cross-reactivity instead of previous HCV priming does not exclude the potential for protection of cross-reactive T cells on HCV exposure, provided the homologous HCV epitopes are involved in the induction of protective responses in natural infection. The demonstration that the comprehensive magnitude of HCV-peptide reactive T cells present in chronically HCV infected patients might even be lower than that of HCV-peptide reactive T cell response found in HCV-negative adults adds further support to the argument for a profound collapse of virus-specific T cell response caused by HCV persistence.


The authors thank Dimitrios Bogdanos for his kind help and advice with the Blast program.