Epstein-Barr virus mRNA vaccine synergizes with NK cells to enhance nasopharyngeal carcinoma eradication in humanized mice

Abstract

The close association between nasopharyngeal carcinoma (NPC) and Epstein-Barr virus (EBV) infection highlights the potential of therapeutic vaccination against viral antigens as an attractive immunotherapy for treating EBV⁺ NPC. Maximizing vaccine efficacy often requires selecting optimal T cell epitopes and incorporating co-treatment strategies. Here, we analyzed genomic mutations of 283 cancer-associated EBV strains and predicted epitopes with broad human leukocyte antigen (HLA) coverage from high-frequency nonsynonymous mutations. A polyepitope mRNA vaccine constructed from the predicted epitopes elicited antigen-specific T cell responses but showed suboptimal efficacy in tumor control in a PBMC-humanized mouse EBV⁺ NPC model. To enhance treatment efficacy, we developed an optimized system for expanding human natural killer (NK) cells with high purity and cytotoxicity as a co-treatment modality. Combined administration of mRNA vaccine and NK cells synergistically improved therapeutic efficacy by durably suppressing or eradicating NPC tumors in humanized mice. The concurrent treatment could improve the infiltration of both human T cells and NK cells into the tumor microenvironment and boost their effector functions. Our study suggests the combined therapeutic vaccination and NK cell therapy as a potential strategy for treating EBV⁺ NPC.

Keywords: Epstein-Barr virus; MT: Regular Issue; combined therapy; epitopes; humanized mouse model; mRNA vaccine; nasopharyngeal carcinoma; natural killer cells.

Graphical abstract

Figure 1

Prediction of immunogenic EBV epitopes with broad HLA coverage from genomic nsSNVs of cancer-related EBV strains (A) Schematic representation of the prediction pipeline. (B) Heatmap plot indicates the distribution of EBV genomic variations in different tumor types. Number of occurrences for each mutation in 283 sequenced strains, their HLA coverage, and exact amino acid substitution information are shown on the right. Mutations highlighted in purple are selected for vaccine construction due to their predicted strong HLA binding affinity and broad HLA coverage. Meta information, including tumor types and gender, is plotted at the bottom.

Figure 2

The nEL mRNA vaccine elicits antigen-specific T cell immune response against EBV⁺ NPC in humanized mice (A) Scheme of the animal experiment. Human PBMC-engrafted NOG mice were subcutaneously implanted with CNE2-EBV cells. One week after tumor implantation, mice were treated with nEL mRNA vaccine (vaccine) or liposomes (control) every 3 to 4 days (n = 5–6 per group). Spleen and tumor samples were collected for subsequent analysis at the end of the experiment. (B) A schematic diagram of the polyepitope nEL mRNA molecule. (C) Tumor growth curves of vaccinated and control mice (ordinary two-way ANOVA; ns, no significance; ∗∗p < 0.01). (D) Average tumor weight of vaccinated and control mice (unpaired two-tailed t test, ∗p < 0.05). (E) Post-vaccination response to the respective epitope for vaccinated and control mice. Representative images of splenic IFN-γ ELISpot assay (left) and the quantification of IFN-γ positive splenocytes (right, unpaired two-tailed t test, ∗∗∗∗p < 0.0001). (F) IHC staining of human CD4 and CD8 in tumor sections; scale bar, 100 μm. All data represented as means ± SEM.

Figure 3

Phenotypic characterization of expanded NK cells from donor peripheral blood (A) Flow cytometric characterization of NKG2D, NKp30, PD1, granzyme B, and perforin expression in expanded NK cells. (B) Average vitality and purity of NK cells, percentages of expanded NK cells expressing NKG2D, NKp30, PD1, granzyme B, and perforin in nine individual experiments. (C) Cytotoxicity assay of expanded NK cells against the leukemic cells K562 and CNE2-EBV cells (n = 3 per group).

Figure 4

Assessment of systemic engraftment and physiological impacts in NOG mice following human PBMC transplantation and allogeneic NK cell administration (A) Scheme of the animal experiment. Two weeks after human PBMCs engraftment (from 3 different donors), NOG mice were intravenously injected with allogeneic NK cells or saline solution containing 1% human serum albumins (control) every 3 to 4 days (n = 3 per group). Tissue samples (liver, spleen, heart, stomach, intestine, lung, brain, kidney, muscle, thymus, and lymph node) were collected for subsequent analysis at the end of the experiment. (B) Body weight and (C) food intake changes of PBMC-humanized mice in the control and NK-injected groups (n = 3) at different time points. (D) Tissue weight analysis of major organs at the experimental endpoint (ordinary one-way ANOVA with Tukey’s multiple comparisons test; ns, no significance; p > 0.05). All data represented as means ± SEM.

Figure 5

The nEL mRNA vaccine synergizes with NK cells to promote the eradication of EBV⁺ NPC in humanized mice (A) Scheme of the animal experiment. Human PBMC-engrafted NOG mice were implanted subcutaneously with CNE2-EBV cells. One week after tumor inoculation, mice were treated with mRNA vaccine and NK cells (Vac+NK), NK cells alone (NK), or liposomes and PBS (Control) every 3 to 4 days (n = 6 per group). Blood and tumor samples were collected for analysis at the end of the experiment. (B) Tumor growth curves of each group (ordinary two-way ANOVA with Tukey’s multiple comparisons test, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001). (C) Photograph of harvested tumor tissues from each group. The cross symbol indicates eradicated tumors by the combined therapy. (D) Average tumor weight of each group (ordinary one-way ANOVA with Tukey’s multiple comparisons test, ∗p < 0.05, ∗∗∗∗p < 0.0001). (E) HE staining and (F) IHC staining of EBNA1 in tumor sections from each group. Scale bars: 500 μm (top), 100 μm (bottom, enlarged boxed regions). (G) Body weight and food intake changes of PBMC-humanized mice in the control, NK, and Vac+NK groups at different time points (n = 6 per group). A group of non-humanized NOG mice was additionally recorded for reference. (H) Tumor growth curves of each group from humanized mice reconstituted with PBMC source from donor #3 (ordinary two-way ANOVA with Tukey’s multiple comparisons test, ∗∗p < 0.01, ∗∗∗∗p < 0.0001). All data represented as means ± SEM.

Figure 6

The combined therapy mobilizes both T cell and NK cell immunity against EBV⁺ NPC in humanized mice (A) Quantification of the serum concentrations of human IFN-γ (left) and human TNF-α (right) in humanized mice from each group (ordinary one-way ANOVA with Tukey’s multiple comparisons test; ns, no significance; p > 0.05, ∗p < 0.05, ∗∗∗∗p < 0.0001; data are represented as means ± SEM). (B) Characterization of T cell subtypes induced by individual predicted EBV antigen by intracellular cytokine staining. Data (means ± SEM) show the percentages of IFN-γ⁺/TNF-α⁺ CD4⁺ and IFN-γ⁺/TNF-α⁺ CD8⁺ T cells and are representative of two experiments. (C) IHC staining of human IFN-γ and granzyme B in tumor sections from each group. (D) IHC staining of human CD56, CD4, and CD8 in tumor sections from each group. Scale bar, 200 μm.