Innate Immune Priming by cGAS as a Preparatory Countermeasure Against RNA Virus Infection

Michael T. Parker; Smita Gopinath; Corey E. Perez; Melissa M. Linehan; Jason M. Crawford; Akiko Iwasaki; Brett D. Lindenbach

Collectively, our results indicate that cGAS binds mtDNA in both infected and 251 uninfected cells, and that VSV infection does not induce the release of mtDNA into 252 the cytosol or increase cGAS-bound mtDNA. Additionally, viral and cellular 253 mRNA-specific cDNAs can be detected, but are of extremely low abundance, less 254 than one copy per 10 4 cells. Taken together, these results suggest that steady state 255 levels of cytosolic DNA, rather than virus-induced DNAs, may provide ligands for 256 cGAS-mediated restriction of RNA virus replication. 257 . CC-BY-NC-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/434027 doi: bioRxiv preprint cGAS primes smoldering baseline ISG expression. Based on the above 258 results, we hypothesized that cGAS may serve to program baseline levels of innate 259 immune activation rather than strictly in response to RNA virus infection. To address 260 this, we analyzed ISRE-driven luciferase expression in uninfected THP-1 cells To examine whether cGAS drives basal levels of innate immune activation in 279 vivo, we examined ISG expression in vaginal tissue from uninfected WT B6J mice or 280 in mice defective for several innate immunity pathways. As shown in Fig. 7 Although cGAS was previously reported to restrict RNA viruses (50), it has been 301 widely assumedthough unproventhat this restriction depends on cGAS's DNA 302 binding and cGAMP synthase activities. Here, we used genetic knockout and 303 transgenic replacement to determine that both DNA binding and cGAMP synthase 304 activities are essential for cGAS-mediated restriction of RNA viruses. One caveat to 305 this approach is that gene knockout can have far-reaching network-level effects on 306 transcription, which are just beginning to be unearthed (51). A second caveat is that 307 reconstituted cGAS was slightly overexpressed in THP-1 cells, which, at least for WT 308 cGAS, can induce ISG expression (50, 52) and may have exaggerated the 309 response. Nevertheless, our results in THP-1 cells were consistent with results 310 obtained from MEFs (Figure 3 ), which did not overexpress cGAS. Taken together, 311 these data establish that DNA binding and cGAMP synthase activities are required 312 for cGAS-mediated RNA virus restriction. 313 . CC-BY-NC-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/434027 doi: bioRxiv preprint Despite the essential role of cGAMP synthase activity and demonstrated 314 detection of cGAMP synthesized after DNA transfection, we were unable to detect 315 cGAMP production in response to VSV-GFP infection. Our results are consistent 316 with results recently reported by Franz et al., who were also unsuccessful in 317 detecting cGAMP production in . While Franz and colleagues 318 concluded that cGAMP is not produced in response to VSV infection, we also 319 considered the possibility that cGAMP levels may be below the limit of detection 320 and/or rapidly degraded. Whereas cGAMP synthesis is readily detected in response 321