IRE1A-XBP1 CONTROLS GROUP 2 INNATE LYMPHOID CELLS IN INTESTINAL INFLAMMATION AND FIBROSIS

Abstract

Abstract BACKGROUNDS Group 2 innate lymphoid cells (ILC2s) are increased in peripheral blood and intestinal samples from patients with ulcerative colitis or Crohn’s disease and murine models of IBD. ILC2s produce a large amount of type 2 cytokines including IL-5 and IL-13 and growth factor amphiregulin (AREG) that induce goblet cell hyperplasia and mitigate acute inflammation in the gut. On the other hand, prolonged or uncontrolled ILC2 activation may contribute to chronic inflammation and tissue fibrosis. However, the regulatory mechanisms and functional impact of ILC2s in IBD remains elusive. The IRE1a-XBP1 pathway orchestrates cellular stress response that plays a key role in intestinal inflammation. METHODS We generated conditional knockout Ire1aflo/floxIl5-Cre (Ire1aΔIl5) with Ire1a deleted in ILC2s. Rag-/-Ire1aΔIl5 mice were also generated to exclude possible contribution of T cells during chronic inflammation. Dextran sodium sulfate (DSS) was given in drinking water to induce acute/chronic colitis and intestinal fibrosis. We also collected colonic biopsies from healthy individuals and sort-purified ILC2s for flow cytometry analysis. RESULTS Mouse intestinal ILC2s express high level of Ire1a. The number of ILC2s increased in small intestine (SI) and colon of Ire1aΔIl5 mice compared to Il5-Cre controls. Additionally, SI and colonic Ire1aΔIl5 ILC2s produced more IL-5, IL-13, and AREG after ex vivo stimulation with IL-33 and IL-25. Upon DSS challenge, Ire1aΔIl5 mice exhibited milder signs of disease including weight loss, clinical scores as well as colon shortening and histology (Figure 1), which was accompanied by elevations in ILC2-derived cytokines and AREG as well as heightened goblet cell differentiation and mucin production. In response to 3 cycles of DSS, Rag-/-Ire1aΔIl5 mice developed worsened chronic colitis, fibrosis, and increased fibrosis markers including collagens and vimentin. Intraperitoneal injection of neutralizing antibody against IL-13 or AREG reversed the phenotype. In sort-purified human ILC2s, selective IRE1 inhibitor 4u8C enhanced production of IL-13, while selective IRE1 activator IXA4 suppressed IL-13 in response to IL-2, IL-7, and IL-33 ex vivo (Figure 2). CONCLUSIONS We demonstrate that IRE1a-XBP1 constrains cytokine production by intestinal ILC2s during inflammation. IRE1a deficiency in ILC2s protected against acute colitis with heightened type 2 cytokines and AREG; while loss of IRE1a exacerbated chronic colitis and fibrosis. We identified IRE1a as a potential therapeutic target for fibrostenotic CD. Figure 1 Ire1aΔIl5 mice harbor elevated ILC2s (A) and produce increased IL-5, IL-13, and AREG by (B) SI and (C) colonic ILC2s ex vivo. (D-H) Ire1aΔIl5 mice are protected from acute DSS colitis. *P<0.001. Figure 2 IRE1 modulators control IL-13 production by human ILC2s. **P<0.01.