P094 Differences in expression of PPARγ in small intestine vs. colon impact the effect of 5-aminosalicylates in inflammatory bowel disease

I. Bakke1,2, T. Bruland1,2, K.S. Eriksen1, H.K. Skovdahl1, S. Gopalakrishnan1, S. Thorsvik1,2,3, A.V.B. Granlund1,4, A.K. Sandvik1,2,3,4, A.E. Østvik MD1,2,3

1Norwegian University of Science and Technology, Department of Clinical and Molecular Medicine- Faculty of Medicine and Health Sciences, Trondheim, Norway, 2St Olavs Hospital, Trondheim University Hospital, Clinic of Medicine, Trondheim, Norway, 3St. Olav`s University Hospital, Department of Gastroenterology, Clinic of Medicine, Norway, Trondheim, Norway, 4Norwegian University of Science and Technology, Centre of Molecular Inflammation Research, NTNU, Trondheim, Norway


The nuclear receptor peroxisome proliferation-activated receptor γ (PPARγ) harbours anti-inflammatory effects. There is evidence that PPARγ mediates the effect of 5-aminosalicylic acid (5-ASA). 5-ASA is the first-line drug in ulcerative colitis (UC, while its use in Crohn`s disease (CD) is debated and not recommended according to guidelines. We hypothesise that the inconsistent therapeutic effect of 5-ASA in Crohn’s disease is caused by different expression of PPARγ between the large and small intestine.


Levels of PPARγ mRNA were measured by RNASeq in mucosal biopsies from (1) active/inactive ileal CD (n = 14/17) and healthy controls (n = 9) and (2) colonic biopsies from patients with active/inactive UC (n = 24/24) and active/inactive CD (n = 24/21) and healthy controls (n = 20). Subsets of ileal and colonic biopsies were examined by western blot, immunohistochemistry (IHC) and in situ hybridisation (ISH). The effects of 5-ASA on PPARγ expression and TNF/IL17/polyI:C induced cytokine release were examined in the colonic cell-line HT29 and primary human IECs (colonoids) using RNASeq and ELISA in.


PPARγ mRNA was strongly downregulated in colonic biopsies from inflamed mucosa of UC (log2 = −1.65, fold change 0.319 (p < 0.001)) and CD (log2 = −1.3, fold change 0.406 (p < 0.001)) compared with healthy controls. In ileal biopsies from CD and controls, PPARγ mRNA was not differentially expressed between inflamed and non-inflamed or healthy mucosa. These findings were confirmed by ISH in a subset of biopsies. Western blot analysis and IHC revealed almost undetectable levels of PPARγ protein in ileum, in the colon however, PPARγ was strongly expressed in healthy controls and inactive IBD; with significantly lower levels in active IBD. PPARγ was downregulated by TNF, TNF+IL17 and TNF + poly(I:C) in colonoids derived from IBD-patients (n = 3) and non-IBD controls (n = 3). 5-ASA attenuated the release of TNF-induced proinflammatory cytokines such as CXCL1, CXCL8 and CXCL10 from both HT29 cells and colonoids. This effect was reversed by the PPARγ antagonist GW9662.


5-ASA harbours anti-inflammatory effects on intestinal epithelial cells mediated by epithelial PPARγ. We suggest that the previously observed lack of effect of 5-ASA in CD is related to differences in PPARγ expression in small intestine vs. colon. These results suggest that patients with Crohn’s colitis may benefit from 5-ASA similarly to UC patients and challenge the current view on use of 5-ASA in CD.