Detection of diverse hepatitis C virus (HCV)-specific cytotoxic T lymphocytes in peripheral blood of infected persons by screening for responses to all translated proteins of HCV

Abstract

Broadly directed hepatitis C virus (HCV)-specific cytotoxic T lymphocytes (CTL) have been identified from liver-infiltrating lymphocytes but have been more difficult to assess in peripheral blood of infected persons. To enhance the detection of CTL from peripheral blood mononuclear cells (PBMC), we cocultured PBMC with autologous Epstein-Barr virus-transformed B-lymphoblastoid cell lines that had been infected with recombinant vaccinia virus constructs so that they expressed the entire translated polyprotein of HCV-H, a type 1a strain. These stimulated cells from HCV-infected as well as exposed seronegative persons were then cloned at limiting dilution and tested for HCV-specific CTL activity using a standard (51)Cr release assay. HCV-specific CTL were detected in PBMC from seven of nine persons with chronic hepatitis, including five of seven in whom CTL had previously been detected from liver biopsy specimens but not PBMC. In a single person with chronic HCV infection, CTL directed against as many as five different epitopes were detected in peripheral blood and were similar in specificity to those detected in liver tissue. This technique was used to evaluate eight subjects identified to be at high risk for HCV exposure due to continued injection drug abuse; no evidence of CTL in PBMC was found. We conclude that CTL can be detected in PBMC from the majority of persons with chronic HCV infection but are present at lower levels or absent in exposed but persistently seronegative persons. The high degree of concordance of HCV epitopes identified from liver and PBMC suggests that this strategy is a reasonable alternative to liver biopsy for characterizing the CTL response to HCV in chronically infected persons.

FIG. 1

HCV-specific CTL are present in PBMC at low frequencies. Subject P1 was selected for study because she had chronic HCV infection and HCV-specific CTL previously identified from liver biopsy-derived lymphocytes. (A) PBMC were stimulated with 12F6, a CD3-specific monoclonal antibody, and tested after 2 weeks for CTL activity in a standard 4-h chromium release assay. HCV-H antigens were expressed on autologous B-LCL target cells using a recombinant HCV-vv infection. No HCV-specific CTL activity was detected compared to cells expressing the control LacZ protein. (B) Fresh (previously unstimulated) PBMC were cloned at limiting dilution and stimulated with 12F6 and allogeneic irradiated PBMC feeders. Of the 105 clones screened for CTL activity, 2 clones (P5-11 and P5-28.6) with CTL activity against target cells expressing the E2/NS2 proteins of HCV were identified. This suggested that HCV-specific CTL were present at low frequencies. (C) PBMC from subject P1 were also stimulated with antigens from the entire translated polyprotein of HCV-H expressed on autologous B-LCL. After 2 weeks in culture, the bulk stimulated cells were tested for CTL activity in a standard 4-h chromium release assay. HCV-specific CTL activity was detected against B-LCL expressing aa 1 to 966 and 827 to 3011 but not the negative control expressing the lacZ gene product.

FIG. 2

CTL identified from PBMC recognize the same epitopes as CTL identified from liver of subject P1. PBMC from subject P1 that had been stimulated with HCV-H antigens expressed by autologous B-LCL were subcloned and tested for recognition of five epitopes previously identified from this patient from liver CTL. Synthetic peptides corresponding to the identified CTL epitopes were incubated with ⁵¹Cr-labeled autologous B-LCL and used as target cells in a standard chromium release assay. CTL clones derived from PBMC (solid bars) and liver (clear bars) specifically recognized the epitopes (A) TINYTIFK, aa 621-628 in E2 envelope, (B) TLGFGAYMSK, aa 1261 to 1270, and (C) HSKKKCDEL, aa 1395 to 1403 in the NS3 protein, (D) TLGFGAYMSK, aa 1636 to 1643 in NS4, and (E) SLTPPHSAK, aa 2510 to 2518 in the NS5b protein.

FIG. 3

HCV-specific PBMC CTL detected in majority of subjects with intrahepatic CTL. (A) Seven subjects (P1 to P7) known to have intrahepatic CTL responses, two subjects with no detectable intrahepatic CTL responses (P8 and P9), and three HCV-seronegative controls (N1 to N3) were tested for PBMC CTL after stimulation with HCV-H antigens expressed on autologous B-LCL using a 4-h chromium release assay. Assays were scored as positive if the total specific lysis was greater than 20% and at least 15% greater than the background nonspecific lysis. HCV-specific CTL activity was detected in bulk stimulated cultures of subjects P1, P2, and P3 (indicated by asterisks). (B) Bulk-stimulated cells were subcloned and tested again for CTL activity. Where the bulk CTL assay had been scored as positive, HCV-specific CTL clones recognizing the appropriate regions identified using vaccinia virus vectors were detected in all three subjects (P1 to P3). Furthermore, PBMC CTL clones targeting epitopes in regions where the bulk CTL assay was scored as negative were found in subject P3 (aa 2152 to 2160) and in an additional two subjects (P4 and P5). All CTL clones recognized the optimal epitope in the context of a specific restricting HLA class I allele with high specificity. Nonspecific lysis to irrelevant peptides for all clones was <5%.