Lassa virus (LASV), a highly prevalent mammalian arenavirus endemic in West Africa, can cause Lassa fever (LF), which is responsible for thousands of deaths annually. LASV is transmitted to humans from naturally infected rodents. At present, there is not an effective vaccine nor treatment. The genetic diversity of LASV is the greatest challenge for vaccine development. The reassortant ML29 carrying the L segment from the nonpathogenic Mopeia virus (MOPV) and the S segment from LASV is a vaccine candidate under current development. ML29 demonstrated complete protection in validated animal models against a Nigerian strain from clade II, which was responsible for the worst outbreak on record in 2018. This study demonstrated that ML29 was more attenuated than MOPV in STAT1^-/- mice, a small animal model of human LF and its sequelae. ML29 infection of these mice resulted in more than a thousand-fold reduction in viremia and viral load in tissues and strong LASV-specific adaptive T cell responses compared to MOPV-infected mice. Persistent infection of Vero cells with ML29 resulted in generation of interfering particles (IPs), which strongly interfered with the replication of LASV, MOPV and LCMV, the prototype of the Arenaviridae. ML29 IPs induced potent cell-mediated immunity and were fully attenuated in STAT1^-/- mice. Formulation of ML29 with IPs will improve the breadth of the host’s immune responses and further contribute to development of a pan-LASV vaccine with full coverage meeting the WHO requirements.

The safety and genetic stability of V4020, a novel Venezuelan Equine Encephalitis Virus (VEEV) vaccine based on the investigational VEEV TC-83 strain, was evaluated in mice. V4020 was generated from infectious DNA, contains a stabilizing mutation in the E2-120 glycoprotein, and includes rearrangement of structural genes. After intracranial inoculation (IC), replication of V4020 was more attenuated than TC-83, as documented by low clinical scores, inflammation, viral load in brain, and earlier viral clearance. During the first 9 days post-inoculation (DPI), genes involved in inflammation, cytokine signaling, adaptive immune responses, and apoptosis were upregulated in both groups. However, the magnitude of upregulation was greater in TC-83 than V4020 mice, and this pattern persisted till 13 DPI, while V4020 gene expression profiles declined to mock-infected levels. In addition, genetic markers of macrophages, DCs, and microglia were strongly upregulated in TC-83 mice. During five serial passages in the brain, less severe clinical manifestations and a lower viral load were observed in V4020 mice and all animals survived. In contrast, 13.3% of mice met euthanasia criteria during the passages in TC-83 group. At 2 DPI, RNA-Seq analysis of brain tissues revealed that V4020 mice had lower rates of mutations throughout five passages. A higher synonymous mutation ratio was observed in the nsP4 (RdRP) gene of TC-83 compared to V4020 mice. At 2 DPI, both viruses induced different expression profiles of host genes involved in neuro-regeneration. Taken together, these results provide evidence for the improved safety and genetic stability of the experimental V4020 VEEV vaccine in a murine model.

Lymphocytic choriomeningitis virus (LCMV) and Lassa virus (LASV) share many genetic and biological features including subtle differences between pathogenic and apathogenic strains. Despite remarkable genetic similarity, the viscerotropic WE strain of LCMV causes a fatal LASV fever-like hepatitis in non-human primates (NHPs) while the mouse-adapted Armstrong (ARM) strain of LCMV is deeply attenuated in NHPs and can vaccinate against LCMV-WE challenge. Here, we demonstrate that internalization of WE is more sensitive to the depletion of membrane cholesterol than ARM infection while ARM infection is more reliant on endosomal acidification. LCMV-ARM induces robust NF-κB and interferon response factor (IRF) activation while LCMV-WE seems to avoid early innate sensing and failed to induce strong NF-κB and IRF responses in dual-reporter monocyte and epithelial cells. Toll-like receptor 2 (TLR-2) signaling appears to play a critical role in NF-κB activation and the silencing of TLR-2 shuts down IL-6 production in ARM but not in WE-infected cells. Pathogenic LCMV-WE infection is poorly recognized in early endosomes and failed to induce TLR-2/Mal-dependent pro-inflammatory cytokines. Following infection, Interleukin-1 receptor-associated kinase 1 (IRAK-1) expression is diminished in LCMV-ARM- but not LCMV-WE-infected cells, which indicates it is likely involved in the LCMV-ARM NF-κB activation. By confocal microscopy, ARM and WE strains have similar intracellular trafficking although LCMV-ARM infection appears to coincide with greater co-localization of early endosome marker EEA1 with TLR-2. Both strains co-localize with Rab-7, a late endosome marker, but the interaction with LCMV-WE seems to be more prolonged. These findings suggest that LCMV-ARM’s intracellular trafficking pathway may facilitate interaction with innate immune sensors, which promotes the induction of effective innate and adaptive immune responses.