Broadly reactive human CD8 T cells that recognize an epitope conserved between VZV, HSV and EBV

Abstract

Human herpesviruses are important causes of potentially severe chronic infections for which T cells are believed to be necessary for control. In order to examine the role of virus-specific CD8 T cells against Varicella Zoster Virus (VZV), we generated a comprehensive panel of potential epitopes predicted in silico and screened for T cell responses in healthy VZV seropositive donors. We identified a dominant HLA-A*0201-restricted epitope in the VZV ribonucleotide reductase subunit 2 and used a tetramer to analyze the phenotype and function of epitope-specific CD8 T cells. Interestingly, CD8 T cells responding to this VZV epitope also recognized homologous epitopes, not only in the other α-herpesviruses, HSV-1 and HSV-2, but also the γ-herpesvirus, EBV. Responses against these epitopes did not depend on previous infection with the originating virus, thus indicating the cross-reactive nature of this T cell population. Between individuals, the cells demonstrated marked phenotypic heterogeneity. This was associated with differences in functional capacity related to increased inhibitory receptor expression (including PD-1) along with decreased expression of co-stimulatory molecules that potentially reflected their stimulation history. Vaccination with the live attenuated Zostavax vaccine did not efficiently stimulate a proliferative response in this epitope-specific population. Thus, we identified a human CD8 T cell epitope that is conserved in four clinically important herpesviruses but that was poorly boosted by the current adult VZV vaccine. We discuss the concept of a "pan-herpesvirus" vaccine that this discovery raises and the hurdles that may need to be overcome in order to achieve this.

Figure 1. Human CD8 T cells recognize a dominant epitope in the VZV ribonucleotide reductase subunit 2.

HLA-A*0201 binding predictions were used to identify candidate VZV-derived epitopes. PBMCs from 21 healthy adult HLA-A*0201+ volunteers were used in IFN-γ ELISpot assays to screen (a) peptide pools and subsequently (b) individual peptides for their ability to stimulate IFN-γ. Asterisks represent p-values<0.001 using Student's t-test. The A2-ILI tetramer was constructed and used to label epitope-specific CD8 T cells. (c) One representative donor is shown. The plot is gated on CD3+CD8+ lymphocytes. The number represents frequency of A2-ILI+ events as a percentage of CD8+ lymphocytes. (d) The frequency of A2-ILI+ CD8 T cells is shown along with the median. The dotted line represents the limit of detectability.

Figure 2. A2-ILI tetramer+ CD8 T cells display distinct phenotypic patterns between individuals.

A2-ILI tetramer+ CD8 T cells were co-stained with anti-CD45RA, CCR7, CD27, CD28, granzyme B, granzyme K, perforin, and Ki-67. Phenotypic markers were analysed by flow cytometry. (a) Subjects were categorized according to the combination of phenotypic markers expressed. One representative donor of each phenotype is shown. Numbers represent percentage of CD3+CD8+ lymphocytes. (b) The frequency of A2-ILI+ CD8 T cells in every subject expressing each marker is summarized. The medians are shown and p-values derived from Student's t-test. (c) PBMCs were collected at baseline (day 0) and one month or six months later. A2-ILI tetramer+ CD8 T cells were co-stained with anti-CD27, CD28, granzyme B, and perforin. Phenotypic markers were analysed by flow cytometry. Representative plots from one subject of each phenotype are shown.

Figure 3. ILI-specific CD8 T cells display functional impairment associated with their phenotype.

(a) PBMCs were stimulated with ILIEGIFFV peptide in vitro for 6 hours then co-stained for intracellular IFN-γ and IL-2. Numbers represent frequency as a percentage of CD8+ lymphocytes. The frequency of (b) IFN-γ+ cells as a proportion of A2-ILI tetramer+ cells and (c) IL-2 producing cells as a percentage of IFN-γ+ CD8 T cells are shown. P-values are derived from Student's t-test. (d) PBMCs from HLA-A*0201+ volunteers were stained with CFSE and stimulated with ILIEGIFFV peptide in vitro for 6 days. A2-ILI tetramer+ cells were then analyzed by flow cytometry. Numbers represent frequency as a percentage of CD8+ lymphocytes. Two representative donors are shown.

Figure 4. Expression of inhibitory receptors is associated with increased ILI-specific CD8 T cell frequency but impaired functionality.

A2-ILI-tetramer+ CD8 T cells were co-stained with anti-PD-1 and 2B4, and analysed using flow cytometry. (a) Representative donors with each phenotype are shown. Numbers represent frequency of events as a percentage of CD3+CD8+ lymphocytes. (b) Non-linear regression and Spearman's rank correlation were used to show the association between the frequency of IFN-γ producing A2-ILI+ cells and their frequency of PD-1 and 2B4 co-expression. (c) Non-linear regression and Spearman's rank correlation were used to show the association between the frequency of A2-ILI tetramer+ CD8 T cells and their expression of both PD-1 and 2B4. P-values for Spearman's rank correlation are shown.

Figure 5. ILI-specific CD8 T cells are broadly reactive, recognizing homologous epitopes conserved between alpha- and gamma-herpesviruses.

Homologous epitopes to ILIEGIFFV were identified by sequence similarity in HSV-1, HSV-2 and EBV. (a) The capacity of each epitope to stimulate cytokine production was measured by in vitro peptide stimulation. Two representative subjects are shown with their serostatus with respect to the herpesviruses tested. Numbers represent the percentage of CD3+CD8+ lymphocytes. (b) PBMCs from HLA-A*0201+ individuals with detectable ILI-specific responses underwent in vitro stimulation with each epitope at a concentration of 10 µg/ml followed by staining with anti-CD3, CD8, IFN-γ and TNF-α. The frequency of cytokine producing cells stimulated by the epitopes from VZV (black), HSV-1 (light blue), EBV (red), and HSV-2 (green) as a percentage of maximal cytokine producing cell frequency is shown. (c) The ability of each epitope to induce cytokine production was assessed by titration of stimulating peptide concentrations. One representative subject (subject 118, seropositive for all herpesviruses tested) is shown. The frequency of cytokine producing cells stimulated by the epitopes from VZV (black), HSV-1 (light blue), EBV (red), and HSV-2 (green) as a percentage of maximal cytokine producing cell frequency is shown.

Figure 6. The live attenuated vaccine Zostavax cannot efficiently induce proliferation of ILI-specific CD8 T cells in most individuals.

HLA-A*0201+ subjects were immunized with Zostavax. Blood was collected pre-vaccination and at 7, 14 and 28 days post-vaccination. (a) Whole blood was stained with anti-CD3, CD8 and A2-ILI tetramer. The frequency of A2-ILI+ cells post-vaccination in donors with detectable responses is shown. The only responder (subject 105) is marked in red and by an asterisk. (b) FACS plots indicating A2-ILI+ CD8 T cells from the responding donor are shown. Plots are gated on CD3+CD8+ lymphocytes and numbers represent the frequency of A2-ILI+ cells as a percentage of CD8+ lymphocytes. (c) PBMCs were co-stained with anti-Ki-67. The proportion of A2-ILI+ cells expressing Ki-67 post-vaccination in donors with detectable responses is shown. The single responder is marked in red and by an asterisk. (d) FACS plots show the expression of Ki-67 in A2-ILI+ CD8 T cells in the single responder. Plots are gated on CD3+CD8+ lymphocytes and numbers represent the percentage of A2-ILI+ cells with or without Ki-67 expression.