Table 1.
HLA-A*02 and -B*27 restricted epitopes.
Figure 1.
Functional avidity of HCV-specific CD8+ T cells.
(A) Representative flow cytometry plots showing intracellular IFN-γ staining of a CTL line exposed to target cells pulsed with a range of peptide concentrations. PBMCs from a subject chronically infected with HCV were stimulated for two weeks with the HLA-A*02-restricted NS31073 epitope. Prior to intracellular IFN-γ staining, the CTL line was cocultured for 5 hours with HLA-A*02+ EBV-immortalized B-cell lines (B-LCLs) pulsed with serially diluted concentrations of the cognate peptide. The background value from the negative control (B-LCLs without peptide loading) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+). (B) Representative data from intracellular IFN-γ staining of a CTL line specific for the HLA-A*02-restricted NS31073 epitope (filled circles) and a CTL line specific for the HLA-B*27-restricted NS5B2841 epitope (open circles) across a range of peptide concentrations. The background value from the negative control (B-LCLs without peptide loading) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+), which were then normalized to the maximum response by defining the smallest value as 0% and the largest value as 100%. (C) The functional avidity of CTL lines, derived from chronically HCV infected subjects, specific for HLA-A*02-restricted epitopes (n = 7, filled circles) and the immunodominant HLA-B*27-restricted epitope NS5B2841 (n = 6, open circles), as well as of CTL lines derived from subjects with resolved infection specific for HLA-A*02-restricted epitopes (n = 3, filled squares) and the immunodominant HLA-B*27-restricted epitope NS5B2841 (n = 4, open squares). Functional avidity was determined as the concentration of peptide required to achieve half-maximal IFN-γ induction (EC50). P-values were calculated using the Mann-Whitney U-test. Horizontal bars represent mean values.
Figure 2.
Functional profile of HCV-specific CD8+ T cells.
(A) Cytokine production and CD107a mobilization of CTL lines, derived from chronically HCV infected subjects specific for HLA-A*02-restricted epitopes (n = 4, black bars) and the immunodominant HLA-B*27-restricted epitope NS5B2841 (n = 4, white bars). Prior to intracellular multi-cytokine staining CTL lines were stimulated for 5 hours with specific peptides. The background value from the negative control (without peptide) was subtracted from all measured response frequencies (cytokine+ CD8+/total CD8+). Data are presented as mean ± SD. (B) Pie charts showing the mean multifunctionality of HLA-A*02-restricted (n = 4) and HLA-B*27-restricted (n = 4) HCV-specific CD8+ T-cell lines (one to five functions: CD107a, IFN-γ, IL-2, MIP-1β and TNF-α). Responding cells are grouped by the number of functions indicated by the numbers in the pie charts and matched to the color code in figure 2C. (C) Detailed functional profile of HLA-A*02-restricted (n = 4, black bars) and HLA- B*27-restricted (n = 4, white bars) HCV-specific CD8+ T-cell lines. Bars represent frequency of CD8+ T cells displaying the indicated combination of functions within the total population of responding CD8+ T cells. P-values were calculated using the Mann-Whitney U-test. Data are presented as mean ± SD.
Figure 3.
Antiviral efficacy of HCV-specific CD8+ T cells.
Antiviral efficacy (white bars) and functional avidity (filled circles) of CTL lines, derived from subjects with a chronic HCV infection specific for (A) HLA-A*02-restricted epitopes (n = 3) and (B) the immunodominant HLA-B*27-restricted epitope NS5B2841 (n = 3) as well as of CTL lines derived from subjects with resolved infection specific for (C) HLA-A*02-restricted epitopes (n = 3) and (D) the immunodominant HLA-B*27-restricted epitope NS5B2841 (n = 3). HuH7A2HCV and HuH7B27HCV cell lines were pulsed with a range of cognate peptide concentrations and cocultured for 24 hours with CTL lines at an E∶T ratio of 1∶1. Inhibition of viral replication was measured by luciferase activity. As a reference, intracellular IFN-γ staining at corresponding peptide concentrations is indicated. Measured RLUs and IFN-γ responses were normalized to maximum viral replication or maximum IFN-γ response, respectively, by defining the smallest value as 0% and the largest value as 100%. Data are presented as mean ± SD.
Figure 4.
Precursor frequency of HCV-specific CD8+ T cells.
(A) Representative flow cytometry plots of PBMCs from a healthy donor displaying staining with the NS5B2841/HLA-B*27 tetramer before and after magnetic-bead enrichment. Percentage of tetramer+ CD8+ T cells is indicated. (B) Precursor frequencies of CD8+ T cells specific for NS31406 (HLA-A*02 tetramer, filled circles) and NS5B2841 (HLA-B*27 tetramer, open circles) detected in healthy donors after tetramer staining, magnetic-bead enrichment and multiparametric flow cytometric analysis. The enriched tetramer+ CD8+ T-cell populations were stained for CD45RA, CD27, CCR7 and CD11a. The number of phenotypically naïve (CD45RAhi, CD27hi, CCR7hi, CD11alow) tetramer+ CD8+ T cells relative to the number of total CD8+ T cells is indicated. Horizontal bars represent mean values.
Figure 5.
Proteasomal digestion of HCV-specific epitope-containing regions.
(A) Peptides of 24 (A2-NS31073) and 25 (A2-NS5B2594 and B27-NS5B2841) amino acids in length spanning the regions containing the respective epitopes were used for incubation with constitutive proteasomes and immunoproteasomes. Epitope sequences are underlined. (B) Production of epitope-containing and non-epitope-containing cleavage products of the 24-mer (A2-NS31073) and two 25-mer (A2-NS5B2594 and B27-NS5B2841) peptides incubated with constitutive proteasomes and immunoproteasomes for 1 to 6 hours using standard conditions. The sum of all fragment intensities was set at 100%. Data are representative of triplicate experiments. (C) The relative production of all cleavage products of the B27-NS5B2841 peptide by constitutive proteasomes and immunoproteasomes for 1 to 6 hours using standard conditions. All fragments were between 4 and 9 amino acids long. Data are representative of triplicate experiments. (D) The relative production of all cleavage products of the B27-NS5B2841 peptide by constitutive proteasomes and immunoproteasomes for 1 to 6 hours using a 4× dilution of both of the proteasomal forms. The abundance of B27-NS5B2841-containing fragments ending in a C-terminal 2857D represents a maximum as this corresponds to the end of the 25-mer peptide substrate. All fragments containing the NS5B2841 epitope are highlighted using black symbols. The NS5B2841 epitope sequence is underlined. Data are representative of triplicate experiments.
Figure 6.
Epitope precursor degradation rates of natural HLA-A*02- and HLA-B*27-restricted epitopes as well as of an artificial chimeric epitope.
The percentage of undigested 25-mer precursor peptide left following 1, 2, 4, and 6 hours of constitutive and immunoproteasomal digestion, respectively. The amino acid sequences of the examined 25-mer peptides are shown with imbedded optimal epitopes underlined and the amount of purified proteasomal solution is indicated. B27-in-A2 signifies the B27-NS5B2841 epitope surrounded by the A2-NS5B2594 flanking regions, while A2-in-A2 and B27-in-B27 show these epitopes in their natural sequence contexts. The grey line indicates 50% degradation of the precursor peptide. Data are representative of triplicate experiments.
Figure 7.
Cross-recognition patterns of all B27-NS5B2841-containing fragments derived from immunoproteasomal digestion.
(A) Representative flow cytometry plots showing intracellular IFN-γ staining with CD8+ T cells from one subject across a range of peptide concentrations. PBMCs from HLA-B*27+ subjects chronically infected with HCV were enriched for CD8+ T cells and expanded non-specifically for 14 days. The background value from the negative control (without peptide loading) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+). Tested peptide sequences are indicated. (B) Representative cross-recognition patterns of all naturally processed B27-NS5B2841-containing fragments by enriched and non-specifically expanded CD8+ T cells (see Figure 7A). Responses were measured by intracellular IFN-γ staining against serial dilutions of all 7 B27-NS5B2841-containing peptide forms. The background value from the negative control (without peptide loading) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+). Symbol code for peptide sequences is indicated in Figure 7C. (C) Representative cross-recognition patterns of the NS5B2841- peptide by PBMCs specifically expanded by all naturally processed B27-NS5B2841-containing fragments. PBMCs from subjects chronically infected with HCV were stimulated for two weeks with all 7 B27-NS5B2841-containing peptide forms separately (see indicated symbol code). Responses were measured by intracellular IFN-γ staining against serial dilutions of the NS5B2841- peptide. The background value from the negative control (without peptide loading) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+).
Figure 8.
Processing and presentation of HCV-specific CD8+ T-cell epitopes.
(A) APCs were infected with vaccinia virus constructs carrying the relevant part of the HCV polyprotein (vHCV 827). After a defined time period (0, 2, 4, 8, 12, 16, 20 and 24 hours) these APCs were added to HCV-specific CTL lines and antigen-specific IFN-γ production was assessed by flow cytometry. (B) Representative flow cytometry plots of a CTL line stimulated with infected APCs. Prior to intracellular IFN-γ staining, the epitope-specific CTL line was stimulated for 5 hours with APCs previously infected with vHCV 827 and pre-incubated for 2, 4, 8 and 12 hours respectively. The background value from the negative control (T7 RNA polymerase (vTF7)) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+). (C) Representative time course showing intracellular IFN-γ staining of an HLA-A*02-restricted CTL line (filled circles) and an HLA-B*27-restricted (open circles) CTL line stimulated with infected APCs pre-incubated for 0 to 24 hours. The background value from the negative control (vTF7) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+), which were then normalized to the maximum response by defining the smallest value as 0% and the largest value as 100%. The complete kinetic (0 to 24 hours) was performed for all subjects, and the normalized responses were used to generate the data shown in Figure 8D and E. (D) Responses of CTL lines against APCs infected with vHCV 827 and pre-incubated for 2, 4 and 8 hours. Normalized IFN-γ responses of CTL lines specific for HLA-A*02-restricted epitopes (n = 9, filled circles) and for the immunodominant HLA-B*27-restricted NS5B2841 epitope (n = 7, open circles) are shown. Responses were normalized to the maximum response, as described for Figure 8C. Horizontal bars represent mean values. (E) Calculated peptide concentration present at the surface of infected APCs after 2, 4 and 8 hours of incubation with vHCV 827. The concentration of presented peptide restricted by HLA-A*02 (n = 9, filled circles; Table 1) and HLA-B*27 (NS5B2841; n = 7, open circles) was calculated by correcting for the individual functional avidity of the respective peptide-specific CTL line. For details of the calculations, see Figure S3. P-values were calculated using the Mann-Whitney U-test. Horizontal bars represent mean values. (F) Representative time course showing intracellular IFN-γ staining of an HLA-B*27-restricted CTL line stimulated with infected APCs incubated for 0 to 8 hours and pre-incubated with (triangles) or without (squares) the proteasome inhibitor epoxomicin. The background value from the negative control (vTF7) was subtracted from all measured response frequencies (IFN-γ+ CD8+/total CD8+). Data are representative of experiments performed with two different HLA-B*27-restricted CTL lines and one HLA-A*02-restricted CTL line.