An mRNA-LNP-based Lassa virus vaccine induces protective immunity in mice

The mammarenavirus Lassa virus (LASV) causes the life-threatening hemorrhagic fever disease, Lassa fever. The lack of licensed medical countermeasures against LASV underscores the urgent need for the development of novel LASV vaccines, which has been hampered by the requirement for a biosafety level 4 facility to handle live LASV. Here, we investigated the efficacy of mRNA-lipid nanoparticle (mRNA-LNP)-based vaccines expressing the LASV glycoprotein precursor (LASgpc) or the nucleoprotein (LCMnp) of the prototypic mammarenavirus, lymphocytic choriomeningitis virus (LCMV), in mice using recombinant (r) LCMV expressing a modified LASgpc and wild-type rLCMV. Two doses of LASgpc- or LCMnp-mRNA-LNP administered intravenously or intramuscularly protected mice from a lethal challenge with rLCMVs. Negligible levels of LASgpc-specific antibodies were induced in mRNA-LNP-immunized mice, but robust LASgpc- and LCMnp-specific CD8+ T cell responses were detected. Our findings and surrogate mouse models of LASV infection provide a critical foundation for the rapid development of mRNA-LNP-based LASV vaccines.

underscores the urgent need for the development of novel LASV vaccines, which has 23 been hampered by the requirement for a biosafety level 4 facility to handle live LASV. 24 Here, we investigated the efficacy of mRNA-lipid nanoparticle (mRNA-LNP)-based 25 vaccines expressing the LASV glycoprotein precursor (LASgpc) or the nucleoprotein 26 (LCMnp) of the prototypic mammarenavirus, lymphocytic choriomeningitis virus (LCMV), 27 in mice using recombinant (r) LCMV expressing a modified LASgpc and wild-type rLCMV.

28
Two doses of LASgpc-or LCMnp-mRNA-LNP administered intravenously or 29 intramuscularly protected mice from a lethal challenge with rLCMVs. Negligible levels of 30 LASgpc-specific antibodies were induced in mRNA-LNP-immunized mice, but robust 31 LASgpc-and LCMnp-specific CD8 + T cell responses were detected. Our findings and 32 Introduction wild-type (WT) LASgpc (rLCMV/LASgpc), was generated 26 . LCMV, thought to be a 82 neglected human pathogen of clinical significance, is a particular threat to 83 immunocompromised or pregnant individuals 27,28,29,30,31 . Although rLCMV/LASgpc can 84 be used to investigate LASgpc functions in the context of the natural infection of cultured 85 cells without the need for a BSL-4 facility, the virus is rapidly cleared from C57BL/6 mice  Intriguingly, studies involving humans and animals indicated that COVID-19 mRNA-LNP 99 vaccines elicited neutralizing antibodies as well as spike protein-specific cellular immunity, 100 suggesting that mRNA-LNP-based vaccines are a feasible LASV vaccine modality 36 . In 101 the present study, we investigated the potential for an mRNA-LNP-based vaccine platform 102 to induce protective immunity in mice, using LASgpc and LCMV NP (LCMnp) as vaccine 103 antigens and rLCMV/LASgpc, containing K465V and G467K mutations 104 (rLCMV/LASgpc 2m ), and WT rLCMV, for challenge experiments.  (Fig. 1a). The intracranial inoculation of WT LCMV in 127 mice caused fatal lymphocytic choriomeningitis and the mice died within 8 days of 128 infection 41 . Next, we investigated whether intracranial (i.c.) inoculation of 129 rLCMV/LASgpc 2m to mice also caused a fatal infection. C57BL/6 mice inoculated i.c. with 130 10 3 FFU of rLCMV/LASgpc 2m all succumbed to the infection albeit with slightly extended 131 survival periods compared with those commonly observed in mice i.c. inoculated with WT 132 LCMV ( Supplementary Fig. 2b).

137
The levels of LASgpc and LCMnp antibodies in plasma 14 days post-second 138 immunization were examined by ELISAs (Fig. 1c). Consistent with previous reports of 139 virus vector-based LASV vaccine candidates 42, 43 , we observed negligible levels of 140 LASgpc-specific antibodies (Fig. 1c); however, we confirmed that our ELISA could detect 141 commercially available LASV GP2 antibodies in a dose-dependent manner 142 ( Supplementary Fig. 3). Conversely, two doses of LCMnp-mRNA-LNP induced a robust 143 LCMnp-specific antibody response in mice (Fig. 1c). To examine whether two doses of 144 LASgpc-mRNA-LNP or LCMnp-mRNA-LNP conferred protection in mice against lethal 145 virus exposure, mRNA-LNP-or mock-immunized C57BL/6 mice were inoculated i.c. with 146 a lethal dose of rLCMV/LASgpc 2m 28 days after the second immunization. As expected, 147 all mock immunized mice succumbed to rLCMV/LASgpc 2m infection within 12 days post-148 virus exposure. By contrast, all mice immunized with LASgpc-mRNA-LNP or LCMnp-149 mRNA-LNP survived without developing overt clinical signs of disease (Fig. 1d). Low 150 levels of LASgpc-specific antibodies were found in mice immunized with LASgpc-mRNA-151 LNP, and LCMnp is a cytosolic protein, indicating virus antigen-specific cytotoxic T cell 152 responses might have played a critical role in protection against infection. To assess this, 153 we examined an LCMV-specific CD8 + T cell response using a well-defined H-2d-restricted 154 LCMnp T cell epitope peptide (NP396). Erythrocyte-free splenocytes obtained from 155 C57BL/6 mice immunized twice with LCMnp-mRNA-LNP-or mock immunized were 156 cultured in the presence of NP396 and intracellular cytokine expression levels were 157 examined by flow cytometry (Fig. 1e). Significant increases in IFN-g-or TNF-a-expressing 158 CD8 + T cells from LCMnp-mRNA-LNP-immunized mice were detected, and most TNF-a-159 positive (TNF-a + ) cells were also IFN-g-positive (IFN-g + ), indicating the polyfunctional 160 property of antiviral CD8 + T cells (Fig. 1f). with LASgpc-mRNA-LNP significantly reduced virus titers in the plasma to levels close to 175 or below the lower limit of detection (10 2 FFU/mL), whereas high virus titers were detected 176 in the plasma from mock immunized mice (Fig. 2c).  immunized mice to below the minimum detection levels (Fig. 3g).

209
The absence of a strong LASgpc-specific antibody response in CBA mice 210 immunized with LASgpc-mRNA-LNP suggested that a LASgpc-specific CD8 + T cell 211 response was involved in the protection. To assess this, we used a LASgpc peptide 212 cocktail to stimulate erythrocyte-free splenocytes collected from CBA mice immunized 213 twice with LASgpc-mRNA-LNP or mock immunized mice, 28 days post-second 214 immunization, and intracellular cytokine expression levels were examined by flow 215 cytometry (Fig. 3h). We confirmed a significant increase in CD8 + T cell numbers that 216 produced IFN-g in response to stimulation with the LASgpc peptide cocktail following 217 LASgpc-mRNA-LNP immunization (Fig. 3i).

219
LCMnp-mRNA-LNP confers protection against lethality in a WT rLCMV 220 hemorrhagic fever mouse model 221 We found that the mRNA-LNP-based vaccine elicited protective immunity against 222 rLCMV/LASgpc 2m exposure using two antigens (GPC and NP), two immunization routes 223 (i.v. and i.m.), two virus inoculation routes (i.c. and i.v.), and two mouse strains (C57BL/6 224 and CBA), strongly suggesting the feasibility of an mRNA-LNP-based vaccine as a       The two different rLCMVs used in this study were generated by reverse genetics as    days after the second immunization, splenic CD3 + CD8 + T cells that specifically responded 817 to the NP396 peptide were examined for cytokine production by flow cytometry (f). The 818 presented data are the mean ± SD. **p < 0.01, *p < 0.05. week-old CBA mice (n = 5 per group) were immunized i.m. twice with LASgpc-mRNA-840 LNP or mock immunized, 21 days apart. At 28 days after the second immunization, 841 splenic CD3 + CD8 + T cells that specifically responded to the LASgpc peptide cocktail were 842 examined for cytokine production by flow cytometry (i). **p < 0.01.