Adenovirus-based vaccines against rhesus lymphocryptovirus EBNA-1 induce expansion of specific CD8+ and CD4+ T cells in persistently infected rhesus macaques

Abstract

The impact of Epstein-Barr virus (EBV) on human health is substantial, but vaccines that prevent primary EBV infections or treat EBV-associated diseases are not yet available. The Epstein-Barr nuclear antigen 1 (EBNA-1) is an important target for vaccination because it is the only protein expressed in all EBV-associated malignancies. We have designed and tested two therapeutic EBV vaccines that target the rhesus (rh) lymphocryptovirus (LCV) EBNA-1 to determine if ongoing T cell responses during persistent rhLCV infection in rhesus macaques can be expanded upon vaccination. Vaccines were based on two serotypes of E1-deleted simian adenovirus and were administered in a prime-boost regimen. To further modulate the response, rhEBNA-1 was fused to herpes simplex virus glycoprotein D (HSV-gD), which acts to block an inhibitory signaling pathway during T cell activation. We found that vaccines expressing rhEBNA-1 with or without functional HSV-gD led to expansion of rhEBNA-1-specific CD8(+) and CD4(+) T cells in 33% and 83% of the vaccinated animals, respectively. Additional animals developed significant changes within T cell subsets without changes in total numbers. Vaccination did not increase T cell responses to rhBZLF-1, an immediate early lytic phase antigen of rhLCV, thus indicating that increases of rhEBNA-1-specific responses were a direct result of vaccination. Vaccine-induced rhEBNA-1-specific T cells were highly functional and produced various combinations of cytokines as well as the cytolytic molecule granzyme B. These results serve as an important proof of principle that functional EBNA-1-specific T cells can be expanded by vaccination.

Importance: EBV is a common human pathogen that establishes a persistent infection through latency in B cells, where it occasionally reactivates. EBV infection is typically benign and is well controlled by the host adaptive immune system; however, it is considered carcinogenic due to its strong association with lymphoid and epithelial cell malignancies. Latent EBNA-1 is a promising target for a therapeutic vaccine, as it is the only antigen expressed in all EBV-associated malignancies. The goal was to determine if rhEBNA-1-specific T cells could be expanded upon vaccination of infected animals. Results were obtained with vaccines that target EBNA-1 of rhLCV, a virus closely related to EBV. We found that vaccination led to expansion of rhEBNA-1 immune cells that exhibited functions fit for controlling viral infection. This confirms that rhEBNA-1 is a suitable target for therapeutic vaccines. Future work should aim to generate more-robust T cell responses through modified vaccines.

FIG 1

AdC vaccine constructs and in vitro testing. (A) Schematic representation of the chimeric proteins SgD-rhEBNA-1 (top) and NBEFSgD-rhEBNA-1 (bottom). TM, transmembrane domain; light gray band, mutation to create SgD; dark gray bands, mutations that block HVEM binding (20). (B) Western blot analyses of protein lysates harvested from CHO-CAR cells infected with 10¹⁰, 10⁹, or 10⁸ vp of the indicated constructs. (C) Cell surface staining of infected CHO-CAR cells for SgD-EBNA1 and NBEFSgD-EBNA1. Cells were infected for 48 h with 10¹⁰ or 10⁹ vps of AdC68-SgD-EBNA1 or AdC68-NBEFSgD-EBNA1 or 10¹⁰ vps of AdC68-gD and stained with mouse anti-Flag antibody and probed with goat anti-mouse IgG-Alexa Fluor 700. (D) An HVEM binding assay was used to test the ability of the chimeric proteins to bind their receptor. Normalized protein extracts from CHO-CAR cells infected with AdC68 encoding SgD-rhEBNA-1 show enhanced HVEM binding (when comparing AUC by one-way ANOVA: versus AdC68-NBEFSgD-rhEBNA-1, P = 0.0036; versus AdC68-rhEBNA-1, P = 0.0035). (E) Mature DCs from an rhLCV-seropositive rhesus macaque were infected with the indicated Ad vectors or pulsed with rhEBNA-1 peptide and then cocultured with autologous rhEBNA-1-specific CD4⁺-depleted CD8⁺ T cells. rhEBNA-1-specific responses were measured by an IFN-γ ELISPOT assay after 16 h, and results are shown as average spots per 10⁶ CD8⁺ T cells based on quadruple experiments for the DC + T cells and triplicate experiments for the other samples. Significant differences are denoted by asterisks: all P values are <0.0001, except for peptide versus AdC68-rhEBNA-1 (P = 0.0009) and AdC68-gDrhEBNA-1 (P = 0.0036). P values were calculated using one-way ANOVA, with a P value of <0.05 considered significant. All multiple comparisons were Bonferroni adjusted to control for type I errors.

FIG 2

Magnitude of peptide-specific CD8⁺ T cell responses upon vaccination. PBMCs were stimulated with specific peptide pools, and ICS was used to measure production of IFN-γ, IL-2, and TNF-α. Mean counts of rhEBNA-1-specific (A) and rhBZLF-1-specific (B) total cytokine-producing CD8⁺ T cells ± SD are plotted as a change from baseline (BL) for all vaccine responders of groups 1 (×, black line) and 2 (asterisk, gray line) and all animals of group 3 (+, dashed line). Peptide-specific CD8⁺ T cell responses are shown for overall, CM, EM, and EFF subsets (left to right). Raw data were used to calculate changes from baseline, and any values below the limit of detection were excluded from the calculation of the mean. All values are presented as numbers of responding cells per 10⁶ live CD3⁺ cells. To calculate the sum of the peptide-specific response, we subtracted normalized background activity and then summed the 7 possible different combinations of functions. Significant differences were determined by comparing areas under the curve (AUC) and are as follows. For rhEBNA-1 versus rhBZLF-1 for all vaccine responders, CD8⁺ overall, P = 0.048, and EM, P = 0.039. For RhEBNA-1 for all vaccine responders versus controls, CD8⁺ overall, P = 0.048, and EM, P = 0.024. RhEBNA-1 CD8EM responses were also significant for vaccine responders within group 1 (rhEBNA-1 versus rhBZLF-1, P = 0.0081; rhEBNA-1 responders versus controls, P = 0.012). P values were calculated using two-sided Mann-Whitney tests (all responders grouped) or ANOVA (comparisons between all three groups), with a P value of <0.05 considered significant.

FIG 3

Proportions of peptide-specific CD8⁺ and CD4 ⁺T cell subsets upon vaccination. The proportions of rhEBNA-1-specific CD8⁺ (A) and CD4⁺ (B) EFF (dark gray), CM (light gray), and EM (white) subsets are shown as a percentage of the total response at baseline and weeks 2, 8, and 17 after vaccination. For each subset, normalized counts were divided by the sum of all three subsets to determine the relative proportions. Results for individual animals are shown as filled circles, and responses for all vaccine responders are displayed as floating bars (5th to 95th percentiles) with lines at the median. Significant differences were determined by comparing the percentage of a given subset after vaccination to the percentage at baseline and are as follows: rhEBNA1-specific CD8EM and CD4EM cells at week 2, P = 0.035 and 0.045, respectively. P values were calculated using the two-sided Mann-Whitney test, with a P value of <0.05 considered significant.

FIG 4

Magnitude of peptide-specific CD4⁺ T cell responses upon vaccination. Mean counts of rhEBNA-1-specific (A) and rhBZLF-1-specific (B) cytokine-producing CD4⁺ T cells ± SD are plotted as a change from baseline (BL). This figure is organized in the same fashion as Fig. 2. Significant differences are as follows. For rhEBNA-1 versus rhBZLF-1 for all vaccine responders, CD4⁺ overall, P = 0.0003; CM, P = 0.0019; EM, P = 0.02. For RhEBNA-1 for all vaccine responders versus controls, CD4⁺ overall, P = 0.02; CM, P = 0.039. RhEBNA-1 responses were also significant for vaccine responders within group 2 (for group 2 versus group 1, CD4⁺ overall, P = 0.0173; for group 2 rhEBNA-1 versus group 2 rhBZLF-1, CD4⁺ overall, P = 0.002; CM, P = 0.0009; EM, P = 0.021; for group 2 rhEBNA-1 versus control rhEBNA-1, CD4⁺ overall, P = 0.024; CM, P = 0.047; EM, P = 0.038). Responses of RhEBNA-1 CD4EFF cells were significantly lower than those of both CD4CM (P = 0.0058) and CD4EM (P = 0.019) cells. P values were calculated using two-sided Mann-Whitney tests (all responders grouped) or ANOVA (comparisons between all three groups), with a P value of <0.05 considered significant.

FIG 5

Cytokine profiles of rhLCV-specific T cells upon vaccination. Pies represent the average percentages of CD8⁺ and CD4⁺ T cells making each combination of IFN-γ (g), IL-2 (2), and TNF-α (a) following stimulation with specific peptides. RhEBNA-1-specific (A) and rhBZLF-1-specific (B) CD8EFF, CD8CM, and CD8EM responses are shown at baseline (BL) and at weeks 2, 8, and 17 after vaccination. The same time points are shown for rhEBNA-1specific (C) and rhBZLF-1-specific (D) CD4CM and CD4EM responses. Numbers of peptide-specific CD4EFF responses were all very low and are therefore not shown. For all responding animals, normalized cell counts were calculated for each function within every subset. Values for each of the seven possible combinations of functions were then divided by the sum of all seven, and pie charts reflect the ratios of those means. Significant differences are described in Results.

FIG 6

Granzyme B production by rhLCV-specific T cells. Six of the highest vaccine responders (three from each group) were evaluated for granzyme B-producing T cells. PBMCs were stimulated with overlapping peptide pools, and ICS was used to measure production of granzyme B (denoted B), IFN-γ (g), IL-2 (2), and TNF-α (a). Bars reflect the average magnitudes of granzyme B⁺ CD8⁺ (A) and CD4⁺ (C) memory cells producing each combination of cytokines after stimulation with specific peptides. The total sums of all peptide-specific cytokine responses at each time point are listed above the bars. Total values reflect the sum of every combination of the four functions tested; cells that produced only granzyme B were excluded. RhEBNA-1-specific (left) and rhBZLF-1-specific (right) responses are shown at baseline (BL) and at weeks 2, 8, and 17/19 after vaccination. CCR7 expression by rhEBNA-1-specific memory CD8⁺ T cells that produce granzyme B with all combinations of cytokines (red) is shown in comparison to total nonspecific memory (blue), naive (green), and effector (orange) populations. (B) Responses are shown at baseline and week 8 for one representative vaccine responder. Responses at weeks 2 and 17/19 were the same as at week 8 and are therefore not shown. Significant differences are described in Results.