Endothelial STING and STAT1 mediate IFN-independent effects of IL-6 in an endotoxemia-induced model of shock
Nina Martino, Erin K. Sanders, Ramon Bossardi Ramos, Iria Di John Portela, Fatma Awadalla, Shuhan Lu, Dareen Chuy, Neil Poddar, Mei Xing G. Zuo, Uma Balasubramanian, Peter A. Vincent, Pilar Alcaide, Alejandro P. Adam
Nina Martino, Erin K. Sanders, Ramon Bossardi Ramos, Iria Di John Portela, Fatma Awadalla, Shuhan Lu, Dareen Chuy, Neil Poddar, Mei Xing G. Zuo, Uma Balasubramanian, Peter A. Vincent, Pilar Alcaide, Alejandro P. Adam
View: Text | PDF
Research Article Inflammation Vascular biology

Endothelial STING and STAT1 mediate IFN-independent effects of IL-6 in an endotoxemia-induced model of shock

Abstract

Severe systemic inflammatory reactions, including sepsis, often lead to shock, organ failure, and death, in part through an acute release of cytokines that promote vascular dysfunction. However, little is known about the vascular endothelial signaling pathways regulating the transcriptional profile in failing organs. Our work focused on signaling downstream of IL-6, due to its clinical importance as a biomarker for disease severity and predictor of mortality. Here, we show that loss of endothelial expression of the IL-6 pathway inhibitor SOCS3 promoted a type I IFN–like (IFNI-like) gene signature in response to endotoxemia in mouse kidneys and brains. In cultured primary human endothelial cells, IL-6 induced transient IFNI-like gene expression in a noncanonical, IFN-independent fashion. We further show that STAT3, which we had previously demonstrated to control IL-6–driven endothelial barrier function, was dispensable for this activity. Instead, IL-6 promoted a transient increase in cytosolic mitochondrial DNA and required STAT1, cGAS, STING, and IRF1, -3, and -4. Inhibition of this pathway in endothelial cell–specific STING-KO mice or global STAT1-KO mice led to reduced the severity of the response to acute endotoxemic challenge and prevented expression of an endotoxin-induced IFNI-like gene signature. These results suggest that permeability and DNA-sensing responses are driven by parallel pathways downstream of this cytokine, provide potential insights into the complex response to acute inflammatory responses, and offer the possibility of novel therapeutic strategies for independently controlling the intracellular responses to IL-6 in order to tailor the inflammatory response.

Authors

Nina Martino, Erin K. Sanders, Ramon Bossardi Ramos, Iria Di John Portela, Fatma Awadalla, Shuhan Lu, Dareen Chuy, Neil Poddar, Mei Xing G. Zuo, Uma Balasubramanian, Peter A. Vincent, Pilar Alcaide, Alejandro P. Adam

×

Figure 2

Loss of endothelial SOCS3 promotes a type I IFN–like response to LPS in the kidney and brain endothelium.

Options: View larger image (or click on image) Download as PowerPoint
Loss of endothelial SOCS3 promotes a type I IFN–like response to LPS in ...
(A) TRAP-Seq data demonstrating the specificity of the loss of SOCS3 expression in SOCS3iEKO mice. (B) Volcano plot demonstrating a strong alteration of the endothelial transcriptional response to LPS. FC, fold change. (C) Heatmap of the most significant kidney transcriptional changes in response to LPS in SOCS3iEKO mice compared with WT mice. (D) Normalized gene set enrichment analysis scores of kidney and brain TRAP-Seq comparing the LPS response in SOCS3iEKO and WT mice. (E) Strong IFN-α signature in TRAP-Seq from LPS-treated SOCS3iEKO mice. (F) Plasma CXCL10 levels are greatly increased by an LPS challenge in WT and further increased in SOCS3iEKO mice. Box-and-whisker plot (line defines the median, the box extends from the 25th to 75th percentiles, and whiskers define the full range of value). *P < 0.05 (2-way ANOVA and Šidák’s post hoc test; n = 6–10). (G) Correlation between plasma CXCL10 levels and disease severity. Dotted lines represent 95% CI. (H) Immunohistochemistry of kidney to detect CXCL10. Scale bars: 20 μm. Representative of 3 independent cohorts.