Lipid-nanoparticle-encapsulated mRNA vaccines induce protective memory CD8 T cells against a lethal viral infection

Abstract

It is well established that memory CD8 T cells protect susceptible strains of mice from mousepox, a lethal viral disease caused by ectromelia virus (ECTV), the murine counterpart to human variola virus. While mRNA vaccines induce protective antibody (Ab) responses, it is unknown whether they also induce protective memory CD8 T cells. We now show that immunization with different doses of unmodified or N(1)-methylpseudouridine-modified mRNA (modified mRNA) in lipid nanoparticles (LNP) encoding the ECTV gene EVM158 induced similarly strong CD8 T cell responses to the epitope TSYKFESV, albeit unmodified mRNA-LNP had adverse effects at the inoculation site. A single immunization with 10 μg modified mRNA-LNP protected most susceptible mice from mousepox, and booster vaccination increased the memory CD8 T cell pool, providing full protection. Moreover, modified mRNA-LNP encoding TSYKFESV appended to green fluorescent protein (GFP) protected against wild-type ECTV infection while lymphocytic choriomeningitis virus glycoprotein (GP) modified mRNA-LNP protected against ECTV expressing GP epitopes. Thus, modified mRNA-LNP can be used to create protective CD8 T cell-based vaccines against viral infections.

Keywords: CD8 T cells; lipid nanoparticle; mRNA vaccine; modified mRNA; poxvirus; virus.

Graphical abstract

Figure 1

Nucleoside-modified mRNA-LNP vaccination elicits virus-specific CD8 T cell responses without skin pathology (A–E) B6 mice were immunized i.d. with 3, 10, or 30 μg of unmodified or modified mRNA-LNP encoding EVM158. TSYKFESV-specific CD8 T cells were determined by K^b-TSYKFESV dimer staining. (A) Image of hind back of mice vaccinated with unmodified (top) or modified (bottom) mRNA-LNP at 8 dpv. White arrows indicate areas of lesions. Ruler units are in centimeters. (B) Representative flow plot of gating strategy to identify TSYKFESV-specific CD8 T cells using DimerX complexes. (C) Frequency of TSYKFESV-specific of CD8 T cells in PBL. (D and E) Frequency and total number of TSYKFESV-specific CD8 T cells in (D) liver and (E) spleen at 8 dpv. (C–E) Groups were compared to each other by one-way ANOVA analysis with Tukey’s post-tests. All statistical differences indicated are compared to unvaccinated group unless otherwise designated; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 (C–E n = 6).

Figure 2

Modified mRNA-LNP induces greater MPEC formation and primarily effector memory CD8 T cells (A) Percentage of TSYKFESV-specific CD8 T cells in PBL over time following immunization of B6 mice with 10 μg mRNA-LNP. Statistical differences are compared to unvaccinated group unless otherwise designated. (B) Concatenated flow plot of KLRG-1 by CD127 staining of TSYKFESV-specific CD8 T cells at 8 and 15 dpv in the PBL of B6 mice following immunization with unmodified or modified mRNA-LNP. (C) Frequency of short-lived effector cells (SLEC; KLRG-1⁺CD127⁻) or memory precursor (MPEC; KLRG-1⁻CD127⁺) TSYKFESV-specific CD8 T cells in the PBL at 8 and 15 dpv of B6 mice immunized with unmodified or modified mRNA-LNP. (D and E) Frequency and total number of TSYKFESV-specific CD8 T cells in the (D) liver and (E) spleen at 28 dpv in B6 mice. (F) Concatenated flow plots of CD62L by CD127 staining on TSYKFESV-specific CD8 T cells at 28 dpv in B6 mice. (G) Frequency of T effector memory (T_EM; CD127⁺CD62L^-) or T central memory (T_CM; CD127⁺CD62L⁺) TSYKFESV-specific CD8 T cells at 28 dpv in B6 mice. (C and G) Groups were compared using t tests, and (A), (D), and (E) were compared using one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. (A, D, and E, n = 9; C and G, n = 10).

Figure 3

mRNA-LNP vaccination protects susceptible strains of mice against lethal ECTV challenge (A–C) B6.D2-D6 mice were immunized i.d. with 10 μg modified mRNA-LNP encoding EVM158. TSYKFESV-specific CD8 T cells were identified by K^b-TSYKFESV dimer staining. (A) Frequency of TSYKFESV-specific CD8 T cells in PBL following mRNA-LNP vaccination. (B) Survival curve of B6.D2-D6 mice following ECTV infection. (C) Virus titers of immunized B6.D2-D6 mice by plaque assay in the liver and spleen at 7 dpi. (D) Percentage of TSYKFESV-specific CD8 T cells in PBL of B6.D2-D6 mice following either i.d. or i.m. immunization with modified mRNA-LNP. (E) Survival curve for B6.D2-D6 mice immunized with mRNA-LNP i.d. or i.m. (F) Frequency of TSYKFESV-specific CD8 T cells in PBL following modified mRNA vaccination of TLR9-deficient mice. (G) Survival curve of immunized TLR9-deficient mice. (A, C, D, and F) Groups were compared by t test; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. (A and B, n = 8–9; C, n = 9–10; D and E, n = 7–8; F and G, n = 7).

Figure 4

Prime-boost mRNA-LNP immunization increases magnitude of CD8 T cell response and improves survival in TLR9-deficient mice, but sera from immunized mice are not protective (A–D) Previously immunized B6 mice were i.d. vaccinated with 10 μg modified mRNA-LNP 15 days following the first immunization. (A) Frequency of TSYKFESV-specific CD8 T cells in the PBL following prime-boost immunization. (B) Frequency and number of TSYKFESV-specific CD8 T cells in the PBL 28 days following a single or booster immunization. (C and D) Percentage and total number of TSYKFESV-specific CD8 T cells in the (C) liver and (D) spleen at 8 and 28 dpbv. Statistical differences are compared to naive mice unless otherwise designated. (E) Concatenated flow plots of CD62L and CD127 on TSYKFESV-specific CD8 T cells in the PBL of B6 mice following prime-boost immunization at 28 dpbv. (F) Percent T_EM or T central memory (T_CM) of TSYKFESV-specific CD8 T cells in the PBL of B6 mice at 28 dpbv following prime-boost immunization with modified mRNA-LNP. (G) Frequency of TSYKFESV-specific CD8 T cells following prime-boost immunization in Tlr9^−/− mice. (H) Survival curve in prime-boost-vaccinated Tlr9^−/− mice. (I) WT B6 mice received 200 μL of serum from unvaccinated, EVM158-modified mRNA-LNP-immunized, or previously ECTV-infected B6 mice 1 day prior to ECTV challenge. Viral titers of liver and spleens from ECTV-infected, serum-recipient mice at 5 dpi. Groups indicate source of sera. (A, B, F, and G) Groups were compared using t tests; data in (C), (D), and (I) were compared using one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗0.001 (A–D, n = 9; E and F, n = 9; G–I, n = 8–9).

Figure 5

CD8 T cell response induced by modified mRNA-LNP vaccination protects mice against ECTV infection (A–C) Tlr9^−/− mice were immunized with 10 μg modified mRNA-LNP encoding GFP-TSYKFESV. (A) Frequency of TSYKFESV-specific CD8 T cells in the PBL of Tlr9^−/− mice following immunization with GFP-TSYKFESV mRNA. (B) Survival curve of Tlr9^−/−mice immunized with GFP-TSYKFESV mRNA following ECTV infection. (C) Virus titers in liver and spleen at 6 dpi in unvaccinated and GFP-TSYKFESV-immunized Tlr9^−/− mice or unvaccinated B6 mice infected with ECTV. (D) Frequency of TSYKFESV or SIINFEKL-specific CD8 T cells in the PBL at 28 dpbv of Tlr9^−/− mice immunized with mRNA-LNP encoding either GFP-TSYKFESV or chicken OVA. (E) Survival curve of Tlr9^−/− mice immunized with GFP-TSYKFESV or OVA mRNA-LNP following ECTV infection. (F) B6 mice were immunized with mRNA-LNP encoding LCMV GP. Representative flow plot of IFN-γ⁺ CD8 T cells from the spleen following peptide restimulation for 5 h at 8 and 28 dpbv. (G) Survival curve of Tlr9^−/− mice immunized with LCMV GP-encoding mRNA and infected by a recombinant ECTV-MEE strain expressing LCMV epitopes KAVYNFATC and AVYNFATCGI. (A and D) Comparisons were done using t tests, whereas in (C), groups were compared using one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. (A and B, n = 9–10; C, n = 7–11; D and E, n = 8; G, n = 9–10).