CD4+ Treg-cells (regulatory T-cells) probably contribute to the impaired virus-specific T-cell responses in chronic HCV (hepatitis C virus) infection; however, their antigen-specificity has remained elusive. In the present study, we analysed peripheral blood CD4+ Treg-cells in patients with chronic hepatitis C and subjects with self-limited HCV infection and characterized individual Treg-cell clones obtained from both groups at the phenotypic and functional level. Foxp3 (forkhead box p3)+CD25+CD4+ Treg-cells were detected more frequently in patients with chronic hepatitis C than self-limited HCV infection, which responded to HCV core stimulation and inhibited proliferation of reporter cells. Cloning under limiting dilution conditions resulted in 14 and six hypoproliferative Foxp3+CD25+CD127−CD4+ T-cell clones from patients with chronic hepatitis C and subjects with self-limited HCV infection respectively. All clones expressed Treg-cell markers and produced IL (interleukin)-10 upon mitogen stimulation. However, exclusively Treg-cell clones from chronic hepatitis C produced IL-10 in response to HCV core and inhibited proliferation of reporter T-cells. These core-specific Treg-cell clones recognized epitopes in two regions of HCV core (amino acids 1–44 and 79–113). Co-culture inhibition assays demonstrated Treg-cells to inhibit reporter T-cells via secretion of IL-10 and IL-35 rather than cell-contact-dependent mechanisms. Finally, the HCV-specific Treg-cell clones lost their functional capacity, along with Foxp3 expression, if kept in culture without HCV core exposure. In conclusion, we identified functionally active HCV core-specific Treg-cells in patients with chronic hepatitis C, which share their epitopes with conventional T-cells and require the continued presence of antigen to maintain their functional differentiation. Thus HCV core-specific Treg-cells may contribute to the immunoregulatory balance in chronic hepatitis C.
- adaptive regulatory T-cell (Treg-cell)
- cytokine release
- hepatitis C virus (HCV)
- T-cell cloning
HCV (hepatitis C virus) efficiently establishes chronic persistency in the majority of individuals who encounter the infection and, thus, it is one of the leading causes of chronic liver disease in the world . Although strong and multi-specific lymphocyte responses, including CD4+ and CD8+ T-cells, NK (natural killer) cells and dendritic cells, contribute to viral elimination, chronic hepatitis C is associated with inefficient and delayed T-cell responses . Although the mechanisms of impaired antiviral immunity in patients with chronic hepatitis C are only incompletely understood, it has been proposed that CD4+ Treg-cells (regulatory T-cells) might contribute to T-cell dysfunctions during ongoing HCV infection.
Several studies have demonstrated an overall increased frequency of CD25highCD4+ Treg-cells in chronic hepatitis C which produce IL (interleukin)-10  and TGF (transforming growth factor)-β , express Foxp3 (forkhead transcription factor box 3) , can suppress IFN-γ (interferon-γ) production  and proliferation of virus-specific CD8+ T-cells [7,8]. Thus far, analysed Treg-cells carry phenotypic and functional characteristics of so-called natural CD4+ Treg-cells [9,10], because they are Foxp3+ , express high levels of CD25 [3–5] and predominantly inhibit reporter cells via contact-dependent mechanisms [3,4,6,8]. Since, thus far, expansion of Treg-cells in chronic hepatitis C is largely considered to reflect HCV non-specific Treg-cells, the role of HCV-specific adaptive Treg-cells has still remained a matter of debate.
Antigen-induced adaptive Treg-cells are generated in the periphery upon antigen exposure inhibiting effector T-cells via release of the immunosuppressive cytokines IL-10 and TGF-β [11–13]. In general, adaptive Treg-cells exhibit heterogeneous phenotypes with variable expression of CD25 and Foxp3 and can be subdivided further according to their individual phenotypes and cytokine profiles [13,14].
The HCV core protein is highly conserved among different HCV isolates  and is frequently recognized by the host's cellular antiviral immune response . Thus HCV core is an exquisite model antigen to search for HCV-induced adaptive Treg-cells in chronic hepatitis C. Studying subjects with different outcomes of HCV infection offers an opportunity to understand the relationship of Treg-cells with continued antigen exposure. With this approach, we studied HCV core-specific Treg-cells in the peripheral blood of patients with chronic hepatitis C and compared them with corresponding cells in anti-HCV-positive subjects who had resolved their infection spontaneously.
MATERIALS AND METHODS
PBMCs (peripheral blood mononuclear cells) were isolated by Ficoll–Paque gradient centrifugation (PAA Laboratories) from heparinized blood of 31 treatment-naïve non-cirrhotic patients with chronic hepatitis C [15 males and 16 females; age, 46 (30–68) years; median (range)], and 12 subjects with self-limited HCV infection [all males; age, 43 (27–65) years]. Patients with chronic hepatitis C had detectable HCV RNA genotype 1 by PCR for more than 6 months. Stages of fibrosis were obtained either by histology or Fibroscan and are reported as Metavir scores. Subjects with self-limited HCV infection consisted of carefully selected individuals who had spontaneously recovered from a well-documented episode of acute hepatitis infection 27 months to 30 years prior to the study, which in retrospect had unequivocally been identified as hepatitis C. All control subjects were anti-HCV positive (HCV version 3.0, AXSYM System; Abbott), but had consistently normal aminotransferases and a negative HCV PCR on repeated assessment. The clinical data of both study groups are summarized in Table 1.
The study has been carried out in accordance with the Declaration of Helsinki (2000) of the World Medical Association and has been approved by the local ethics committee. Informed consent had been obtained prior to inclusion into the study.
Recombinant HCV core protein [genotype 1, amino acids 1–115; <4.0 pg/μg of LPS (lipopolysaccharide)] was purchased from Mikrogen. For epitope mapping a 25-mer HCV peptide panel (P1–P12) covering region amino acids 1–133 of the HCV core protein with a ten amino acid overlap was synthesized as described previously [17,18].
Human rIL-2 (recombinant IL-2) was obtained from R&D Systems, anti-CD28 from BD Biosciences, OKT-3 (human anti-CD3) from Janssen-Cilag, and CMV (cytomegalovirus) glycoprotein p65 from AUSTRAL Biologicals.
RPMI 1640 medium supplemented with 10% (v/v) human AB serum, 200 mg/ml glutamine, 50 μg/ml gentamycin, 100 units/ml penicillin and 100 μg/ml streptomycin (all from PAA Laboratories) was used for cell culture.
Analysis of CD4+ Treg-cell subsets prior to in vitro expansion
To identify the spectrum of CD4+ Treg-cells prior to in vitro expansion, subsets of CD4+ T-cells carrying Treg-cell markers were quantitatively determined by flow cytometry in freshly isolated PBMCs from patients with chronic hepatitis C and subjects with self-limited HCV infection. Next, PBMCs were studied after a single round of in vitro stimulation with HCV core to search for HCV-responsive cells. In brief, PBMCs (5× 105 cells/well) were stimulated with HCV core protein (10 μg/ml) in culture medium for 10 days. PBMCs were stained with anti-CD25 [PE (phycoerythrin)-labelled], anti-CD4 [allophycocyanin-labelled], and anti-Foxp3, anti-GITR (glucocorticoid-inducible tumour necrosis factor receptor), anti-CTLA-4 (cytotoxic T-lymphocyte antigen-4), anti-PD1 (programmed death 1) and anti-CD134 (all FITC-labelled).
In co-culture experiments, we analysed the inhibitory capacity of freshly isolated CD25highCD4+ and CD25intCD4+ T-cells (where int is intermediate) on CD25−CD4+ reporter T-cells. Briefly, CD25highCD4+ and CD25intCD4+ T-cells were purified from unexpanded PBMCs by stepwise immunomagnetic separation using the CD4+CD25+ Treg-cell isolation kit and high-affinity CD25 Microbeads (Miltenyi Biotech). Resulting Treg-cells were routinely >95% pure and comprised <5% non-vital cells (as determined by 7-aminoactinomycin D staining). The remaining CD25−CD4+ T-cells were used as reporter cells in co-culture stimulation experiments. CD25highCD4+ or CD25intCD4+ Treg-cells were added to CD25−CD4+ reporter T-cells at a 1:1 ratio and stimulated with HCV core or anti-CD3/anti-CD28 presented via 10% autologous monocytes. Proliferation was assessed via [3H]thymidine uptake after 5 days.
Generation and characterization of HCV-specific Treg-cell clones
Generation of Treg-cell clones was achieved by adapting previously established protocols [19,20]. In detail, freshly isolated PBMCs (5×105 cells/ml) were stimulated with core protein (10 μg/ml) and cultured in medium with rIL-2 (100 units/ml). After 10 days, cells were re-stimulated with HCV core protein and supplemented with irradiated autologous feeder cells (5×105 PBMCs/ml). After 20 days, cells were re-stimulated for 72 h with PHA (phytohaemagglutinin; 1 μg/ml) together with irradiated feeder cells. Next, cells were seeded in Terasaki plates (Greiner Bio-one) under conditions of limiting dilution (0.1 and 0.3 cells/well). After 10 days, clones were picked from plates with <10% positive wells and re-stimulated at 10–14 day intervals in the presence of rIL-2.
Clones were classified as Treg-cells by analysing proliferation and cytokine secretion (IL-4, IL-10, IFN-γ, TGF-β1) in response to either HCV core protein or anti-CD3/anti-CD28. Recombinant CMV p65 (10 μg/ml) served as a control antigen to check HCV specificity. Functional and phenotypic studies were always performed in the resting phase 10 days after re-stimulation. In the presence of antigen, T-cell clones maintained their original functionality for culture periods of up to 100 days.
Flow cytometric analysis of Treg-cell clones
Treg-cell clones (5×105 cells) were stained with allophycocyanin-labelled anti-CD4, and PE-conjugated anti-CD25 (clone M-A251), anti-CD45R0 (clone UCHL1), anti-CD62L (clone Dreg-56) and anti-CD127 (clone SB/199) respectively. FITC-labelled antibodies were used for detection of Treg-cell markers GITR (clone 110416), CTLA-4 (clone 48815), PD-1 (clone J116), CD134 (clone ACT35) and CD30 (clone Ki-1) (all from BD Biosciences). Moreover, we analysed the expression of GARP [glycoprotein A repetitions predominant; also known as LRRC21 (leucine-rich repeat-containing protein 21); clone Plato-1; ALEXIS] and nuclear expression of Foxp3 using the Foxp3 staining kit [clone PCH101, IgG2a isotype control (clone eBR2a); NatuTec] according to the manufacturer's instructions. T-cells were examined on a FACSCalibur flow cytometer and were analysed by Flow Jo software (Tree Star). CD25+ T-cells were segregated into CD25int cells [MFI (mean fluorescence intensity), 10–300] and CD25high cells (MFI, >300) with reference to KARPAS-299 cells, which carry the CD25highCD4+ phenotype of natural Treg-cells (see Supplementary Figure S1 at http://www.ClinSci.org/cs/119/cs1190097add.htm).
Proliferative responses of HCV core-specific Treg-cell clones
Cells (5×104) from each prospective HCV-specific Treg-cell clone were stimulated with either HCV core protein (10 μg/ml) or anti-CD3/anti-CD28 (1 and 2.5 μg/ml), together with 1×105 irradiated autologous APCs (antigen-presenting cells). After 4 days, cultures were pulsed with 1 μCi of [3H]thymidine for 18 h. Incorporated radioactivity was measured by liquid-scintillation spectrometry (TopCount®NXT; PerkinElmer Life Sciences). SIs (stimulation indices) were calculated as the ratio of [3H]thymidine uptake relative to the medium controls. SIs >5 were considered positive.
Cytokine profiles of HCV core-specific Treg-cell clones
Cells (1×105) of each candidate Treg-cell clone were stimulated separately either with HCV core protein or anti-CD3/anti-CD28. At 72 h after stimulation, supernatants were harvested to measure IL-4, IFN-γ, IL-10 and TGF-β1 by ELISA. TGF-β1 was determined using the OptEIA human TGF-β1-Set from BD Biosciences. IL-4, IL-10 and IFN-γ were captured with primary antibodies 8D4-8, JES3-9D7 and NIB42 and bound cytokines were detected with the corresponding antibodies MP4-25D2, JES3-268 and S.B3 (all BD Biosciences) respectively. Plates were measured in an ELISA reader (Orgentec) at 450 nm. Cytokine values >50 pg/ml were considered positive and defined as high if >250 pg/ml. Clones were classified as Th (T-helper) 1, Th2 and Treg-cells based on cytokine and proliferation profiles in response to anti-CD3/anti-CD28 stimulation (see Supplementary Table S1 at http://www.ClinSci.org/cs/119/cs1190097add.htm).
HLA-restriction of HCV core-specific Treg-cell clones
HLA-restriction was determined by blocking HCV core-specific IL-10 production with antibodies L243 (HLA-DR), Tü22 (HLA-DQ; both from A.T.C.C.) or HI43 (HLA-DP; Biozol) respectively. Briefly, 1×105 cells of the HCV-specific Treg-cell clone were stimulated with HCV core protein (10 μg/ml) in the presence or absence of HLA-DR, -DQ or -DP antibodies. After 72 h, supernatants were harvested to measure IL-10 by ELISA.
Identification of Treg-cell epitopes on HCV core
Individual Treg-cell clones (1×105) were stimulated with single peptides P1–P12 (1 μg/ml) in the presence of autologous APCs (1×105). After 72 h, supernatants were harvested to measure IL-10 by ELISA.
Analysis of in vitro suppressor function of HCV core-specific Treg-cell clones
The suppressive function of individual Treg-cell clones was studied using co-culture assays with stimulated autologous Th1 and Th2 reporter clones. Reporter cells (5×104 cells/well) were stimulated either with anti-CD3/anti-CD28 or recombinant HCV core protein in the presence of irradiated autologous APCs (5×104 cells/well). Putative Treg-cell clones were added at 1:4, 1:2 and 1:1 ratios. After 5 days, proliferation was measured in both Th1 and Th2 reporter clones via [3H]thymidine incorporation.
To decipher the predominant mechanism of inhibition, we compared the inhibitory capacity of anti-CD3/anti-CD28-stimulated Treg-cell clones in direct co-culture with Th1 and Th2 reporter clones (1:1 ratio) compared with transwell culture, which disrupts cell–cell contacts between Treg-cells and reporter cells (tissue culture inserts, 0.2 μm Anopore® Membrane; Nunc). In addition, we performed blocking experiments using either a neutralizing antibody against IL-10 (clone 23738, 10 μg/ml; R&D Systems) or against the shared β-chain of IL-35 (EBI3-ATI136, 10 μg/ml; Axxora).
All calculations were performed using Prism software package 4.0 (Graph Pad Software). Differences between experiments were compared by Fisher's exact test, Kruskal–Wallis test, Wilcoxon signed-rank test and Student's t test as appropriate. P<0.05 was considered as statistically significant.
HCV reactivity of freshly isolated Treg-cell subsets in peripheral blood
In line with previous reports, we observed significantly higher baseline numbers of Foxp3+CD25+CD4+ T-cells in freshly isolated PBMCs from patients with chronic hepatitis C than from subjects with self-limited HCV infection. Using KARPAS-299 cells which carry the Foxp3+CD25highCD4+ phenotype of natural Treg-cells as a reference , we found that both Foxp3+CD25highCD4+ and Foxp3+CD25intCD4+ T-cell subsets were increased in chronic hepatitis C (Table 2).
To study the functional contribution of both CD25+CD4+ T-cell subsets, we purified CD25high CD4+ and CD25intCD4+ T-cells from PBMCs of patients with chronic hepatitis C (Figure 1A) and analysed their suppressive activity. As shown in Figures 1(B) and 1(C), addition of CD25highCD4+ and CD25intCD4+ T-cell subsets significantly decreased proliferation of CD25−CD4+ reporter T-cells both after stimulation with mitogen and HCV core. We also observed that Foxp3+CD25highCD4+ and Foxp3+CD25intCD4+ T-cells from patients with chronic hepatitis C increased after in vitro stimulation with HCV core protein (Table 2). Of note, expansion of Foxp3+CD25+CD4+ T-cells after core stimulation corresponded to increased numbers of CTLA-4+CD25high and CTLA-4+CD25intCD4+ T-cells (Table 2), whereas expression of the other Treg-cell markers GITR, PD-1 and CD134 remained unaffected (results not shown). In contrast with chronic hepatitis C, CD4+ Treg-cell subsets from subjects with self-limited HCV infection remained unaltered after in vitro stimulation with HCV core. Thus patients with chronic hepatitis C, but not subjects with previous self-limited HCV infection, harbour Foxp3+CD25+CD4+ Treg-cells, which specifically respond to stimulation with HCV core.
When numbers of Foxp3+CD25high, Foxp3+CD25int, CTLA-4+CD25high and CTLA-4+CD25int T-cells in patients with chronic hepatitis C were correlated with age, viral loads, liver enzymes and Metavir scores, significant trends towards increased numbers of Foxp3+CD25int Treg-cells in advanced stages of fibrosis were detected both at baseline (r=0.46, P=0.02) and after HCV core stimulation (r=0.41, P=0.04). In contrast, none of the other T-cell subpopulations revealed any correlation to clinical parameters (results not shown).
To characterize HCV-specific Treg-cells in detail, we studied T-cell clones generated in the presence of HCV core antigen and abundant IL-2.
Identification of antigen-specific Treg-cell clones in chronic hepatitis C
Overall, we succeeded in generating 242 CD4+ T-cell clones from seven patients with chronic hepatitis C and 65 clones from six subjects with self-limited HCV infection. On the basis of their proliferation characteristics and cytokine profiles in response to anti-CD3/anti-CD28 stimulation, we identified 23 and 14 Th1 clones, 34 and 21 Th2 clones as well as 14 and six Treg-cell clones from patients with chronic hepatitis C and subjects with self-limited HCV infection respectively (see Supplementary Tables S1 and S2 at http://www.ClinSci.org/cs/119/cs1190097add.htm).
All putative Treg-cell clones both from chronic hepatitis C (#1–#14) and self-limited HCV infection (#15–#20) were Foxp3+ and CD25int (MFI range, 49–297) (Figure 2A). In addition, they expressed variable levels of the other Treg-cell markers CTLA-4, GITR, PD-1, CD30, CD134 and GARP, but were consistently CD127− (Supplementary Figure S2 at http://www.ClinSci.org/cs/119/cs1190097add.htm). Thus all clones matched the expected phenotype of Treg-cells. However, Treg-cell clones from patients with chronic hepatitis C expressed significantly higher levels of CD25 and CTLA-4 than Treg-cell clones from subjects with self-limited HCV infection (Figure 2B). Concerning their cytokine profiles in response to anti-CD3/anti-CD28, all Treg-cell clones produced IL-10 (Supplementary Table S2); and two clones from chronic hepatitis C also produced TGF-β1 (clone #5, 187.0 pg/ml; and clone #6, 963.1 pg/ml).
Of note, there was a fundamental difference in antigen-specificity between Treg-cell clones from chronic hepatitis C compared with self-limited HCV infection: 12 out of the 14 Treg-cell clones from chronic hepatitis C produced IL-10 after stimulation with HCV core antigen, whereas none of the Treg-cell clones from self-limited HCV infection recognized the core antigen (Figure 3A). HCV core-specificity of IL-10 responses in Treg-cell clones from chronic hepatitis C was confirmed by the absence of reactivity to a recombinant CMV antigen (Figure 3A). HCV core-specific IL-10 responses were restricted by HLA-DR (Supplementary Figure S3 at http://www.ClinSci.org/cs/119/cs1190097add.htm). Using overlapping peptides, we identified single epitopes which were clustered in the regions of amino acids 1–44 and 79–113 on the HCV core protein (Figure 3B).
Inhibitory capacity of Treg-cell clones in chronic hepatitis C
When Treg-cell clones were analysed in suppressor assays against autologous Th1 and Th2 reporter clones, all Treg-cell clones from chronic hepatitis C, but none of the clones from self-limited HCV infection, inhibited proliferation of reporter cells in a dose-dependent manner upon mitogenic stimulation (Figures 4A and 4B, and Table 3). In two Treg-cell clones from chronic hepatitis C (#4 and #5), sufficient numbers of cells were available to also study the inhibitory activity against autologous reporter cells after activation with HCV core. Of note, both Treg-cell clones inhibited proliferation of reporter cells also upon antigen-specific activation (Figure 4C).
Since adaptive Treg-cells are considered to require ongoing stimulation to maintain their regulatory activity, we next studied the in vitro stability of Treg-cell clones from chronic hepatitis C if they were further expanded with Treg-cell expander beads (Dynal) instead of HCV antigen. As shown in Figure 5, Treg-cell clones gradually lost their suppressor function after multiple rounds of non-specific re-stimulation. This loss of function was closely correlated with reduced Foxp3 expression.
Mechanism of suppression mediated by HCV-specific Treg-cell clones
Treg-cell cells can act by contact-dependent and contact-independent mechanisms. Therefore we performed suppressor assays in a transwell system and in the presence of neutralizing antibodies. Figure 6 shows that suppression of reporter cells remained unaffected if Treg-cells were directly co-cultured with reporter cells or kept in transwells. Thus disrupting contact between Treg-cells and reporter T-cells did not abrogate their inhibitory capacity, suggesting that inhibition was mediated via soluble factors. In line with this concept, suppression of reporter cells by the Treg-cell clones was markedly reduced when co-culture experiments were performed in the presence of a neutralizing IL-10 antibody (Figure 6). We suspected additional soluble factors other than IL-10, possibly explaining the functional difference between Treg-cell clones from chronic hepatitis C and self-limited HCV infection. As IL-35 has been described as a novel inhibitory cytokine of Treg-cells [22,23], we also co-cultured Treg-cells and reporter cells in the presence of a neutralizing IL-35 antibody and found that addition of anti-IL-35 also significantly blocked Treg-cell-mediated inhibitory effects (Figure 6).
Accumulating evidence indicates that CD4+ Treg-cells are expanded in humans during acute HCV infection, maintained at increased levels during the chronic stage and lowered to normal levels upon spontaneous or treatment-induced recovery , suggesting that a certain balance exists between Treg-cells and effector T-cells. Our present results support this concept and provide several levels of evidence that HCV core-specific adaptive Treg-cells are present in chronic hepatitis C. First, we demonstrated that freshly isolated CD25high and CD25intCD4+ T-cells in chronic hepatitis C express Foxp3, as well as other markers of Treg-cells, and can inhibit CD25−CD4+ reporter T-cells not only after stimulation with mitogen, but notably also after activation with HCV core. In line with previous reports [3–5,16,25,26] our present findings also confirmed increased numbers of Foxp3+CD25+CD4+ T-cells in patients with chronic hepatitis C. In addition, in vitro stimulation with HCV core further increased the Foxp3+CD25highCD4+ and Foxp3+CD25intCD4+ Treg-cell subsets in patients with chronic hepatitis C, but not in subjects with previous self-limited HCV infection, despite the fact that HCV-specific CD4+ T-cells usually persist long-term after elimination of the virus . Finally, numbers of Foxp3+CD25intCD4+ Treg-cells appeared to increase in patients with more advanced fibrosis. A rather similar relationship between Foxp3+CD25+CD4+ Treg-cells and hepatic fibrosis in chronic hepatitis C has also been reported by Marché and co-workers in the liver ; however, their study did not address antigen specificity.
To further study HCV core-specific Treg-cells, we established T-cell clones in the presence of HCV core including abundant IL-2 and succeeded in generating 14 and six Treg-cell clones from patients with chronic hepatitis C and subjects with self-limited HCV infection respectively. At a first glance, Treg-cell clones from either group exhibited rather similar features: all clones were hypoproliferative, expressed Foxp3, as well as other Treg-cell markers, and produced substantial amounts of IL-10. However, Treg-cells clones were CD25int and expressed rather moderate levels of Foxp3, which distinguished them from natural Treg-cells [9,10].
Nevertheless, we found marked differences between Treg-cell clones from chronic hepatitis C and self-limited HCV infection: Treg-cell clones from self-limited HCV had lower CD25 and CTLA-4 expression levels, were not active in autologous suppressor assays and did not respond to HCV core. In contrast, Treg-cell clones from chronic hepatitis C inhibited proliferation of autologous Th1 and Th2 reporter clones, and were specifically activated by recombinant HCV core in an HLA-DR-restricted fashion. Although our present experiments were somewhat limited by the low number of cells available, we also found that activation with HCV core led to inhibition of reporter T-cells by the two Treg-cell clones, which we could study in antigen-specific suppressor assays. Taken together, these findings indicated that we had cloned HCV core-specific adaptive Treg-cells from chronic hepatitis C, whereas T-cells with a rather similar phenotype from self-limited HCV infection exhibited neither HCV-specificity nor functional activity.
Epitope mapping with 25-mer peptides confirmed that Treg-cells from chronic hepatitis C were activated by single epitopes which clustered to two regions on HCV core (amino acids 1–44 and 79–113). It is noteworthy that the same regions have been shown previously to harbour major epitopes of HCV core-specific effector T-cells [17,18]. Thus shared HCV core epitopes recognized by both conventional effector T-cells and Treg-cells may be a hint that adaptive Treg-cells in chronic hepatitis C are activated in parallel to maintain a certain balance in the regulation of the immune response .
HCV core protein is abundant during chronic infection but absent after viral elimination. Importantly, HCV core-specific adaptive Treg-cells appear to be closely linked to continued antigen exposure. This idea is supported by our present in vitro studies, which indicated that Treg-cell clones gradually lost their Foxp3 expression and inhibitory activity if cultured without HCV core protein. Thus, in analogy to our experiments in Treg-cell clones, freshly isolated Foxp3+CD25+CD4+ T-cells from subjects with previous self-limited HCV infection might have not responded to HCV core stimulation, because HCV core-specific Treg-cells had disappeared after antigen elimination. This mechanism can also explain why we were unable to establish HCV core-specific Treg-cell clones from subjects with self-limited HCV infection.
Our HCV core-specific Treg-cell clones differ from Treg-cells in HCV infection described by other groups with respect to several features [3–8]. For instance, they differ from the HCV-specific Treg-cells described by Ebinuma et al. , which had been studied after stimulation with non-structural proteins NS3 and NS4, because their Treg-cells were CD25high and did not produce IL-10 or TGF-β1. Thus it remains an interesting, but unproven idea, that several types of distinct adaptive CD4+ Treg-cell subsets exist, the phenotypic and functional differentiation of which may reflect activation by distinct HCV antigens (core in our present study compared with NS3/NS4 in the study by Ebunima et al. ).
A further difference concerns the mechanisms underlying the inhibitory activity in our Treg-cell clones, because most Treg-cells in hepatitis C appear to involve contact-dependent mechanisms of inhibition [3,4,6,8]. Despite some heterogeneity in the individual expression of Treg-cell markers, which is a characteristic feature of adaptive Treg-cells [11–13], all Treg-cell clones from chronic hepatitis C expressed CTLA-4 or PD-1, which are pivotally involved in contact-mediated T-cell inhibition. Moreover, CTLA-4 expression was substantially higher on Treg-cell clones from chronic hepatitis C than from self-limited HCV infection. Nevertheless, our transwell co-culture experiments clearly indicated that Treg-cell clones from chronic hepatitis C suppressed reporter T-cells via the inhibitory cytokines IL-10 and IL-35. These findings are in line with recent observations in mice suggesting that activation of Treg-cells might require some cell contact, whereas in activated cells immunosuppressive functions were mediated via release of IL-10 and IL-35 . Foxp3 expression appears to be an important factor in this process, since Foxp3 can enhance IL-10 secretion, because in vitro experiments transfecting Foxp3 into CD25−CD4+ T-cells resulted in increased IL-10 mRNA levels [32,33]. This relationship between Foxp3 and cytokine enhancement possibly explains our observation that Treg-cell clones lost their immunosuppressive capacity when Foxp3 expression disappeared.
Several studies have already provided some indications that Treg-cells may be antigen-specific during HCV infection [4,5,34]. Li et al.  reported that, unlike controls, in vitro culture of PBMCs from HCV-infected patients with HCV-derived peptides resulted in rapid induction of Foxp3+CD25+CD4+ Treg-cells. Furthermore, in vivo expansion of Foxp3+CD25+CD4+ Treg-cells has been observed during acute hepatitis C [34–36]. Nevertheless, the functional role of Treg-cells has remained unclear. Although Perella et al.  identified Treg-cell expansion as a risk factor for chronic infection, other groups reported that increases in Treg-cell numbers were independent of the clinical outcome of disease [34,36]. Finally, the finding of a relationship between numbers of Treg-cells and fibrosis in the study by Marché et al.  and our present study suggests that Treg-cells may be linked in some way to tissue damage in hepatitis C. To this discussion our present results also add the novel finding that functionally active Treg-cells with specificity for HCV core exist in chronic hepatitis C. These cells require continued antigen exposure to maintain their Foxp3 expression and inhibitory capacity. Since we could not detect such adaptive Treg-cells in subjects after spontaneous resolution of their HCV infection, this type of Treg-cell is either not generated during acute HCV infection or is lost after spontaneous resolution in the subjects with self-limited HCV infection. Thus evolution of Treg-cells should be studied further during the various phases of infection as well as in the liver in order to better understand their contribution to the altered balance of immunoregulation and tissue damage in hepatitis C.
This work was supported by the Deutsche Forschungsgemeinschaft (DFG) [grant numbers SP 483/4-1, SFB TRR 57 ‘Organ fibrosis’].
Abbreviations: APC, antigen-presenting cell; CMV, cytomegalovirus; CTLA-4, cytotoxic T-lymphocyte antigen-4; Foxp3, forkhead transcription factor box 3; GARP, glycoprotein A repetitions predominant; GITR, glucocorticoid-inducible tumour necrosis factor receptor; HCV, hepatitis C virus; IFN-γ, interferon-γ; IL, interleukin; int, intermediate; MFI, mean fluorescence intensity; PBMC, peripheral blood mononuclear cell; PD1, programmed death 1; PE, phycoerythrin; PHA, phytohaemagglutinin; rIL-2, recombinant IL-2; SI, stimulation index; TGF, transforming growth factor; Th, T-helper; Treg-cell, regulatory T-cell
- © The Authors Journal compilation © 2010 Biochemical Society