Distinct memory CD4+ T-cell subsets mediate immune recognition of Epstein Barr virus nuclear antigen 1 in healthy virus carriers

Abstract

CD4+ T cells, specific for transforming latent infection with the Epstein Barr virus (EBV), consistently recognize the nuclear antigen 1 of EBV (EBNA1). EBNA1-specific effector CD4+ T cells are primarily T-helper 1 (TH1) polarized. Here we show that most healthy EBV carriers have such IFN-secreting EBNA1-specific CD4+ T cells at a frequency of 0.03% of circulating CD4+ T cells. In addition, healthy carriers have a large pool of CD4+ T cells that proliferated in response to EBNA1 and consisted of distinct memory-cell subsets. Despite continuous antigen presence due to persistent EBV infection, half of the proliferating EBNA1-specific CD4+ T cells belonged to the central-memory compartment (TCM). The remaining EBNA1-specific CD4+ T cells displayed an effector-memory phenotype (TEM), of which a minority rapidly secreted IFN upon stimulation with EBNA1. Based on chemokine receptor analysis, all EBNA1-specific TCM CD4+ T cells were TH1 committed. Our results suggest that protective immune control of chronic infections, like EBV, includes a substantial reservoir of TCM CD4+ TH1 precursors, which continuously fuels TH1-polarized effector cells.

Figure 1

The EBNA1-specific CD4⁺ T-cell clone A4. E116 responds to EBNA1 peptide pools containing the EBNA1 peptide sequence 513-527. (A) Scheme of EBNA1 amino acid sequence represented by overlapping peptide pools (EBNA1, all EBNA1 peptides; pool I, aa's 400-461; pool II, aa's 452-508; pool III, aa's 499-548; pool IV, aa's 539-593; and pool V, aa's 584-641). (B) ELISPOT assay indicating spots per 10⁴ cells in response to dilutions of peptide 513-527, and a noncognate peptide sequence as a negative control. Error bars indicate standard deviations of duplicates. (C) Intracellular cytokine staining assay for IFNγ in response to EBNA1 peptide pools, known CD8⁺ T-cell epitopes from CMV and EBV (Table S1), medium alone, and Staphylococcus enterotoxin B (SEB); note significantly positive response to the pool of all EBNA1 peptides and pool III, which both contain peptide 515-529. (D) CFSE proliferation assay in response to EBNA1 peptide pools. The data are representative of 3 independent experiments.

Figure 2

EBNA1-specific CD4⁺ T-cell responses among 40 healthy volunteers. (A) IFNγ secretion of PBMCs in response to all EBNA1 peptides (Master mix) and subpools I-V by ELISPOT assays after stimulation for 1 week with autologous dendritic cells pulsed with all EBNA1 peptides. Ten of 20 representative PBMC stimulations from leukocyte concentrates are shown. (B) Each data point represents the frequency of IFNγ-expressing CD4⁺ T cells from each volunteer to one of 3 stimuli. The stimuli are indicated along the x-axis: medium (no stimulus), CMV (HCMV-derived CD8⁺ T-cell epitopes; Table S1), and EBNA1 (the master mix of all 51 overlapping peptides of EBNA1; Table S1). The frequency of IFNγ-positive CD4⁺ T cells of each individual is indicated by a diamond. The average frequency of IFNγ-positive CD4⁺ T cells for each of the 3 stimuli is indicated by the black bars.

Figure 3

CD4⁺ T cells respond consistently to EBNA1 with IFNγ secretion and proliferation. EBNA1-specific CD4⁺ T cells were analyzed by intracellular IFNγ staining and CFSE proliferation assay. CD4⁺ T-cell responses in response to medium (no stimulus), Staphylococcal enterotoxin B (SEB), HCMV-derived CD8⁺ T-cell epitopes (CMV), all EBNA1 peptides (EBNA1), and subpools of EBNA1 peptides are shown. (A) Whole-blood assay after gating on lymphocytes based on size and on CD4⁺ T cells. The frequency of IFNγ-positive CD4⁺ T cells is indicated. Forward-scatter (FSC; x-axis) and intracellular cytokine staining for INFγ (y-axis) are depicted. The top row represents detected IFNγ responses from one of 20 EBV-positive carriers and the bottom row from one EBV-negative volunteer upon stimulation with the indicated stimuli. Gates for IFNγ-positive cells are based on isotype controls. (B) CFSE dilution assays characterize CD4⁺ T-cell proliferation in response to the indicated antigens and influenza virus infection (FLU). The frequencies of CD4⁺ T cells with diluted CFSE are indicated. The top row displays CD4⁺ T-cell responses from one representative of 20 healthy volunteers. The proliferation responses of an EBV-negative volunteer are depicted in the bottom row.

Figure 4

Frequencies of EBNA1-specific CD4⁺ T-cell responses are stable over time. (A) The frequency of EBNA1-specific IFNγ-expressing CD4⁺ T cells (◆) was assessed at the indicated time points in comparison to the frequencies of IFNγ-expressing CD4⁺ T cells in response to medium (◇) and CMV-derived CD8⁺ T-cell epitopes (□). (B) The average EBNA1 subpool (EBNA1_400-461; Table S1) recognition was determined from the assays conducted at the 7 different time points, indicated in panel A. Standard deviations are depicted. The data are representative for 3 healthy EBV carriers that were analyzed at 7 or more time points.

Figure 5

Frequency of EBNA1-specific proliferating CD4⁺ T-cell precursors. Each data point represents the frequency of CD4⁺ T-cell precursor cells from the original culture from each volunteer relative to one of 3 stimuli. The stimuli are indicated along the x-axis: medium (nothing added), CMV (HCMV-derived CD8⁺ T-cell epitopes; Table S1), and EBNA1 (the master mix of all 51 overlapping peptides of EBNA1; Table S1). The frequency of precursors from each individual is indicated by a diamond. The average frequency of precursors from each of the 3 stimuli is indicated by the black bars.

Figure 6

Phenotype of EBNA1- and influenza-specific IFNγ-secreting and proliferating CD4⁺ T cells. EBNA1- and influenza-specific CD4⁺ T-cell responses of one representative volunteer are shown. (A) After gating on CD3⁺CD4⁺ T cells, EBNA1-specific cells were identified by intracellular IFNγ staining (y-axis), and the expression of the indicated surface markers was analyzed (x-axis). The frequencies of EBNA1-specific CD3⁺CD4⁺ cells are shown in the appropriate quadrants. Gates were determined following an analysis with isotype controls (data not shown). One of 3 experiments is shown. (B) As described for panel A but characterizing IFNγ-positive influenza-specific CD4⁺ T cells. One of 2 experiments is shown. (C) After gating on CD3⁺CD4⁺ T cells, EBNA1-specific proliferation was identified by CFSE dilution (x-axis), and the expression of the indicated surface markers was analyzed (y-axis). The frequencies of CFSE dilute or proliferating EBNA1-specific CD4⁺ cells are shown in the appropriate quadrants. One of 3 experiments is shown. (D) As described for panel C but following CD4⁺ T-cell proliferation after influenza stimulation. One of 2 experiments is shown.

Figure 7

Stability of the central-memory phenotype and chemokine receptor expression on proliferating EBNA1-specific CD4⁺ T cells. (A) CFSE-labeled CD3⁺CD4⁺CD62L⁺ cells, purified by flow-cytometric cell sorting, were stimulated with the indicated antigens: medium (no stimulus), Staphylococcus enterotoxin B (SEB), influenza A infection (FLU), and the cognate EBNA1 peptide pool I. After 6 days, CD3⁺CD4⁺ T cells were analyzed for CD62L expression (y-axis) and CFSE dilution (x-axis). The frequencies of CD3⁺CD4⁺ T cells in each quadrant are indicated. One representative of 2 experiments is shown. (B) After gating on CD3⁺CD4⁺CD62L⁺ T cells, proliferation in response to 4 stimuli (medium, no stimulus; SEB, Staphylococcus enterotoxin B; influenza, influenza infection; and EBNA1, master mix of all 51 overlapping peptides of EBNA1) was identified by CFSE dilution (x-axis), and the expression of the indicated chemokine receptors (CXCR3, CCR4, and CXCRR5) was analyzed (y-axis). The frequency of CFSE dilute, and therefore proliferating, CD4⁺ cells is indicated in the appropriate quadrants. Gates and quadrants were determined after analysis with isotype controls (data not shown). One of 3 experiments is shown.