Ursodeoxycholic acid inhibits colonic mucosal cytokine release and prevents colitis in a mouse model of disease

Toll-like receptors (TLR) promote cytokine release from intestinal epithelia, thereby driving inflammation. Although ursodeoxycholic acid (UDCA) is well known to exert anti-inflammatory effects in the liver, its potential for treating intestinal inflammation has not been well-studied. Here, we investigated UDCA effects on TLR-driven colonic mucosal cytokine release and induction of inflammation in a mouse model of colitis. Resected human colonic mucosa and T84 colonic epithelial cells were treated with specific TLR agonists (TLR3: Polyinosinic-polycytidylic acid, 25 μg/ml; TLR4: lipopolysaccharide (LPS), 100 ng/ml), in the presence or absence of UDCA. Released cytokines were measured by ELISA and luciferase reporter assays were performed in HEK293T cells. Dextran sodium sulphate (DSS) colitis was induced by administering DSS (2.5 %) in the drinking water of male C57BL/6 mice, along with daily intraperitoneal injections of UDCA (30 or 100 mg/kg) or vehicle for 5 days. Colitis severity was recorded as disease activity index (DAI) and by histological inflammation score. Results were analysed by ANOVA using Tukey multiple comparisons test and are expressed as the mean ± standard error of the mean. In T84 cells, UDCA (200 μM) attenuated TLR3-stimulated tumour necrosis factor α (TNFα) release from 32.9 ± 4.1 to 23.8 ± 3.7 pg/ml; interleukin-8 (IL-8) release from 1197.5 ± 110.1 to 923.8 ± 110.3 pg/ml and IL-1β release from 2.0 ± 0.2 to 1.6 ± 0.2 pg/ml (n = 4, p <0.05). Apical LPS (6 hours) induced basolateral IL-8 secretion from human colonic mucosa from 435 ± 46 in unstimulated controls to 959 ± 22 pg/ml and caused a 3 fold increase in secretion of Regulated on Activation, Normal T cell Expressed and Secreted (RANTES); increases that were abolished by UDCA (250 μM) (n = 3, p <0.001). Using NFκB and P125 luciferase reporter assays we determined that UDCA (200 μM) exerts its anti-inflammatory effects at the level of TANK-binding kinase 1 (TBK1) (n = 5, p <0.01). In the DSS model of colitis, UDCA (30 mg/kg and 100 mg/kg) significantly reduced disease severity from a DAI of 10.0 ± 0.3 in mice treated with DSS alone to 7.2 ± 0.7 and 5.8 ± 0.5, respectively (n = 6 - 12, p <0.001). Furthermore, histologically the inflammation score was also significantly reduced in UDCA (100 mg/kg)-treated animals to 24.3 ± 4.4, compared to 37.3 ± 0.8 in mice treated with DSS alone (n = 4 - 6, p <0.05). In conclusion, our data show that UDCA prevents inflammation in a mouse model of colitis and that the protective effects of UDCA may be due to inhibition of TLR-induced proinflammatory cytokine release. These findings suggest that UDCA represents a good target for developing new therapeutic approaches to treat IBD.