Identification and characterization of a novel transcript isoform of STING that suppresses innate immune response

Abstract

DNA derived from microbial pathogens constitutes a danger signal in the cytosol to be detected by DNA sensors, such as cyclic GMP-AMP synthase (cGAS), DDX41, IFI16 and Mre11. Following the detection of cytosolic DNA, cGAS can produce endogenous cyclic dinucleotide (CDN) 2’3’-cGAMP, which directly binds to stimulator of IFN genes (STING), also known as TMEM173, MPYS, MITA or ERIS, to induce the production of type I interferons (IFNs) and pro-inflammatory cytokines through IRF3 and NF-κB pathways. STING also functions as a direct sensor of bacterial CDNs such as c-di-GMP and c-di-AMP. DDX41, IFI16, and Mre11 also associate with STING to induce type I IFN production. STING therefore plays a central role in DNA-induced innate immune signaling pathways. However, innate immunity is a double-edged sword. To avoid excessive innate immune activation that may result in inflammation and tissue damage, STING activity should be tightly controlled. Although several regulators of STING have been reported, more studies are still necessary in order to get a comprehensive understanding of the regulatory mechanisms of STING activity. In this study, I identified and characterized a novel transcript isoform of STING designated STINGβ, which functions as a dominant inhibitor of the original STINGα isoform. STINGβ transcript was found to be widely expressed in different cell lines and human tissues. Importantly, the expression of STINGβ was inversely correlated with IFNβ induction in cells infected with Sendai virus (SeV), herpes simplex virus 1 (HSV1) or vesicular stomatitis virus (VSV). The mRNA level of STINGβ, but not SITNGα, was also dampened in peripheral blood leukocytes of patients with systemic lupus erythematosus. STINGβ was expressed from its own promoter regulated independently of STINGα. Although STINGβ has no transmembrane domain, it was also found to localize primarily in the endoplasmic reticulum together with SITNGα. Functionally, STINGβ effectively suppressed the production of type I IFNs and pro-inflammatory cytokines by inhibiting the activation of IRF3 and NF-κB pathways induced by STINGα, RIG-I-like receptors and TRIF as well as CDNs and various RNA and DNA viruses. STINGβ also inhibited the phosphorylation of TBK1 and IRF3 induced by these stimuli. By co-immunoprecipitation, STINGβ was shown to interact with STINGα, MAVS, TBK1 and IKKε. In HeLa cells, STINGβ co-localized with both STINGα and TBK1. Overexpression of STINGβ prevented TBK1 from complex formation with STINGα or TRIF. Both STINGα and STINGβ were pulled-down by 2’3’-cGAMP agarose. Moreover, overexpression of STINGβ reduced the binding affinity of STINGα to 2’3’-cGAMP. STINGβ also antagonized the antiviral function of STINGα. Depletion of STINGβ could boost the induction of type I IFNs by cytosolic DNA and enhanced antiviral effect. Taken together, STINGβ generated by alternative transcriptional initiation functions as a physiological suppressor of innate immune response.