OP018 High-fat diet and inflammation drive intestinal fibrosis enhancing epithelial–mesenchymal transition through the activation of S1P3 signalling
C. Mascaraque1, S. Elangovan2, L. Petti1, A. Piontini2, C. Correale1, V. Arena3, B. Romano1, F. Ungaro2, V. Garlatti2, S. D’Alessio2, G. Fiorino1, A. Spinelli2,4, S. Danese1,2, S. Vetrano2*
1Humanitas Research Institute, IBD Center Laboratory of Immunology in Gastroenterology, Rozzano, Italy, 2Humanitas University, Department of Biomedical Sciences, Rozzano, Italy, 3Catholic University of Rome, Institute of Pathology, Rome, Italy, 4Humanitas Reseach Institute, Colon and Rectal Surgery, Rozzano, Italy
Epithelial-mesenchymal transition (EMT) is considered to contribute to intestinal fibrosis. However, the mechanisms underlying this process remain largely unknown. Sphingosine-1-phosphate (S1P), a pleiotropic lipid mediator crucially involved in inflammation and cancer, has been identified as a master regulator of fibrosis by activation of sphingosine-1-phosphate receptor 3 (S1P3). Recently S1P has been demonstrated being able to induce EMT in cancer. However, so far no data are available on the functional role of S1P/S1P3 signalling in the gut and its involvement in EMT process associated to intestinal fibrosis. Therefore, our aim was to explore the role of S1P/S1P3 signalling in intestinal fibrosis.
Fibroblasts isolated from fibrotic and non-fibrotic biopsies of Crohn’s disease patients were analysed before and after S1P and TGF-β stimulation, in the presence or in the absence of S1P3 inhibitors and/or palmitic acid or lipid mixture. S1P levels and S1P3 expression were quantified in cell supernatants and whole tissue. Pro-fibrotic factors were analysed by RT-PCR, whereas collagen deposition by Sirius red. Overexpression of S1P3 was carried out on Caco-2 cells by lentivirus infection. Nine EMT-genes were analysed by RT-PCR. WT and S1P3 ko mice treated with TNBS via enema were fed either a high-fat diet (HFD) or normal diet for up to 12 weeks. Fibrosis and inflammation were quantified by histologic scoring, collagen synthesis, and cytokine expression.
S1P3 expression significantly increased in both whole fibrotic tissue and fibrotic fibroblasts. S1P, TGF-β, and HFD stimulation augmented on fibrotic fibroblasts S1P3 expression, which was also associated with increased expression of all pro-fibrotic factors analysed. The treatment with specific S1P3 inhibitors reduced proliferation and migration of fibroblasts as well as S1P- and TGF-β-dependent collagen deposition, and pro-fibrotic factors. HFD-fed WT mice exhibited a marked increase of epithelial S1P3 rather baseline displaying a higher accumulation of collagen in the mucosa. In contrast, S1P3 ko mice did not show any changes. Furthermore the combined association of HFD with chronic inflammation resulted in worsening intestinal fibrosis and alfa-SMA expression from epithelial cells. In vitro a long-HFD stimulation augmented expression of S1P3 on Caco-2 cells, which displayed EMT changes as well as S1P3 overexpressing cells.
Overall, our data demonstrate for the first time that HFD and inflammation cooperate to intestinal fibrosis modulating S1P3 receptor on epithelium, which promotes EMT. These results pave the way for a new specific anti-fibrotic therapy for stricturing complications in CD patients.