Generation and Characterization of Human-Mouse STING Chimeras That Allow DENV Replication in Mouse Cells

Our group was the first to describe direct antagonism of the cyclic GMP-AMP synthase (cGAS)/stimulator of interferon genes (STING) pathway by dengue virus (DENV) in human cells, and here, we report new findings on the characterization of the interaction between the DENV nonstructural protein 2B (NS2B)-NS3 (NS2B3) protease complex and STING. We demonstrate interactions between NS2B and the transmembrane domains of human STING and between NS3 and a portion of the cytoplasmic C-terminal domain of human STING. One significant obstacle we face today in the DENV field is the lack of small animal models available that can effectively recapitulate DENV pathogenesis in the early events of infection. The existing mouse models are either immunocompromised mice lacking interferon (IFN) receptors or "humanized" mice reconstituted with human stem cells. However, both approaches fail to capture important aspects of human pathogenesis because they lack critical innate immunity components or have deficiencies in immune cell development or maintenance. As an important step toward developing an immunocompetent mouse model for DENV, we have generated two chimeric human-mouse STING constructs that have promise in retaining both cleavability by NS2B3 and signaling capacity in the mouse. IMPORTANCE This article characterizes the interaction between human STING and DENV viral protease complex NS2B3 by constructing serial deletion mutants of STING. Our findings suggest that DENV nonstructural protein NS2B interacts with the transmembrane domains and NS3 with the C-terminal cyclic dinucleotide binding domain of human STING. Furthermore, as there exists no ideal immunocompetent murine model that can simultaneously support robust DENV replication and recapitulate the clinical manifestation of dengue disease observed in humans, we expressed and characterized two promising human-mouse chimeric STING constructs that can be used for developing a relevant transgenic mouse model to study dengue in the future. Both constructs can activate normal IFN responses in the overexpression system and be cleaved under infection conditions. We believe our findings offer a roadmap to the further development of a murine model that can greatly facilitate antiviral discoveries and vaccine research for DENV.