Intrinsic and viral regulation of the innate immune receptor RIG-I

Retinoic acid inducible gene I (RIG-I) is a helicase and innate immune receptor that recognizes viral 5’ triphosphorylated (5’PPP), blunt-ended dsRNAs. RIG-I activation stimulates the production of Type I interferons and cytokines that establish a cellular antiviral state and trigger the adaptive immune system. Due to the potent immune response from RIG-I activation, it is imperative that RIG-I remains autoinhibited in the absence of foreign RNA; recognition of self-RNAs, which are cytoplasmically abundant, leads to autoimmunity. The mRNA 5’ cap, a common self-RNA motif, combines the presence of an inverted G-cap and 2’-O ribose methylation to evade RIG-I recognition. Recently, RNAs capped with noncanonical moieties like NAD⁺, rather than the canonical G-cap, have been identified; however, it is unknown whether these non-canonically capped RNAs would activate RIG-I. Viruses have mechanisms to inactivate and evade RIG-I signaling, including influenza B virus nonstructural protein 1 (NS1B), whose expression leads to downregulation of RIG-I signaling. However, the exact mechanism by which the RNA-binding NS1B protein downregulates RIG-I signaling was unknown. My thesis addressed the following three questions: What intrinsic mechanisms prevent RIG-I from binding self RNA? Does RIG-I recognize the non-canonical capped dsRNAs? Does the influenza virus NS1B have RNA specificity and does it compete with RIG-I for RNA binding? The signaling domains in RIG-I are connected to the helicase domain through a ~50 amino acids linker (CARD2-Hel1 linker or CHL), which is intrinsically disordered and negatively charged. Such regions often have regulatory functions; hence, I set out to understand the role of CHL using an array of biophysical, biochemical, and cellular assays with a panel of self and nonself-RNAs. My studies demonstrated that the CHL is responsible for three functions essential for RIG-I regulation. First, the CHL stabilizes the autoinhibitory CARD2:Hel2i interface known to enhance RIG-I’s RNA discriminatory ability by sequestering the signaling CARDs. Second, the CHL utilizes its negative charge to compete with incoming RNAs for the nonspecific helicase domain, ensuring C-terminal domain’s (CTD) fidelity for 5’PPP RNA. Third, the CHL destabilizes the RIG-I:RNA binding complex through an unidentified mechanism. This work was the first description of the CHL's role during intrinsic RIG-I regulation and thus is the first to recognize this linker as a crucial regulatory domain. While it is well-understood that RIG-I discriminates against the host mRNA cap, recently, a new class of RNA caps has been identified in various organisms, including humans. Metabolites, such as NAD⁺ and FAD, can be used as the initiating nucleotide during transcription. The resulting RNA has a capped structure independent of the normal mRNA capping enzymes. These caps have not been identified with the same critical 2’-O ribose methylation in the mRNA cap that contributes to RIG-I evasion. I demonstrated using biochemical and cell-based assays that RIG-I binds to these alternatively capped RNAs with similar affinities as the 5’ PPP RNAs. The capped RNAs also stimulated the RIG-I signaling pathway comparable to the 5’PPP RNA. Thus, my work has discovered a new class of immunogenic RNAs that can be functionalized and leveraged in RIG-I based therapeutics. Non-canonical capped RNAs are found in bacteria, fungi, and mitochondria, and it remains to be determined whether RIG-I recognizing these non-viral RNAs is important for host immune function. Influenza B virus, like all viruses, has multiple avenues to evade or down-regulate the host immune system. One such method is viral NS1 protein, which inhibits the RIG-I signaling pathway. NS1 of influenza B (NS1B) has a novel RNA binding domain in its C-terminal domain (CTD) which is not found in the closely related NS1 of influenza A. Using biophysical, biochemical, and cell-based assays and demonstrated that NS1B-CTD specifically recognizes blunt-ended, 5’ PPP dsRNA, RNA which potently activates RIG-I. The segmented influenza viral panhandle genomes also contain 5’PPP dsRNA ends. My work suggests that NS1 can specifically bind to these ends, obscuring viral genomic RNA from RIG-I recognition.Ph.D.Includes bibliographical reference