ROLE OF SPHINGOSINE-1-PHOSPATE PATHWAY IN INTESTINAL EPITHELIAL CELLS AND ITS INVOLVEMENT IN INTESTINAL TUMORIGENESIS.

Title: Role of Sphingosine-1-phosphate pathway in intestinal epithelial cells and its involvement in intestinal tumorigenesis. Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid metabolite, involved in several cellular processes. S1P is both an intracellular second messenger and the ligand of five EDG family G protein-coupled receptors 1-5 (S1PR1-5). S1PRs are widely expressed on many cell types of different tissues including the gut. However, so far there are no evidences about which S1P receptor is expressed on intestinal epithelial layer. Therefore, the first proposal of this project is to characterize the S1P pathway on intestinal epithelial cells and to explore the physiological functions of the S1PRs on intestinal epithelial cells. I characterized the expression of the ubiquitous expressed S1P receptors S1PR1, S1PR2 and S1PR3 on human and murine primary intestinal epithelial cells isolated from healthy controls by qRT-PCR and western blot. Interestingly, both the analyses revealed that the S1PR2 is the most abundantly expressed S1PR in the colonic epithelium, among the three S1PRs analyzed. Consequentially, I decided to focus specifically on S1PR2 for the study of the physiological functions of the S1PRs in the epithelium. Recently, in vitro studies have been highlighted the capacity of S1P to enhance epithelial barrier function suggesting this pathway as new potential target of intestinal barrier restoration (Vetrano et al., 2011; Greenspoon et al., 2011). However, so far there are no evidences of which receptor mediates this effect. Since the expression of the S1PR1 and S1PR3 transcript is very low also in the human epithelial cell line Caco-2, whereas the S1PR2 transcript is constitutively expressed, I assessed the hypothesis of the S1PR2 involvement in the S1P-mediated regulation of the epithelial barrier permeability. I evaluated the intestinal permeability in vivo in S1pr2 knock out mice (S1P2-/-) at baseline and after DSS-induced colitis by the Evans Blue method. Surprisingly, no difference in terms of intestinal permeability and susceptibility to DSS-induced colitis was found between S1P2-/- and WT mice. In line with this, the expression of tight junction (TJ) proteins at baseline in S1P2-/- mice was comparable to WT mice thus excluding direct defects of the epithelial junction proteins in absence of S1PR2. Several studies reported that the S1P signalling controls the cell growth and few papers bring to light the anti-proliferative role of the S1PR2 in vitro and in vivo in different cell lines (Ikeda et al., 2003; Goparaju et al., 2005). Therefore, it was of interest to analyse the role of S1PR2 in the proliferation of the colonic epithelial cells. I evaluated the proliferation in vitro and in vivo respectively in Caco-2 cell line under the treatment with different concentrations of S1PR2 inhibitor and in S1P2-/- mice by BrdU incorporation assay. Both analyses have showed an increase in the epithelial proliferation rate in absence of S1PR2. Since the abnormal proliferation is associated with tumor development and the S1P2 acts as a tumor suppressor in different types of tumors, I tested whether the loss of S1PR2 could be a pro-tumorigenic key for the development of colorectal cancer. To this end, a AOM/DSS model of colitis associated cancer was induced in S1P2-/- and WT mice. All mice were treated with a single injection of azoxymethane (AOM) (10mg/Kg) a known cancerogenic agent, followed by four cycles of DSS (2,5%). During the experiment, the animals were monitored for body weight loss, rectal bleeding and fecal consistency, and analyzed by endoscopic procedure to monitor the tumor growth in vivo. Interestingly, although no differences in terms of grade and extension of intestinal inflammation were observed between two groups, S1PR2 mice were characterized by a higher number of tumors compared to WT mice. These results were confirmed by three independent experiments thus indicating a crucial role of S1PR2 in the intestinal tumorigenesis. The incidence of high grade adenomas is higher and the few carcinomas observed are bigger in s1pr2-/- compared with wild type mice. These observations indicate that the loss of S1PR2 leads to an increased onset and a faster progression of colorectal cancers. No difference in the clinical parameters of inflammation (body weight loss, disease activity index, colon length and Rachmilewitz score) has been observed between S1PR2 knockout and wild type mice, suggesting that the increase in the cancer development does not depend on a higher inflammatory response in the S1P2-/- mice. The analysis of tumors in the Apcmin/+ genetic model of intestinal and colon cancer further confirms this hypothesis. Indeed, I observed a higher incidence and size of tumors in the colon of s1pr2-/-/Apcmin/+ mice compared to s1pr2+/+/Apcmin/+ controls and, while control mice mainly develop few adenomas, almost all the tumors observed in S1PR2-null mice are carcinomas. Overall these data point out a protective role of S1PR2 during the development of the colorectal cancer, both spontaneous and inflammation-driven. By immunohistochemistry staining and western blot, I observed a strong decrease of the S1PR2 expression both in early staged human carcinomas and in the murine adenomatous polyps and adenocarcinomas of both AOM/DSS and genetic models, suggesting that S1PR2 down-regulation is an early event required for the development of colorectal cancers. Since the pharmacological inhibition of S1PR2 increases the proliferation of epithelial cells in vitro and the expression of S1P2 is mainly observed in the differentiated epithelial cells localized in the upper region of the crypts in the colon and at the top of the crypts and in the villi in the small intestine, I investigated whether S1PR2 is involved in the regulation of the intestinal epithelial cell differentiation. To verify this hypothesis, I took advantage from the mini-gut culture system and I observed that the pharmacological inhibition of S1PR2 during the early phases of the organoid development blocks the complete formation of the branches and keep the organoids in a round shape and undifferentiated. Interestingly, the organoids treated with S1PR2 inhibitor show an enrichment of stem cell markers such as Lgr5 and Olfm4 compared to untreated organoids. This finding raises the unrknown role of S1PR2 in determining the differentiation of the intestinal epithelial cells, thus suggesting the S1PR2 as a master player in the maintenance of the crypt-villus differentiation axis. Overall these results suggest that the loss of S1PR2 affects the differentiation of epithelial cells promoting their proliferative potential. Taken together these results suggest that S1PR2 functions as a negative regulator of epithelial cell proliferation in the colon and may contrast tumorigenesis by promoting epithelial cells differentiation and impeding their malignant transformation.