High frequency of functional anti-YMDD and -mutant cytotoxic T lymphocytes after in vitro expansion correlates with successful response to lamivudine therapy for chronic hepatitis B

Abstract

Background: Many determinants for a sustained response to lamivudine therapy have been reported but the role of T cell responsiveness remains unclear. The finding that tyrosine-methionine-aspartate-aspartate (YMDD) motif of the reverse transcriptase domain of hepatitis B virus (HBV) DNA polymerase carries a HLA-A2 restricted cytotoxic T lymphocyte (CTL) epitope makes quantitative measurement of the numbers of peptide specific CTLs feasible using MHC tetramer-peptide complex staining.

Aim: To investigate the correlation between anti-YMDD motif CTL activity and the efficacy of lamivudine therapy in HLA-A2 positive patients with chronic hepatitis B (CH-B).

Methods: The function and phenotype of peptide and interleukin 2 expanded peripheral blood mononuclear cells were quantified by cell lytic assay and immunocytochemical analysis by staining with HLA-A2-peptide tetramer complexes.

Results: After in vitro expansion, sustained responders had more potent CTL responses against YMDD, YVDD, and YIDD, as well as other epitopes on HBV antigens than non-responders. The frequency of YMDD/YVDD/YIDD motif specific CTLs increased significantly with an effective cell lytic function during and after therapy in sustained responders but not in non-responders. YMDD specific CTLs cross reacted with YIDD and YVDD mutant epitopes, and shared T cell receptor gene usages with YIDD and YVDD specific CTLs.

Conclusions: Sustained responders, at least HLA-A2 patients, elicited a more potent CTL immunity against YMDD and its mutants. YMDD specific CTLs are cross reactive with YVDD and YIDD mutant epitopes, which may further contribute to immune clearance of the mutant viruses and a successful response to lamivudine therapy in CH-B patients.

Figure 1

Protocol of the modified CTL response index of the peptide (CRI-p) culture method. On the first day of blood sampling, 5×10⁶ peripheral blood mononuclear cells (PBMCs) in complete medium in triplicate cultures in a six well flat bottomed plate (Nunc, Roskilde, Denmark) with a mixture of the panel of peptides to be tested (each 1 μg/ml), 1 μg/ml recombinant hepatitis B core antigen (rHBcAg), and 20 U/ml recombinant human interleukin 2 (rhIL-2) were incubated at 37°C in a humidified 5% CO₂ incubator. On days 3, 5, 7, 9,11, and 13, the culture medium was refreshed with 20 U/ml of rhIL-2. Usually, the total number of cells could proliferate up to 3∼5×10⁷ cells on day 13, and sometimes to 10⁸ cells by day 14. Cells harvested on days 14–15 were used for cell lytic assays and phenotypic study, with surface staining of MHC-A2-peptide tetrameric complexes. CTL, cytotoxic T lymphocyte; TCR, T cell receptor.

Figure 2

Cell lytic analysis of cells expanded using the CTL response index of the peptide (CRI-p) method. (A) Pretreatment and three months post treatment frequency of peptide specific CD8 T cells from four complete response (CR; patients 2, 8, 9, and 14) and four non-response (NR)+ partial response (PR) (patients 1, 6, 11, and 12) cases. Both pre- and post treatment levels of peptide specific CD8 T cell frequencies were higher in CR than in NR+PR patients. CTL, cytotoxic T lymphocyte. (B) Cell lytic activities of cells from patient No 3 (CR to lamivudine therapy) and patient No 4 (NR to lamivudine therapy) were assayed after in vitro expansion by the modified CRI-p method. Both cases had enhancement of killing capacity against various peptide epitopes of wild-type hepatitis B virus (HBV) antigens three months after treatment. Patient No 3 elicited more vigorous anti-YMDD motif killing activity than patient No 4. The broken line set at 30% is the mean level (+3 SD) of the 11 non-HLA-A2 patients. Levels above this value were considered positive responses. (C) Phenotypic assay of peptide specific CD8 T cells. Representative fluorescence activated cell sorter analysis of tetramer assay in patient No 3 (top) and patient No 4 (bottom panel) using cells cultured from the modified CRI-p culture for surface staining. Compared with patient No 4, patient No 3 showed significantly higher percentages of tetramer staining for YMDD, YVDD, and YIDD as well as for HBV core 18–27 tetramers. (D) Frequencies of peptide specific CD8 T cells were higher in patients with CR than in those with NR/PR. Data for the frequencies of peptide specific CD8 T cells were pooled and compared between patients with CR (seven patients) and those with NR/PR (seven patients) (see table 1 ▶) using the Mann-Whitney U test.

Figure 3

Cell lytic activities of peptide specific cytotoxic T lymphocyte (CTL) cells. Positively sorted peptide specific CD8 T cells after in vitro expansion by the modified CTL response index of the peptide (CRI-p) method were cocultured with hepatitis B virus (HBV) infected or peptide pulsed B-LCL target cells. CTLs including YMDD, YIDD, and YIDD tetramer stained CD8+ T cells sorted by fluorescence activated cell sorter analysis from different complete response patients were used as effector cells (see materials and methods) with three different effector/target cell ratios (E/T ratio = 40, 20, and 10). Target cells were either peptide pulsed, HBV transfected, or mock transfected B-LCL cells.

Figure 4

Cross reactivity of YMDD motif specific cytotoxic T lymphocytes (CTLs) against YIDD and YVDD but not hepatitis B core antigen (HBcAg) 18–27. YMDD cell fraction sorted from patient No 3 after in vitro expansion also killed B-LCL cells pulsed with 1 μg/ml YIDD or YVDD peptide as efficiently as target cells pulsed with YMDD peptide (all induced approximately 35–40% of specific target cell lysis at an effector/target cell (E/T) ratio of 20) but not with hepatitis B virus (HBV) core 18–27 peptide (A, B, and C v D). Non-HLA-A2 B-LCL target cells infected with HBV or pulsed with YMDD were used as controls (E).

Figure 5

Authenticity of T cell receptor α variable (AV) and β variable (BV) primers for polymerase chain reaction (PCR) analysis of variable gene usages in phytohaemagglutinin stimulated peripheral blood mononuclear cells. (A) Ethidium bromide staining of PCR products. (B) Autoradiography profile. The size of the amplified products using 5′ Vα (or Vβ??and 3′ Cα (or Cβ) primers ranged from 320 to 520 base pairs (BP), and that of 5′ and 3′ actin primers was approximately 500 base pairs, which was used as the PCR control (data not shown).

Figure 6

Shared T cell receptor α variable (AV) and β variable (BV) gene usages in YMDD motif specific T cells. Only data from patient No 3 are shown. Consistent results were also detected in patient Nos 2, 5, 8, 9, 10, and 14.

Figure 7

Frequency of YIDD+YVDD specific cytotoxic T lymphocytes (CTLs) was higher than that of YMDD specific CTLs in complete response (CR) patients. Total frequency of YMDD mutant (that is, YIDD+YVDD) tetramer stained CD8+ T cells after expansion using the modified CTL response index of the peptide (CRI-p) method from patient No 2 was compared with that of YMDD specific and hepatitis B core antigen (HBc 18–27) specific CTLs (A) three months after therapy. Before therapy, the frequency of YIDD was 2.9% (table 1 ▶) after in vitro expansion. Overall, CR patients had a significantly higher YIDD and YVDD specific CTL frequency than YMDD specific cells (B). ***p = 0.001, Mann-Whitney U test.