Identification of protective T-cell antigens for smallpox vaccines

Abstract

Background aims: E3L is an immediate-early protein of vaccinia virus (VV) that is detected within 0.5 h of infection, potentially before the many immune evasion genes of vaccinia can exert their protective effects. E3L is highly conserved among orthopoxviruses and hence could provide important protective T-cell epitopes that should be retained in any subunit or attenuated vaccine. We have therefore evaluated the immunogenicity of E3L in healthy VV-vaccinated donors.

Methods: Peripheral blood mononuclear cells from healthy volunteers (n = 13) who had previously received a smallpox vaccine (Dryvax) were activated and expanded using overlapping E3L peptides and their function, specificity and antiviral activity was analyzed. E3L-specific T cells were expanded from 7 of 12 (58.3%) vaccinated healthy donors. Twenty-five percent of these produced CD8+ T-cell responses and 87.5% produced CD4+ T cells. We identified epitopes restricted by HLA-B35 and HLA-DR15.

Results: E3L-specific T cells killed peptide-loaded target cells as well as vaccinia-infected cells, but only CD8+ T cells could prevent the spread of infectious virus in virus inhibition assays. The epitopes recognized by E3L-specific T cells were shared with monkeypox, and although there was a single amino acid change in the variola epitope homolog, it was recognized by vaccinia-specific T-cells.

Conclusions: It might be important to include E3L in any deletion mutant or subunit vaccine and E3L could provide a useful antigen to monitor protective immunity in humans.

Keywords: E3L; smallpox; vaccinia virus (VV); variola virus; virus-specific T cells.

Figure 1

Response of healthy vaccinated donors to E3L. PBMCs were stimulated with autologous DCs loaded with E3L pepmixes on days 0 and 9 in the presence of IL-4 and IL-7. The frequency of E3L-specific T cells generated from 12 donors was measured on day 16 by IFN-γ ELISpot in response to stimulation with the E3L pepmix (black bars) is shown. Results are expressed as spot-forming cells (SFC) per 1 × 10⁵ cells. Control was IFN-γ release in response to stimulation with irrelevant pepmix (gray bars). The data shown represent repeated experiments with all responders at least three times, whereas experiments for non-responders were repeated twice.

Figure 2

Identification of CD8+ T-cell epitopes within E3L from HLA-B35 donors. (A) CD8 PBMCs from donor 1 were stimulated with autologous DCs loaded with E3L pepmix in presence of IL-4 and IL-7. E3L-specific CD8 T-cells were measured by IFN-γ ELISpot in response to stimulation with E3L peptide pools on day 16. Results are expressed as spot-forming cells (SFC) per 1 × 10⁵ cells. (B) Design of peptide pools. Each pool contained 5 to 6 peptides and was assigned a number from 1 to 12. The numbers of the pools (top low and left column) are shown in bold. Individual peptides (n = 35) in these 12 pools correspond to the numbers in the respective columns and row. Each peptide was uniquely present in two pools. (C) E3L-specific CD8+ T-cells were stimulated with individual E3L peptides 20 to 27, and IFN-γ production measured in by ELISPOT. (D) The cytotoxic activity of E3L-specific CD8+ T-cells from donor 1 was measured in a 6 hours ⁵¹Cr release assay using E3L peptide 22-loaded autologous OKT3 blast target cells with or without blocking antibodies to HLA class I and II at an E:T ratio of 20:1. (E) Specific lysis of autologous or allogeneic OKT3 blasts matched at single HLA alleles and pulsed with E3L peptide 22 was evaluated in a 6-h Cr⁵¹ release assay, at E:T ratios of 40:1, 20:1, 10:1 and 5:1 . (F) Specific lysis of autologous and HLA-B35 matched allogeneic OKT3 blast pulsed with E3L peptide 22 the presence or absence of a HLA class I blocking antibody was evaluated in a 6-h ⁵¹Cr release assay in at E:T ratio of 20:1. (G) Fine mapping of E3L 9-mer peptide within peptide 22 (20-mer). E3L-specific CD8+ T cells from donor 1 and donor 2 (HLA-B35 donors) were stimulated with individual 9-mer peptides and IFN-γ production measured in ELISpot assays. (H) E3L-specific CD8+ T-cells from donors 1 and 2 were stained with an HLA-B35/NPV tetramer and anti-CD8 antibody and analyzed by FACS.

Figure 3

Identification of CD4+ T-cell epitopes within E3L from HLA-DR15 donors. (A) CD4+ T-cells from donor 3 were stimulated with autologous DCs loaded with E3L pepmixes in presence of IL-4 and IL-7, and the response to stimulation with E3L peptide pools was measured on day 9. Results are expressed as spot-forming cells (SFC) per 1 × 10⁵ responder cells. (B) Identification of reactive peptides 31 and 32 within the peptide pools. (C) The γ-IFN response of E3L-specific CD4+ T cells to single peptides 31 and 32 with or without blocking antibodies to HLA class II DR. (D) E3L-specific CD4+ T cells from donors 3 to 6 (HLA-DR15 donors) were stimulated with peptides 31 and 32, with or without HLA class II DR, DP and DQ blocking antibodies and the response was measured by IFN-γ ELISpot assay. (E) To fine map the minimal HLA class II–restricted epitope within peptides 31 and 32 (20-mers), E3L-specific CD4+ T cells from donors 3 to 6 were stimulated with 15-mer peptides and the responses measured by IFN-γ ELISpot.

Figure 4

Variant E3L epitope in variola and modified vaccinia nkara (MVA). (A) E3L-specific CD8+ T-cells from donor 1 and donor 2 were restimulated with vaccinia E3L peptide (aa 117-125:NPV) or the variola/MVA variant E3L peptide (aa117-125:NPV V→I). (B) Specific lysis of target cells alone or loaded with vaccinia and variola peptides in was evaluated in a 6-h Cr⁵¹ release assay at E:T ratios of 40:1, 20:1, 10:1 and 5:1. (C) E3L-specific CD8+ T cells from donors 1 and 2 were stained with the variola variant tetramer - HLA-B35/NPV V→I (and anti-CD8 antibody and analyzed by FACS.

Figure 5

Vaccinia-specific T cells can kill VV-infected LCL and prevent VV replication. (A) E3L-specific CD8+ T cells from donors 1 and 2 were cultured with autologous or HLA-B35 matched allogeneic LCLs at 2, 4 and 8 h after infection with VV. Specific lysis was evaluated in 6-h Cr⁵¹ release assay at E:T ratio of 40:1, 20:1, 10:1 and 5:1. (B) VV-infected LCLs were cultured alone or with E3L-specific CD8+ T cells at a T-cell to VV–infected LCL ratio of 5:1, or with stock VV or medium alone. Virus titers were measured as described in material and methods.

Figure 6

Comparative function of T cells specific for E3L and other vaccinia proteins using cytotoxicity and virus inhibition assay. (A) The cytotoxic activity of C7L- and E3L-specific CD8+ T cells from donor 2 against VV infected or pepmix-pulsed, HLA-B35 matched allogeneic LCLs plated 2, 4 and 8 hours after infection was evaluated in a 6-h Cr⁵¹ release assay at E:T ratio of 40:1, 20:1, 10:1 and 5:1. E3L- and C7L-specific CD8+ T cells both killed VV infected HLA-B35 matched allogeneic donor LCL in all the conditions of 2, 4 and 8 h after infection with VV. Only E3L prevented VV replication in co-culture virus inhibition assay (right panel). (B) B22R-specific CD8+ T cells from donor 4 were able to kill pepmix pulsed, but not VV-infected, LCLs in chromium release assay (left panel), and were unable to prevent infectious virus spread in cocultures (right panel). (C) The cytotoxic activity and viral inhibition assay of A10L-specific CD8+ T-cells from donor 5. A10L-specific CD8+ T cells killed pepmix-pulsed LCLs and VV infected LCLs if added 8 h after LCL infection, but they did not kill VV-infected LCL when added 2 or 4 h after infection with VV. A10L-specific T cells did not prevent infectious virus spread in infected LCLs. (D) CD4+ T cells from donor 4 that were specific for A10L-, H3L-, G5R- and E3L as measured by IFN-γ ELISpot (left panel) did not prevent VV replication in coculture with VV-infected autologous LCLs (right panel). (E) CD4+ T cells from donor 5 secreted IFN-γ in response to A10L-, H3L- and E3L in ELISPOT assays (left panel) did not inhibit VV replication in cocultures with VV-infected autologous LCLs.