IPS-1 Is Essential for the Control of West Nile Virus Infection and Immunity

Suthar, Mehul S.; Ma, Daphne Y.; Thomas, Sunil; Lund, Jennifer M.; Zhang, Nu; Daffis, Stephane; Rudensky, Alexander Y.; Bevan, Michael J.; Clark, Edward A.; Kaja, Murali-Krishna; Diamond, Michael S.; Gale, Michael

Although infection of primary DCs, macrophages, and neuronal cells failed to induce type I IFN upon WNV infection, WNVinfected IPS-1 2/2 mice showed enhanced systemic type I IFN responses. This finding agrees with previous studies that indicate both IPS-1-dependent and -independent pathways contribute to the systemic type I IFN production in vivo [8, 9, 23, 25] . Most importantly, the enhanced tissue tropism and rapid viral entry into the CNS observed in the IPS-1 2/2 mice is not affected by the elevated systemic IFN responses. This suggests that local tissuespecific and intracellular responses triggered by RLR-dependent signaling are more essential for reducing viral burden and dissemination. One possible explanation is that a distinct set of RLR-responsive genes function to control virus replication at the site of infection. This could explain, in part, the elevated levels of virus replication, enhanced tissue tropism and cell-to-cell spread in mice or cells deficient in IRF3 or IRF-7, each of which are downstream transcription factors of RLR signaling [8, 9, 10] . Additionally, it is likely that pDCs, which are specialized dendritic cells for producing systemic type I IFN during a viral infection [46] , are likely contributing to the IFN responses observed during WNV infection. Studies by Silva et al. have shown that WNV triggers IFN induction in pDCs through a replication-independent manner [47] . Interestingly, within the DLN, we observed similar expansion of pDCs between wild type and IPS-1 2/2 infected mice, yet at the same timepoint (24 hours pi), elevated systemic type I IFN responses were observed in IPS-1 2/2 mice. This suggests two possibilities: 1) splenic pDCs or circulating pDCs are likely responding to the high levels virus in the serum from the IPS-1 2/2 infected mice to produce IFN at 24 hours pi and/or 2) various other cell types that express TLR3 and/or TLR7 are responding to WNV infection and contributing to systemic IFN responses. Taken together, these studies indicate that RLR signaling and the actions of IRF-3/7 are important in triggering IFN production and ISG expression to limit WNV replication and spread, and that TLR signaling may impart additional, RLRindependent defenses that regulate immunity against WNV infection. The production of and response to type-I IFN is a major linkage point between innate and adaptive immunity, as IFN-a and IFN-b sustain B cell activation and differentiation [48, 49, 50] , expand antigen-specific CD8+ T cells [51, 52] , CD4+ T cells [53] , and activation of NK cells [54] . One of the most intriguing aspects of this study was the global alteration of the immune response elicited in the IPS-1 2/2 mice, indicating that RLR signaling couples innate immunity with regulation of the adaptive immune response. Infection of IPS-1 2/2 mice exhibited increased IgM and IgG WNV-specific antibodies, enhanced WNV-specific CD8+T cell response, and increased expansion of neutrophils, NK cells and NK-T cells. One trivial explanation for these differences is that there is an increased antigen load in the absence of IPS-1 and, as a result, enhanced virus-specific (e.g. CD8+ T cells, IgG and IgM antibodies) and nonspecific (e.g. Neutrophils, NK cells) responses. However, there are several key findings from this study that argue against these responses simply being attributed to higher antigen load: (1) In the absence of IPS-1, the CD4 and CD8 T cells, which are protective against WNV infection [34, 3