An mRNA vaccine encoding five conserved Group A Streptococcus antigens

Abstract

A commercial vaccine to address the high global burden of Group A Streptococcus (GAS) disease is an urgent and unmet medical need. Messenger RNA (mRNA) lipid-nanoparticle (LNP) vaccines represent a largely untapped platform for targeting bacterial pathogens. Here, we evaluate the immunogenicity and preclinical efficacy of a multicomponent mRNA-LNP vaccine formulation based on the GAS vaccine, Combo#5. Combo#5 mRNA-LNP antigens confer protection from infection in mouse intraperitoneal and subcutaneous challenge models. Combo#5 mRNA-LNP vaccination generates significantly increased frequencies and numbers of effector type CD4+ and CD8 + T cells in the spleen, enhances T follicular helper cells, germinal center B cells and memory B cells in the spleen and draining lymph nodes, and boosts the production of antigen-specific antibodies. These findings demonstrate the potential of the mRNA-LNP platform for the development of vaccines against bacterial pathogens.

Fig. 1. Individual mRNA-LNP constructs encoding SCPA, SLO, SpyCEP, ADI and TF stimulate robust antibody responses.

Schematic illustrations of a soluble-expressing and b transmembrane-anchored (TM) mRNA constructs encoding SCPA, SLO, SpyCEP, ADI and TF antigens. The first and last amino acid positions and detoxifying mutations within antigen coding sequences are indicated. c–h Sera were collected from BALB/c mice (n = 5 per group) at 7 days after the primary immunization (day 7, blue), first booster (day 27, orange) or second booster (day 35, purple) with either NTFIX mRNA-LNP, PBS/alum, mRNA-LNP constructs expressing single GAS antigens as soluble or transmembrane (TM) proteins protein (2 µg mRNA), or single purified recombinant proteins formulated with alum (5 µg total protein/alum). Dot plots show (c) IgM, d total IgG, e IgG1, f IgG2a, g IgG2b and h IgG3 responses against the indicated antigens, as measured by multiplex Luminex assays, given as median fluorescence intensity (MFI). Each dot represents data from one mouse and the geometric mean for each group is indicated. The dotted lines indicate the median lower limit of quantitation (LLOQ) of all analytes. Source data are provided as a Source Data file.

Fig. 2. Protective efficacy of various mRNA-LNP formulations expressing Combo#5 antigens.

Kaplan-Meier survival curves of BALB/c (a, b) and C57BL/6 hPLG (c, d) mice immunized with M1/alum (red), PBS/alum (black), NTFIX mRNA-LNP (gray), Combo#5-soluble mRNA-LNP (orange) or Combo#5-TM mRNA-LNP (green). Mice were infected 2 weeks post-final-vaccination (day 42) either intraperitoneally (IP) with ~4 × 10⁷ CFU of M1T1 GAS strain 5448 (a, b) or subcutaneously (SC) with ~1 × 10⁸ CFU GAS 5448 (c, d). Data from a and c are from 3 (n = 30 mice/group) or 2 independent experiments (NTFIX, n = 20 mice/group). Data from b and d are from 7 (n = 70 mice/group) and 9 (n = 90 mice/group) independent experiments, respectively. Survival curves were compared using the log-rank Mantel-Cox test (a) M1/alum vs. PBS/alum ***P < 0.001, Combo#5-soluble vs. PBS/alum ***P < 0.001; b M1/alum vs. PBS/alum ***P < 0.001, Combo#5-TM vs. PBS/alum ***P < 0.001; c M1/alum vs. PBS/alum ***P < 0.001, Combo#5-soluble vs. M1/alum **P = 0.006; d M1/alum vs. PBS/alum ***P < 0.001, Combo#5-TM vs. PBS/alum ***P < 0.001). Source data are provided as a Source Data file.

Fig. 3. Combo#5-TM mRNA-LNP immunization regime boosts B cell responses.

a Flow cytometric contour plots showing the proportion of plasma cells (PCs; gated as live B220^loCD138⁺CD98⁺) in the spleen at day seven post-final vaccination. Frequency of total PCs within the total B cells cell gate are shown. b Frequency (left) and total number (right) of PCs in the spleen at day seven post-final vaccination. c Flow cytometric contour plots showing the proportion of germinal center (GC) B cells (live B220⁺CD138^-CD38^loCD95^hi) in the spleen at day seven post-final vaccination. d Frequency (left) and total number (right) of GC B cells in the spleen at day seven post-final vaccination. e Representative flow cytometric contour plots show the gating strategy employed to identify follicular helper T (Tfh) cells and non-Tfh cells in the spleen at day seven post-final vaccination. Geometric mean fluorescence intensity (gMFI) of Bcl-6 for each subset are shown. f Frequency (left) and total number (right) of Tfh cells in the spleen at day seven post-final vaccination. Tfh cells were identified as live CD45⁺CD3⁺TCRβ⁺CD4⁺PD1⁺CXCR5⁺ cells. g Representative flow cytometry contour plot showing immunoglobulin class-switched (IgM^-) and IgM⁺ memory B cells (MBC; defined as live B220^hi/loCD138^-CD38^hiCD95^loIgD^lo). h Total number of IgM⁺ (left) and class-switched (right) MBC in the spleen at day seven post-final vaccination. i Contour plot showing CD80 and PD-L2 expression switched memory B cells in the spleen at day seven post-final vaccination. j Total number of PD-L2⁺ single positive (PD-L2 SP), PD-L2⁺CD80⁺ double positive (DP), CD80⁺ single positive (CD80 SP) and PD-L2^-CD80^- double negative (DN) memory B cell subsets in the spleen at day seven post-final vaccination. Data in a–i represent findings from two independent experiments. Data in b–j are pooled from two independent experiments (n = 6 mice/group/experiment, equal numbers of males and females). Each dot represents one mouse and bars indicate the median. Statistical analyses were performed using a Kruskal-Wallis test followed by Dunn’s test for multiple comparisons. P-values for comparisons between the PBS/PBS and Combo#5-TM or M1/alum versus Combo#5-TM groups are indicated. Source data are provided as a Source Data file.

Fig. 4. Combo#5-TM mRNA-LNP immunization enhances effector CD4⁺ T cell responses.

a Representative flow cytometric pseudo-color plots showing activated CD4⁺ T cells (gated as live CD45⁺CD3⁺TCRβ⁺CD4⁺CD62L^lo lymphocytes) in the spleen expressing CD69, ICOS, KLRG1 and CXCR3 at day seven post-final vaccination. b Frequencies and c total number of activated CD4⁺ T cells in the spleen expressing CD69, ICOS, KLRG1 and CXCR3 at day seven post-final vaccination. d Flow cytometric histograms of T-bet, RORγt, GATA-3, TCF-1 and Ki67 expression in activated CD8⁺ T cells. Values for each histogram show geometric mean fluorescence intensity (gMFI) of T-bet, RORγt, GATA-3, TCF-1 and Ki67 within the T-bet⁺, RORγT⁺, GATA-3⁺, TCF-1⁺ and Ki67⁺ activated CD4⁺ T cells. e Representative flow cytometric pseudo-color plot showing effector CD4⁺ T cells (gated as live CD45⁺CD3⁺TCRβ⁺CD4⁺CD44^hi lymphocytes) producing IFN-γ alone (IFN-γ⁺ SP), TNF-α alone (TNF-α⁺ SP) and both IFN-γ and TNF-α (IFN-γ⁺TNF-α⁺) in the spleen at day seven post-final vaccination. f Total numbers of splenic CD4⁺ T cells producing IFN-γ alone (left panel), TNF-α alone (middle panel) and both IFN-γ and TNF-α (right panel). Data in a, d and e show representative plots from two independent experiments. Data in b, c and f were pooled from two independent experiments (n = 6 BALB/c mice/group/experiment). Each dot represents one mouse and bars indicate the median. Statistical analyses were performed using a Kruskal-Wallis test followed by Dunn’s test for multiple comparisons. P-values for comparisons between the PBS/PBS and Combo#5-TM or M1/alum versus Combo#5-TM groups are indicated. Source data are provided as a Source Data file.

Fig. 5. Combo#5-TM mRNA-LNP immunization regime enhances effector CD8⁺ T cell responses.

a Representative flow cytometric pseudo-color plots showing activated CD8⁺ T cells (gated as live CD45⁺CD3⁺TCRβ⁺CD8⁺CD62L^lo lymphocytes) in the spleen, expressing CD69, CX3CR1, KLRG1 and CXCR3 at day seven post-final vaccination. b Total activated CD8⁺ T cells expressing CD69, CX3CR1, KLRG1 and CXCR3. c Representative flow cytometric pseudo-color plot showing effector CD8⁺ T cells (gated as live CD45⁺CD3⁺TCRβ⁺CD8⁺CD44^hi lymphocytes) producing IFN-γ alone (IFN-γ⁺ SP), TNF-α alone (TNF-α⁺ SP) and both IFN-γ and TNF-α (IFN-γ⁺TNF-α⁺) in the spleen at day seven post-final vaccination. d Total activated splenic CD8⁺ T cells producing IFN-γ alone (left panel), TNF-α alone (middle panel) and both IFN-γ and TNF-α (right panel) following in vitro stimulation with PMA/ionomycin. Data in a and c represent findings from two independent experiments. Data in b and d were pooled from two independent experiments (n = 6 BALB/c mice/group/experiment). Each dot represents one mouse and bars indicate the median. Statistical analyses were performed using a Kruskal-Wallis test followed by Dunn’s test for multiple comparisons. P-values for comparisons between the PBS/PBS and Combo#5-TM or M1/alum versus Combo#5-TM groups are indicated. Source data are provided as a Source Data file.

Fig. 6. Antibodies from Combo#5-TM-immunized mice bind to the surface of live GAS and neutralize toxin activity.

a Pooled serum from PBS/alum (black), M1/alum (red), NTFIX mRNA-LNP (gray), and Combo#5-TM mRNA-LNP-immunized BALB/c mice (green) were incubated with live M1 GAS strain 5448, M3 GAS strain 89437, and M28 GAS strain PS001. Binding was detected by flow cytometry, and T(x) values for each group, located to the right of each histogram, were determined by the probability binning algorithm in FlowJo comparing GAS samples incubated with M1/alum immune sera to PBS/alum immune sera, and Combo#5-TM mRNA-LNP immune sera with NTFIX mRNA-LNP immune sera. A T(x) value of > 200 was considered significant (P < 0.01). b Opsonophagocytic killing of GAS M1 strain 5448 by differentiated HL-60 cells. The opsonic index was determined as the highest sera dilution that resulted in 50% killing. Mouse serum was from PBS/alum, M1/alum, NTFIX mRNA-LNP, and Combo#5-TM mRNA-LNP-immunized C57BL/6 mice. The percentage of killing was determined by normalizing sera responses to control wells containing active complement alone. Data from 3 independent experiments and mean +/- standard deviation are indicated. c Percent inhibition of sheep red blood cell hemolysis following 30 min incubation with recombinant SLO protein and pooled cholesterol oxidase-treated serum from PBS/alum, M1/alum, NTFIX mRNA-LNP, and Combo#5-TM mRNA-LNP-immunized BALB/c mice. Combo#5-TM was statistically significant (P < 0.0001) when compared to all other groups at dilutions 1:40, 1:80 and 1:160 by ordinary two-way ANOVA with Tukey’s multiple comparisons test. Data from three independent experiments and mean values +/−standard deviation are indicated. Source data are provided as a Source Data file.