Characterization of antigenic variants of hepatitis C virus in immune evasion

Abstract

Background: Antigenic variation is an effective way by which viruses evade host immune defense leading to viral persistence. Little is known about the inhibitory mechanisms of viral variants on CD4 T cell functions.

Results: Using sythetic peptides of a HLA-DRB1*15-restricted CD4 epitope derived from the non-structural (NS) 3 protein of hepatitis C virus (HCV) and its antigenic variants and the peripheral blood mononuclear cells (PBMC) from six HLA-DRB1*15-positive patients chronically infected with HCV and 3 healthy subjects, the in vitro immune responses and the phenotypes of CD4+CD25+ cells of chronic HCV infection were investigated. The variants resulting from single or double amino acid substitutions at the center of the core region of the Th1 peptide not only induce failed T cell activation but also simultaneously up-regulate inhibitory IL-10, CD25-TGF-β+ Th3 and CD4+IL-10+ Tr1 cells. In contrast, other variants promote differentiation of CD25+TGF-β+ Th3 suppressors that attenuate T cell proliferation.

Conclusions: Naturally occuring HCV antigenic mutants of a CD4 epitope can shift a protective peripheral Th1 immune response into an inhibitory Th3 and/or Tr1 response. The modulation of antigenic variants on CD4 response is efficient and extensive, and is likely critical in viral persistence in HCV infection.

Figure 1

Proliferation in response to peptides of wild-type NS3_358-375 and its variants and binding affinity of those peptides to HLA-DRB1*1501 molecules. A and B: PBMC from a HCV-infected patient B3019 were cultured with wild type NS3_358-375 and variant peptides at the given concentrations, to six days, pulsed overnight with 1.0 μCi/well tritiated thymidine. Radioactive label incorporation (cpm) was measured. PBMC with tissue culture medium alone was used as a negative control. Experiments were performed in triplicate. A representative of five experiments is shown. C and D: Binding of biotinylated peptides to HLA-DRB1*1501 molecules were assessed using an antigen-capture ELISA. Overnight binding affinities of variant peptides were compared to wild type peptide NS3_358-375. A representative of three experiments is shown. E and F: PBMCs were pre-pulsed with 0.1 μM variant peptides for three hours, then re-challenged with wild type NS3_358-375 at a concentration of 10 μM. Rest of the experiment was as same as mentioned in A. A representative of five experiments is shown.

Figure 2

Secreted IL-10, CD25⁺ and TGF-β-bound cells were involved in the inhibition by HCV variants. A-B: PBMCs were cultured with peptide NS3_358-375 (5 μM), variant-peptide pool VP1, or VP2 (1 μM/each peptide) for three hours, then 15 μg/ml of each antibody to human IL-10, CD25 and TGF-β were added. Parallel cultures with either NS3_358-375 or VP1 and VP2 without antibody, and PBMC with medium alone were used as controls. At day four, cells were pulsed overnight with 1.0 μCi/well tritiated thymidine. Radioactive label incorporation (cpm) was measured at day five. C-D. CD25⁺ or TGF-β-bound cells were deleted from PBMCs using antibodies to CD25⁺ or TGF-β and magnetic beads. The remaining cells were cultured with either NS3_358-375 or VP1 and VP2. Proliferation measurement and controls were the same as described in A-B. E: 50 μl of a supernatant pool alone or the pool with 15 μg/ml TGF-β antibody was added into each culture of PBMC with VP1 or VP2. The supernatant pool consisted of each culture of VP1 and VP2 described in A. F. PBMCs without TGF-β-bound cells were cultured with VP1 or VP2 and 50 μl of the supernatant pool. All experiments were repeated twice, a single representative experiment is shown G. Levels of IFN-γ, IL-10 and TGF-β induced by peptide NS3_358-375, VP1 and VP2 were determined by cytokine-specific ELISA. Each 100 μl of supernatant from the cultures described in A was used for ELISA analysis of IFN-γ, IL-10 and TGF-β. Experiments were performed in duplicate. A single representative of five experiments is shown.

Figure 3

Treg phenotypes induced by peptide NS3_358-375 and variant pools VP1 and VP2. Cell culture adn purification of CD4⁺ cells and controls were the same as described in Figure 2. Intracellular staining and four-color flow cytometry using fluorescent-labeled antibodies to human CD25, IFN-γ, TGF-β and IL-10 were performed with purified CD4⁺ cells. A pool from each test was used as an isotype control. A: Percentages of CD25⁺ gated CD4⁺ cells. Note that the percentage of CD25⁺ cells stimulated by VP2 is significantly higher. B: Percentages of Th3 and Tr1 cells. Note that 58~67% of the CD25⁺ cells were TGF-β⁺ (sum of the upper two quadrants). A majority of the CD25⁺IL-10⁺ cells were also TGF-β⁺ (the upper right quadrant). C: Percentages of CD25^- gated CD4⁺ cells. D: Percentages of Th3 and Tr1 cells. Note that the percentages of CD25^-TGF-β⁺ cells stimulated by VP1 and VP2 are significantly higher than stimulation with wild type NS3_358-375 and medium alone. Two representatives of eight experiments are shown.

Figure 4

Characteristics of antigen-driven CD25⁺ and CD25^- Tregs. Cellular culture and separation of CD4⁺ T cells and controls were the same as described in Figure 2. The purified CD4⁺ cells were stained with fluorescent-labeled antibodies for human CD25, CTLA-4 and Foxp3 followed by a flow cytometry assay. The effect of peptide NS3_358-375 and variant pools VP1 and VP2 on expression of Foxp3 (A) and CTLA-4 (B) on CD4⁺CD25^- (blue) and CD4⁺CD25⁺ (red) T cells is shown. Median values for each population are shown in the boxes. One representative of two experiments is shown.