Development of an engineered extracellular vesicles-based vaccine platform for combined delivery of mRNA and protein to induce functional immunity

Abstract

mRNA incorporated in lipid nanoparticles (LNPs) became a new class of vaccine modality for induction of immunity against COVID-19 and ushered in a new era in vaccine development. Here, we report a novel, easy-to-execute, and cost effective engineered extracellular vesicles (EVs)-based combined mRNA and protein vaccine platform (EV^X-M+P vaccine) and explore its utility in proof-of-concept immunity studies in the settings of cancer and infectious disease. As a first example, we engineered EVs, natural nanoparticle carriers shed by all cells, to contain ovalbumin mRNA and protein (EV^OvaM+P vaccine) to serve as cancer vaccine against ovalbumin-expressing melanoma tumors. EV^OvaM+P administration to mice with established melanoma tumors resulted in tumor regression associated with effective humoral and adaptive immune responses. As a second example, we generated engineered EVs that contain Spike (S) mRNA and protein to serve as a combined mRNA and protein vaccine (EV^SpikeM+P vaccine) against SARS-CoV-2 infection. EV^SpikeM+P vaccine administration in mice and baboons elicited robust production of neutralizing IgG antibodies against RBD (receptor binding domain) of S protein and S protein specific T cell responses. Our proof-of-concept study describes a new platform with an ability for rapid development of combination mRNA and protein vaccines employing EVs for deployment against cancer and other diseases.

Fig. 1.

EV^OvaM+P elicits antigen specific immune responses to prevent tumor growth. (a) NTA of wide type 293F EV and EV^OvaM+P (EVs carrying OVA mRNA and protein). TEM images of EVs (inset). Scale bar, 100 nm. (b-c) EVs production from cells quantified by NTA (b) and microBCA assay (c) (n = 3). (d) Flow cytometry analysis of EVs markers CD9, CD47, CD63, and CD81 (n = 3). (e) Quantification of OVA mRNA from EV^OvaM+P. Data are shown as 1/∆CT (n = 3). n.d., not detected. (f) Representative western blot of OVA protein and CD81 expression in EV^OvaM+P (n = 3). (g) Schematic illustration of vaccination and tumor challenge schedule with EV^OvaM+P in mice. C57BL/6J mice were intramuscularly injected with EVs, EV^OvaM+P, or recombinant OVA protein (Ova^P). (h) Quantification of anti-OVA IgG antibodies from vaccinated C57BL/6J mice by ELISA (n = 6 for EV vaccination, n = 8 for EV^OvaM+P, n = 7 for OVA^P). (i) Tumor volume and weight measurement of wide type (WT) and OVA tumor in EV^OvaM+P vaccinated mice (n=8 per group). (j) Tumor volume and weight measurement of WT and OVA tumor in EV vaccinated mice (n = 6 per group for tumor volume, n = 5 per group for EV tumor weight). (k) Tumor volume and weight measurement of WT and OVA tumor in OVA^P vaccinated mice (n = 7 per group). (l) Frequency of CD11b⁺ cells out of CD45⁺ cells from tumor tissue (n = 5 for EV vaccination, n = 7 for EV^OvaM+P, n = 7 for OVA^P). (m) Frequency of NK1.1⁺ cells out of CD45⁺ cells from tumor tissue (n = 5 for EV vaccination, n = 7 for EV^OvaM+P, n = 7 for OVA^P). (n) Representative images of CD4⁺ T cell and CD4⁺ T cell proliferation by TSA staining in WT tumor and OVA tumor tissue from EV^OvaM+P vaccinated mice. Scale bar, 100 µm. (o) Quantification of CD4⁺ T and proliferative CD4⁺ T cells (Ki67⁺) of images in (n). Data are presented as mean ± s.e.m. Significance was determined by unpaired t-test in b-c, the right panel of i-k, l-m, o, two-way ANOVA with Tukey’s multiple comparisons test in h, mixed-effects analysis with Bonferroni’s multiple comparisons test in the left panel of i, and two-way ANOVA with Bonferroni’s multiple comparisons test in the left panel of j-k. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, ns: not significant.

Fig. 2.

Generation and validation of EV^SpikeM+P. (a) NTA of wild type 293F EV, EV-SC2S (EVs carrying Spike mRNA and protein), and EV^SpikeM+P (EVs carrying Spike-2P mRNA and protein). TEM images of EVs (inset). Scale bar, 100 nm. (b-c) EVs production from cells quantified by NTA (b) and microBCA assay (c) (n = 6). (d) Flow cytometry analysis of EVs markers CD9, CD47, CD63, and CD81 (n = 3). (e) Quantification of S mRNA from EV^SpikeM+P. Data are shown as 1/ΔC_T (n = 3). n.d., not detected. (f) Representative western blot of S protein, syntenin-1, and CD81 expression in EV-SC2S and EV^SpikeM+P (n = 3). Data are presented as mean ± s.e.m. Significance was determined by one-way ANOVA with Dunnett’s multiple comparisons test in b-c. ns: not significant.

Fig. 3.

EV^SpikeM+P vaccination elicits humoral and cellular responses in mice. (a) Schematic illustration of vaccination schedule with EV^SpikeM+P in mice. Balb/c mice were intramuscularly injected with EVs, EV-SC2S and EV^SpikeM+P. (b) Body weight change of Balb/c mice during vaccination expressed as percentage of starting weight (n = 6 per group). (c) Quantification of anti-S-RBD IgG antibodies from vaccinated Balb/c mice by ELISA (n = 6 per group). (d) Schematic illustration of the pseudovirus neutralization assay. (e) Neutralizing capacity of the antibodies from vaccinated mice evaluated with pseudovirus neutralization assay (n = 6 per group). Data are expressed as relative luminescence units (RLU). SFM + virus, serum-free media with pseudovirus; SFM, serum-free media alone. (f-g) S protein-specific CD4⁺ T cells in Balb/c mice splenocytes showing T cell activation (f) and Th1 response (g) by intracellular cytokine staining (n = 6 for PBS, n = 5 for EV vaccination, n = 5 for EV^SpikeM+P). Splenocytes were re-stimulated with S peptide pool. (h) Schematic illustration of vaccination schedule with EV^SpikeM+P in mice. C57BL/6J mice were intramuscularly injected with PBS, EV, EV^SpikeM+P. (i) Body weight change of C57BL/6J mice during vaccination expressed as percentage of starting weight (n = 10 per group). (j) Quantification of anti-S-RBD IgG antibodies from vaccinated C57BL/6J mice by ELISA (n = 10 per group). Data are presented as mean ± s.e.m. Significance was determined by mixed-effects analysis with Tukey’s multiple comparisons test in c and j, two-way ANOVA with Tukey’s multiple comparisons test in e, unpaired t-test in f, and Mann-Whitney test in g. * P < 0.05, ** P < 0.01, *** P < 0.001, ns: not significant.

Fig. 4.

EV^SpikeM+P vaccinated baboons develop antibodies and T cell responses against S protein. (a) Schematic illustration of vaccination schedule with EV^SpikeM+P of 3 dosages in baboons. (b) Quantification of anti-S-RBD IgG antibodies from vaccinated baboons by ELISA (n = 3 per group). (c) Titration of anti-S-RBD IgG antibodies at different time points plotted as log AUC (area under curve, n = 3 per group). (d-f) Quantification of cytokines (IL-2, IFNy, and TNFa) releasing spots from S protein-specific T cells in circulation re-stimulated with S peptide pool (d) or recombinant S protein (e-f) by ELISpot (n = 3 per group). Data are shown as SFCs per 2 × 10⁶ cells. (g-i) Frequency of S protein-specific CD4⁺ T cells in circulation producing single (IL-2 (g), IFNy (h)) and two of Th1 cytokines (IL-2, IFNy, and TNFa, (i)) determined by intracellular cytokine staining (n = 3 per group). PBMCs were re-stimulated with S peptide pool. (j-k) Frequency of S protein-specific CD4⁺ T cells in circulation producing single (TNFa (j)) or three (k) of Th1 cytokines (IL-2, IFNy, and TNFa) determined by intracellular cytokine staining (n = 3 per group). PBMCs were re-stimulated with S recombinant protein. Data are presented as mean ± s.e.m. was determined by two-way ANOVA with Tukey’s multiple comparisons test in b-c, and by unpaired t-test in d-k. * P < 0.05, ** P < 0.01, **** P < 0.0001, ns: not significant, or exact p-values reported.