Gelsevirine is a novel STING-specific inhibitor and mitigates STING-related inflammation in sepsis

**Background:** Sepsis is a life-threatening condition characterized by uncontrolled inflammatory responses and multiple organ failure, causing approximately 11 million deaths annually worldwide. The stimulator of interferon genes (STING) detects cytosolic DNA from pathogens or damaged host tissues, leading to type I interferon and proinflammatory cytokine production, which plays a detrimental role in sepsis. Gelsevirine (GS) is an alkaloid from Gelsemium elegans with known anti-inflammatory properties, but its effect on STING signaling was unknown.

**Methods:** The study used in vitro models (Raw264.7 murine macrophages, THP-1 human monocytes, HEK293T cells) and an in vivo mouse model of cecal ligation and puncture (CLP)-induced sepsis. Cells were pretreated with GS (10 μM) for 6 hours before stimulation with STING agonists (2'3'-cGAMP 5 μg/ml, ISD 2 μg/ml, or poly(dA:dT) 5 μg/ml). mRNA expression of Ifnb1, Cxcl10, and Il6 was measured by RT-PCR. In silico docking and surface plasmon resonance (SPR) assessed GS binding to STING. Biotin pull-down assay confirmed competitive binding. STING dimerization was analyzed by native gel electrophoresis. Ubiquitination was assessed by immunoprecipitation and immunoblot. In vivo, C57BL/6J mice (2-month-old) received GS (10 or 20 mg/kg) intraperitoneally 5 hours after CLP. Survival was monitored for 7 days. At 15 hours post-CLP, lung injury (H&E scoring, wet-to-dry ratio, BALF protein), serum markers (IL-6, TNF-α, BUN, creatinine, AST, ALT), and STING pathway proteins (STING, p-TBK1, p-p65, p-IRF3) were evaluated. STING-deficient mice were used to confirm specificity.

**Key Results:** GS dose-dependently inhibited 2'3'-cGAMP-induced Ifnb1 expression with IC50 values of 5.365 μM in Raw264.7 and 0.766 μM in THP-1 cells. GS also suppressed Cxcl10 and Il6 mRNA induced by 2'3'-cGAMP, ISD, and poly(dA:dT). In silico docking predicted GS binding to the CDN-binding pocket of STING, and SPR showed a Kd of 27.6 μM. Biotin-GS pull-down confirmed binding, which was competed by excess GS or 2'3'-cGAMP. GS inhibited STING dimerization and reduced phosphorylation of TBK1, IRF3, and p65. GS promoted K48-linked ubiquitination of STING and increased interaction with TRIM21, upregulating TRIM21 expression. In CLP mice, GS (20 mg/kg) significantly improved survival (p < 0.05 vs CLP group), reduced lung injury scores (from ~8 to ~4), decreased lung wet-to-dry ratio, lowered BALF protein, and reduced serum IL-6, TNF-α, BUN, creatinine, AST, and ALT. GS also reduced F4/80 and S100A9 positive cells in lungs and suppressed STING, p-TBK1, and p-p65. In STING-deficient mice, GS provided no additional protection, confirming STING-dependent mechanism.

**Clinical Implications:** This study identifies GS as a novel STING-specific inhibitor with dual mechanisms: competitive binding to the CDN-binding pocket and promotion of K48-linked ubiquitination and degradation via TRIM21. GS shows therapeutic potential for sepsis, especially given its efficacy when administered 5 hours after CLP, mimicking clinical timing. The compound's good biosafety (no cytotoxicity up to 160 μM in most cells) and oral availability (implied by prior work) suggest translational promise. However, further studies are needed to evaluate pharmacokinetics, optimal dosing, and efficacy in other STING-driven diseases (e.g., autoimmune disorders). The Kd of 27.6 μM is relatively high compared to some inhibitors (e.g., Astin C Kd 53 nM), but GS's additional ubiquitination mechanism may compensate. The study is limited by use of a single animal model and lack of human clinical data.