Conserved moonlighting protein pyruvate dehydrogenase induces robust protection against Staphylococcus aureus infection

Abstract

Developing an effective Staphylococcus aureus (S. aureus) vaccine has been a challenging endeavor, as demonstrated by numerous failed clinical trials over the years. In this study, we formulated a vaccine containing a highly conserved moonlighting protein, the pyruvate dehydrogenase complex E2 subunit (PDHC), and showed that it induced strong protective immunity against epidemiologically relevant staphylococcal strains in various murine disease models. While antibody responses contributed to bacterial control, they were not essential for protective immunity in the bloodstream infection model. Conversely, vaccine-induced systemic immunity relied on γδ T cells. It has been suggested that prior S. aureus exposure may contribute to the reduction of vaccine efficacy. However, PDHC-induced protective immunity still facilitated bacterial clearance in mice previously exposed to S. aureus. Collectively, our findings indicate that PDHC is a promising serotype-independent vaccine candidate effective against both methicillin-sensitive and methicillin-resistant S. aureus isolates.

Keywords: S. aureus vaccine; moonlighting protein; γδ T cell.

Fig. 1.

PDHC is conserved among S. aureus strains and induces robust protective immunity against epidemiologically relevant staphylococcal strains in multiple murine disease models. (A) Phylogenetic tree of pyruvate dehydrogenase subunit E2 of species in bacteria and eukaryote. The corresponding percentage of sequence identity to PDHC of S. aureus was illustrated in the right panel. (B) Sequence identity of PDHC between different S. aureus strains collected since 2013 around the world. The numbers above the bar indicated the number of strains in continent. (C) SDS-PAGE of recombinant PDHC purified from E. coli expression system. (D) Schematic diagram of vaccination, ELISA, and flow cytometry. (E) Antibodies titer specific for PDHC was measured by ELISA (n = 6). (F) The splenocytes were induced by PDHC in vitro and subjected to flow cytometry analysis for IFN-γ and IL-17 responses (n = 4). (G) Schematic diagram of vaccination, bacterial challenge (intravenous), and pathological studies in C57BL/6 J mice. (H) Survival curves of PDHC-immunized and unvaccinated mice in the bloodstream infection model (n = 10). (I) Bacterial load in blood, kidney, lung, and spleen tissue of PDHC-immunized and unvaccinated mice after 3 d.p.i. (n = 10). (J) Representative images of the H&E-stained kidney tissues of PDHC-vaccinated and unvaccinated C57BL/6 J WT mice after 3 d.p.i. (K) Schematic diagram of vaccination, bacterial challenge (intraperitoneal), and pathological studies in C57BL/6 J mice. (L) Survival curves of PDHC-immunized and unvaccinated mice in peritonitis model (n = 10). (M) Schematic diagram of vaccination, bacterial challenge (subcutaneous), and pathological studies in C57BL/6 J mice. (N) Measurement of skin lesions in PDHC-immunized and unvaccinated mice after S. aureus infection (n = 6). Data are representative of two independent experiments (E, F, H, I, L, and N). Data are presented as mean ± SEM. Statistical significance was calculated using the one-way ANOVA test or unpaired t test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). Survival data were analyzed by the log-rank (Mantel-Cox) test. Fig. 1 D, G, K, and M were created with BioRender.com.

Fig. 2.

Humoral immunity is not required for PDHC-induced protection against S. aureus. (A) Schematic diagram of vaccination, bacterial challenge, and pathological studies in bloodstream infection model. (B) Survival curves of PDHC-immunized and unvaccinated WT and Rag1^−/− mice (n = 10). The statistical significance was indicated in the inlet box. (C) Bacterial load in kidney, spleen, liver, and lung tissue of PDHC-immunized and unvaccinated WT and Rag1^−/− mice after 3 d.p.i. (n = 10). (D) Antibodies titers specific for PDHC in blood samples of PDHC-immunized WT and μMt mice were measured by ELISA (n = 5). (E) Survival curves of PDHC-immunized and unvaccinated WT and μMt mice (n = 10). The statistical significance was indicated in the inlet box. (F) Bacterial load in kidney, spleen, liver, and lung tissue of PDHC-immunized and unvaccinated WT and μMt mice after 3 d.p.i. (n = 10). (G) Schematic diagram of vaccination, bacterial challenge, and bulk RNA-seq of spleen and kidney samples in the bloodstream infection model. (H) Scatter plot showing gene ontology where significantly differentially expressed genes (μMt PDHC vs μMt unvacc.) in spleen tissue are enriched. Scatter color indicated the adjusted P value from gene ontology enrichment test. Scatter size indicated the proportion of genes demonstrating significant differential expression in the given gene ontology. (I) Volcano plots of gene expression changes in spleen tissue between PDHC-vaccinated and unvaccinated μMt mice. Representative genes involved in cellular responses that were significantly up-regulated in PDHC-immunized μMt mice were highlighted. (J) Dynamic comparison of proinflammatory chemokine levels in spleen, liver, and kidney tissue of PDHC-immunized WT or μMt mice with unvaccinated counterparts after S. aureus bloodstream infection. Density plot above the scatter plot indicated the number of chemokines across the expression level fold changes in WT and μMt mice. Significantly up-regulated proinflammatory chemokines were highlighted. P-values were obtained by two-way ANOVA test. (K) Survival curve of WT mice administered with PDHC-specific IgG and control IgG after intravenous injection of S. aureus (n = 10). (L) Percentage of MHCII+ inflammatory macrophages in macrophage population in liver and kidney tissue of naïve WT mice administered with PDHC-specific IgG and control IgG collected at 72H after intravenous injection of S. aureus (n = 10). (M) In vitro assessment of anti-PDHC IgG function. Murine bone marrow–derived macrophages (BMDM) were incubated with cell-free supernatant of WT S. aureus culture in the presence of anti-PDHC IgG for 24 h. IL-6 levels secreted by BMDM were measured by ELISA. C29 is a TLR2 inhibitor. (N) Schematic diagram of vaccination, neutrophil/macrophage depletion, bacterial challenge, and pathological studies in bloodstream infection model. (O) Survival curves of PDHC-immunized WT mice injected with Ly6G monoclonal antibody and isotype antibody (n = 10). (P) Survival curves of PDHC-immunized WT mice administered with Clodronate liposome and control liposome (n = 10). Data are representative of two independent experiments (B–F, J–M, P, and Q). Data are presented as mean ± SEM. Statistical significance was calculated using the one-way ANOVA test or unpaired t test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). Survival data were analyzed by the log-rank (Mantel-Cox) test. Fig. 2 A, G and N were created with BioRender.com.

Fig. 3.

γδ T cells facilitate the bacterial clearance via neutrophil and macrophage activation in PDHC-vaccinated mice. (A) Schematic diagram of vaccination, CD4/8 + T cell depletion, bacterial challenge, and pathological studies in bloodstream infection model. (B) Survival curves of PDHC-immunized WT mice injected with aCD4/8 monoclonal antibodies and PBS control (n = 10). The statistical significance was indicated in the inlet box. (C) Bacterial load in kidney, spleen, liver, and lung tissue of PDHC-immunized WT mice with CD4/8⁺ T cell depletion after 3 d.p.i. (n = 10). (D) Schematic diagram of vaccination, bacterial challenge, and pathological studies of WT and γδ T^−/− mice in the bloodstream infection model. (E) Survival curves of PDHC-immunized and unvaccinated WT and γδ T^−/− mice (n = 10). The statistical significance was indicated in the inlet box. (F) Bacterial load in kidney, spleen, liver, and lung tissue of PDHC-immunized and unvaccinated WT and γδ T^−/− mice after 3 d.p.i. (n = 10). (G) Schematic diagram of vaccination, bacterial challenge, and bulk RNA-seq of spleen and kidney samples of PDHC-immunized WT and γδ T^−/− in bloodstream infection model. (H) Schematic diagram of vaccination, bacterial challenge, and flow cytometry of spleen, liver, and kidney samples of PDHC-immunized WT and γδ T^−/− mice in the bloodstream infection model. (I) Scatter plot showing gene ontology where significantly differentially expressed genes in spleen tissue (γδ T^−/− PDHC vs WT PDHC) are enriched. Scatter color indicated the adjusted P value from gene ontology enrichment test. Scatter size indicated the proportion of genes demonstrating significant differential expression in the given gene ontology. (J) Mean expression of myeloperoxidase (MPO) in neutrophils in spleen, liver, and kidney samples of PDHC-immunized and unvaccinated WT and γδ T^−/− mice after 24 or 72 h.p.i. (n = 6). (K) Quantification of MHCII+ inflammatory macrophages in spleen, liver, and kidney samples of PDHC-immunized and unvaccinated WT and γδ T^−/− mice after 24 or 72 h.p.i. (n = 6). (L) Percentage of CD3⁺CD44⁺Granzyme B⁺ γδ T cells in spleen and liver tissue from vaccinated or unvaccinated WT mice after S. aureus infection (n = 6). (M) Representative dot plots showing CD3⁺CD44⁺Granzyme B⁺ γδ T cells in spleen and liver tissue from vaccinated or unvaccinated WT mice after 24 h.p.i. (N) Percentage of γδ T cell subsets in spleen and liver tissue from vaccinated or unvaccinated WT mice after 72 h.p.i. (n = 6). (O) Representative breakdown of γδ T cell subsets in spleen and liver tissue from vaccinated or unvaccinated WT mice after 72 h.p.i. Data are representative of two independent experiments (B, C, E, F, and J–O). Data are presented as mean ± SEM. Statistical significance was calculated using the one-way ANOVA test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). Survival data were analyzed by the log-rank (Mantel-Cox) test. Fig. 5 A, D, G, and I were created with BioRender.com.

Fig. 4.

PDHC self-assembles into a nanoparticulate scaffold and renders protection to mice with previous multiple exposure to S. aureus. (A) Schematic diagram of vaccination, bacterial challenge, and pathological studies of WT mice with prior S. aureus intraperitoneal injections in bloodstream infection model. (B) Antibodies titers specific for PDHC in blood samples of PDHC-immunized WT mice were measured by ELISA (n = 6). Blood samples were taken 7 d after each dose of PDHC vaccination (D7 = 7 d after the first dose of PDHC; D21 = 7 d after the second dose of PDHC; D35 = 7 d after the final dose of PDHC). (C) Survival curves of PDHC-immunized and unvaccinated WT mice with or without prior S. aureus exposure (n = 10). The statistical significance was indicated in the inlet box (D) Bacterial load in kidney, spleen, liver, and lung tissue of PDHC-immunized and unvaccinated WT mice with or without prior S. aureus exposure after 3 d.p.i. (n = 10). (E) Transmission electron micrograph of PDHC stained with 2% uranyl acetate. (F) Antibodies titers specific for PDHC in blood samples of PDHC-immunized WT mice adjuvanted with Alum or without Alum were measured by ELISA (n = 6). Blood samples were taken 7 d after each dose of PDHC vaccination. (G) Survival curves of PDHC-immunized WT mice with or without Alum (n = 10). (H) Schematic diagram of vaccination, ELISA, bacterial challenge, and skin pathological studies of New Zealand White (NZW) rabbits. (I) Antibodies titers specific for PDHC in blood samples of PDHC-immunized NZW rabbits collected on the 7th day after the final dose of PDHC vaccination (n = 3). (J) Measurement of skin lesions in PDHC-immunized and unvaccinated NZW rabbits after S. aureus infection (n = 3). Data are representative of two independent experiments (B–D, F, and G). Data shown are from one experiment (I and J). Data are presented as mean ± SEM. Statistical significance was calculated using the one-way ANOVA test or unpaired t test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). Survival data were analyzed by the log-rank (Mantel-Cox) test. Fig. 4 A and H were created with BioRender.com.