Herpes zoster mRNA vaccine induces superior vaccine immunity over licensed vaccine in mice and rhesus macaques

Abstract

Herpes zoster remains an important global health issue and mainly occurs in aged and immunocompromised individuals with an early exposure history to Varicella Zoster Virus (VZV). Although the licensed vaccine Shingrix has remarkably high efficacy, undesired reactogenicity and increasing global demand causing vaccine shortage urged the development of improved or novel VZV vaccines. In this study, we developed a novel VZV mRNA vaccine candidate (named as ZOSAL) containing sequence-optimized mRNAs encoding full-length glycoprotein E encapsulated in an ionizable lipid nanoparticle. In mice and rhesus macaques, ZOSAL demonstrated superior immunogenicity and safety in multiple aspects over Shingrix, especially in the induction of strong T-cell immunity. Transcriptomic analysis revealed that both ZOSAL and Shingrix could robustly activate innate immune compartments, especially Type-I IFN signalling and antigen processing/presentation. Multivariate correlation analysis further identified several early factors of innate compartments that can predict the magnitude of T-cell responses, which further increased our understanding of the mode of action of two different VZV vaccine modalities. Collectively, our data demonstrated the superiority of VZV mRNA vaccine over licensed subunit vaccine. The mRNA platform therefore holds prospects for further investigations in next-generation VZV vaccine development.

Keywords: Shingrix; Varicella Zoster Virus; immune mechanisms; mRNA vaccine; nonhuman primate.

Figure 1.

Antibody and memory B cell responses induced by ZOSAL and Shingrix in mice. a. Experimental design. C57BL/6 mice (n = 6) were i.m. immunized with escalating doses of ZOSAL or 0.1 human dose of Shingrix on day 0 and day 14. Blood draws were taken at the indicated time points. Spleens and dLNs were collected 28 days after the boost. b. Anti-gE IgG titres were measured by ELISA and endpoint titres are shown. c. Anti-gE IgG1 and IgG2c titres at day 28 were measured by ELISA and endpoint titres are shown. d. Ratio of IgG2c/IgG1 is shown. e-f. ADCD and ADNP functions of Abs were analyzed using sera collected on day 28. gE-coated microbeads were incubated with diluted and heat-inactivated sera. ADCD (e) was detected by fluorescently labelled anti-C3 Abs and MFIs are shown. ADNP (f) was determined by beads-positive primary neutrophils and phagocytic scores are shown. g-h. Frequencies of class-switched (IgD^-IgM^-) gE-specific MBCs in dLNs and spleens were assessed by flow cytometry. Data are shown as mean ± SEM. Mann-Whitney U test was used for statistical analysis. *p ≤ 0.05, **p ≤ 0.01.

Figure 2.

T cell responses induced by ZOSAL and Shingrix in mice. C57BL/6 mice (n = 6) were i.m. immunized with escalating doses of ZOSAL or 0.1 human dose of Shingrix on day 0 and day 14. Spleens were collected 28 days after the boost immunization. a. Splenocytes were stimulated with or without gE antigen (2 μg/ml) for 8 h in the presence of Brefeldin A. Frequencies of IFN-γ, IL-2, TNF, or IL-21-secreting CD4⁺ T cells were analyzed by flow cytometry. Data from one representative animal is shown. b–d. Splenocytes were stimulated with or without gE antigen (2 μg/ml) for 20 h. Frequencies of AIM ⁺CD4⁺ T cells (b) and AIM+ Tfh cells (c,d) were determined by flow cytometry. e. Expression of ICOS on OX40 ⁺CD137 ⁺CXCR5 ⁺CD4⁺ Tfh cells was evaluated. MFI value is shown. Data represent mean ± SEM. Mann-Whitney U test was used for statistical analysis. *p ≤ 0.05, **p ≤ 0.01.

Figure 3.

Alteration of intermediate monocytes and modulation of gene expression after ZOSAL and Shingrix vaccination in rhesus macaques. a. Experimental design. Rhesus macaques were i.m. immunized twice with ZOSAL (n = 4) or Shingrix (n = 3) at an interval of 4 weeks. Blood and serum samples were taken at the indicated time points for analysis. b. Differentiation of three monocyte subsets 24 h after prime and boost immunization is shown. Classical monocyte (CM), intermediate monocyte (IM), or nonclassical monocyte (NCM). Pie charts indicate the percentage of each monocyte subset out of total monocytes. c. Frequencies of CD14 ⁺CD16⁺ intermediate monocytes are shown. d-h. Transcriptomic analyses of PBMCs isolated 24 h post the boost immunization. d. The Venn diagram indicates the number of altered genes shared by the two vaccine groups. D1 > D0 and D1 < D0 represent up-regulated and down-regulated genes, respectively. e. Volcano plots display genes that were altered by ZOSAL or Shingrix vaccination. Criteria used are p value < 0.05 calculated using a paired two-tailed Student’s t test and at least 2-folded change after vaccination. Representative genes are annotated. f. Heatmaps of significantly altered DEGs by ZOSAL or Shingrix are shown. Values in heatmaps are z-score standardized. g. DEGs in ZOSAL-vaccinated animals versus Shingrix-vaccinated animals were employed for GO analysis, with a focus on the indicated GO biological process terms. h. GSEA analysis using all DEGs. Each box represents a specific module and colours indicate normalized enrichment score (NES). The asterisk denoted in the box represents False Discovery Rate (FDR) values < 0.25. Two-way ANOVA was used for statistical analysis in figure c. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Figure 4.

Antibody and memory B cell responses induced by ZOSAL and Shingrix in rhesus macaques. Rhesus macaques were i.m. immunized twice with ZOSAL (n = 4) or Shingrix (n = 3) at an interval of 4 weeks. a. Endpoint titres of anti-gE IgG were measured by ELISA longitudinally. b. Frequencies of class-switched IgD^-IgM^- gE-specific MBCs in PBMCs were assessed by flow cytometry. Data from representative animals (left panel) and cell frequencies are shown as mean ± SEM (right panel). c. Correlation of class-switched gE⁺ MBCs and anti-gE IgG titres. d. gE-coated microbeads were incubated with diluted and heat-inactivated sera, followed by incubation with THP-1 cells. ADCP effect of Abs was determined as frequencies of beads-positive cells and phagocytic scores are shown. e. ADCD effect of Abs was detected by fluorescently labelled anti-C3 Abs and MFIs are shown. Pearson’s correlation analysis was used. Data are shown as mean ± SEM. ****p ≤ 0.0001.

Figure 5.

A stronger Th1-biased CD4+ T cell response was induced by ZOSAL than Shingrix in rhesus macaques. Rhesus macaques were i.m. immunized twice with ZOSAL (n = 4) or Shingrix (n = 3) at an interval of 4 weeks. a-b. PBMCs collected at different time points before and after vaccination were stimulated with gE antigen (2 μg/ml) for 20 h. Frequencies of IFN-γ or IL-2-secreting T cells at the indicated time points were measured by ELISpot. Data from representative animals are shown. Numbers of gE-specific cytokine-producing T cells were enumerated and are shown as spots per million stimulated cells. c. PBMCs were stimulated with or without gE overlapping peptides pool (10 μg/ml) for 16 h in the presence of Brefeldin A. Frequencies of IFN-γ, IL-2, TNF-secreting CD4⁺ T cells were analyzed by flow cytometry. Data from representative animals are shown. d. Quantification of cytokine-producing CD4⁺ T cells upon antigen stimulation. Two-way ANOVA with multiple comparison tests was used for analysis of statistical significance. ***p ≤ 0.001.

Figure 6.

Correlation and cluster analysis of immune parameters. a. A multivariate nonparametric Spearman’s test was used to analyze the correlation among 44 immune parameters measured in the study. The heatmap shows the correlation coefficient with four clusters denoted. *p ≤ 0.05, **p ≤ 0.01. b. Immune parameters correlating with gE-specific Th1 T cell responses were isolated and shown.

Figure 7.

A preliminary evaluation of safety profiles of ZOSAL and Shingrix in rhesus macaques. Rhesus macaques were i.m. immunized twice with ZOSAL (n = 4) or Shingrix (n = 3) at an interval of 4 weeks. a. Levels of serum biochemical parameters at the indicated time points were measured. b. GSEA analysis revealed distinct differences in the level of indicated modules related to platelet function and pain response. Each box represents a specific module and colours indicate normalized enrichment score (NES). The asterisk denoted in the box represents False Discovery Rate (FDR) values < 0.25. c. Expression of indicated genes after vaccination. Data is shown as fold change normalized to pre-vaccination.