Zika virus infection in Collaborative Cross mice

Abstract

AbstractThe 2015-2016 emergence of Zika virus (ZIKV) in the Americas, and recognition that ZIKV infection during pregnancy can result in birth defects, revealed a need for small animal models to study ZIKV pathogenic mechanisms and evaluate candidate vaccines and antivirals. Mice would be an attractive system for such studies, but ZIKV replicates poorly in laboratory mice because it fails to antagonize murine STAT2 and STING. To address this, most ZIKV pathogenesis studies have used mice with impaired interferon signaling (e.g. Ifnar1−/− or treatment with IFNAR1-blocking antibodies). However, using mice with severe defects in innate antiviral signaling confounds studies of viral pathogenic mechanisms. Collaborative Cross (CC) mice have proven to be a valuable system for developing new mouse pathogenesis models for viral infections that are not well modeled in conventional laboratory mouse lines. To test whether CC mice could provide an immune-competent model for ZIKV pathogenesis, we infected CC lines with ZIKV and assessed weight loss, viremia, and production of neutralizing antibodies. We tested 21 CC lines (CC001, CC002, CC003, CC004, CC005, CC006, CC011, CC012, CC013, CC019, CC024, CC028, CC040, CC041, CC042, CC046, CC051, CC059, CC061, CC068, and CC072, 13 of which have non-functional alleles of the flavivirus restriction factor Oas1b) and 3 ZIKV strains (MR766, H/PF/2013, and a mouse-adapted variant of Dakar 41525). ZIKV infection did not induce weight loss compared to mock-infected controls and accordingly only low levels of viral RNA were detected in serum. Only a subset of mice developed neutralizing antibodies to ZIKV, likely due to overall low levels of infection and viremia. Our results are consistent with other studies demonstrating poor ZIKV infection in interferon-intact mice and suggest that the tested CC lines do not include polymorphic host genes that greatly increase susceptibility to ZIKV infection.