The Sentinel's Betrayal: A STING‐Driven Necroptotic Axis in Caspase‐8‐Deficient Pathology and STING‐Associated Vasculopathy With Onset in Infancy (SAVI)

In a recent publication in Nature, Kelepouras et al. [1] reveal a new function of STING in regulating programmed cell death. The study identifies that the absence of Caspase-8 leads to abnormal activation of the cGAS–STING signaling pathway, which induces the upregulation of ZBP1 and MLKL and licenses necroptosis independently of TNFR1 and FADD. Crucially, the authors also demonstrate that aberrant activation of STING, whether caused by Caspase-8 loss or by gain-of-function mutation in Sting1, invariably drives inflammatory necroptosis [1]. Necroptosis is a programmed cell death pathway executed by RIPK3 and mixed lineage kinase domain-like (MLKL), that has been shown to be involved in various inflammatory diseases and tissue injuries [2]. Caspase-8 is a key negative regulator of necroptosis Its deficiency permits aberrant activation of the pathway and causes lethal inflammation in development or in a tissue-selective manner. Deletion of TNFR1 can modestly delay dermatitis by inhibiting canonical necrosome assembly. These findings indicate that necroptosis can be activated independently of TNF-induced, FADD-mediated recruitment of RIPK1 and RIPK3 [3]. However, the upstream signal source and the main regulatory factors remain unclear at this point. STING as a core factor in cellular DNA recognition and the interferon pathway, has traditionally been considered to mainly participate in antiviral and innate immunity, but whether it can be associated with necroptosis has no direct evidence in the past. Deletion of Sting ameliorated the lethal dermatitis caused by loss of Caspase-8 in keratinocytes, providing direct genetic evidence for STING involvement in the observed skin pathology. Genetic and biochemical evidence indicates that STING agonists enhance TNF-induced cell death in a RIPK3- and MLKL-dependent manner. ZBP1 is essential for the sensitization mediated by STING, and its deletion can completely block STING-induced necroptosis. Genetic and biochemical analysis further shows that STING upregulates ZBP1 and promotes formation of a FADD- and TNFR1-independent ZBP1–RIPK1–RIPK3 complex, thereby triggering necroptosis. Mechanistically, STING upregulates Zbp1 expression via downstream interferon-stimulated gene (ISG) signaling and promotes cytosolic Z-DNA accumulation, which binds to the Zα domains of ZBP1 to initiate activation. ZBP1 then interacts with RIPK1 and RIPK3 through its RHIM domains, forming a FADD- and TNFR1-independent complex that drives necroptosis. In vivo experiments showed that Casp8E-KO mice exhibited severe dermatitis and early death, while simultaneous knockout of STING or TNFR1 could notably delay the progression of inflammation and improve survival, suggesting that STING plays a key role in the pathogenesis of skin inflammation. Histological and molecular analysis further confirmed that STING deficiency specifically inhibited necroptosis and interferon response in keratinocytes, revealing the independent pathological role of the STING-ZBP1-RIPK3-MLKL axis in inflammatory diseases. These data position the STING–ZBP1–RIPK3–MLKL axis as an independent pathological driver of inflammatory disease. Using compound gene knockout mice, the study demonstrated that heterozygous deletion of Zbp1 or Sting partially mitigated the pathology of Casp8E-KO and Tnfr1−/− mice, although these mice still succumbed by 4 weeks due to dermatitis and inflammation. In contrast, homozygous Zbp1 deletion or keratinocyte-specific Sting knockout fully or largely rescued the phenotype, normalizing survival and behavior. Histological analysis confirmed that complete deletion of ZBP1 or STING eliminated epidermal inflammation, immune cell infiltration, and necroptosis markers, whereas heterozygous knockouts displayed residual epidermal thickening, apoptosis, and upregulation of ZBP1 and p-MLKL [4]. Persistent ZBP1 deletion fully prevented adult dermatitis, while STING deletion prolonged survival but did not completely abrogate inflammation, emphasizing the critical role of the STING–ZBP1 axis in Caspase-8-deficient inflammatory pathology. They found that Caspase-8 deficiency causes chromosomal instability and DNA damage, leading to nuclear DNA leakage into the cytoplasm, thereby activating the cGAS-STING pathway as a potent signal [5]. Indeed, Casp8E-KO and cGAS−/− mice completely replicated the protective effect of Casp8E-KO and Sting−/−, while MAVS deletion had no effect, proving that STING activation after Caspase-8 deficiency is completely dependent on cGAS. Z-NAs (both Z-DNA and Z-RNA) accumulate and co-localize with ZBP1. STING agonism increases Z-NAs signal intensity, indicating that STING promotes the generation or stabilization of Z-NAs that activate ZBP1. Functionally, this STING-dependent ZBP1 activation promotes the recruitment and phosphorylation of RIPK1 and RIPK3 and triggers necroptosis. Translational studies further confirmed this in STING gain-of-function mutant mice. ZBP1 was notably upregulated, in the mean while necroptosis and inflammatory lesions were observed in skin, lungs, thymus, and spleen. Notably, the complete absence of RIPK3 completely blocked necroptosis and markedly improved the immune homeostasis and tissue pathology of STING-Associated Vasculopathy with onset in Infancy (SAVI) mice, demonstrating that STING-induced necroptosis is the core driving force of SAVI disease (Figure 1). The translational implications are compelling. In SAVI patient samples, sustained interferon responses are accompanied by transcriptional priming toward necroptosis, with upregulation of ZBP1, RIPK3, and MLKL. In a preclinical N153S STING gain-of-function mouse model, necroptosis in skin and multi-organ inflammation mirrors the clinical manifestations. Strikingly, deletion of RIPK3 rescues survival and normalizes immune homeostasis across tissues. Together, these findings position STING-driven necroptosis as a central pathogenic mechanism in SAVI and support therapeutic strategies targeting the ZBP1–RIPK3–MLKL axis. Beyond its established role in necroptosis, the STING–ZBP1 axis may also intersect with other inflammatory signaling pathways, such as NF-κB or pyroptosis, particularly under conditions of Caspase-8 deficiency. Exploring such crosstalk represents an important direction for future studies. In summary, this study revealed a novel mechanism by which STING can trigger necroptosis by upregulating ZBP1 and driving its interaction with RIPK1-RIPK3, filling an important gap in the non-classical cell death pathways under Caspase-8 deficiency. This discovery not only enriches our understanding of the molecular network of necroptosis but also provides potential intervention targets for the treatment of inflammatory-related diseases such as dermatitis, colitis, and certain autoimmune diseases. However, this study was mainly based on mouse models, and the specific role of the STING-ZBP1 pathway in the human immune system and disease environment remains to be validated. Apart from that, the dual role of STING activation in antiviral immunity and pro-inflammatory responses also suggests that more precise regulatory methods are needed in the future to achieve a balance between disease treatment and immune homeostasis. Overall, this study provides a key breakthrough in understanding the upstream regulatory network of necroptosis and provides a conceptual framework for therapeutic targeting of the STING–ZBP1 pathway in inflammatory disorders. Long Zhang conceived and supervised the manuscript. Xin He and Bin Wang wrote and revised the manuscript. Xin He and Bin Wang contributed equally to this study. All authors have read and approved of the final manuscript. The current work was supported by the Chinese National Natural Science Funds (322525002 and W2411011), a Key R&D Program of Zhejiang Province (2024C03142) and the Joint Project of Pinnacle Disciplinary Group, the Second Affiliated Hospital of Chongqing Medical. We thank G. Xiao and Z. Lin from the Core Facilities, Zhejiang University School of Medicine for their support. The authors have nothing to report. The authors have nothing to report. The authors declare no conflicts of interest. The authors have nothing to report.