Retinoids, bile acids and PPARs in Barrett's oesophagus.

Intestinal metaplasia of the oesophagus (Barrett’s oesophagus) is the most important risk factor for oesophageal adenocarcinoma. Although the role of mixed bile and acid gastro-oesophageal reflux is well established, the contribution of nuclear receptor signalling in the pathogenesis of Barrett’s oesophagus remains largely unknown. The recent article by Chang et al. (Gut 2007;56:906–17) showing that retinoic acid and lithocholic acid play an important role in the development of oesophageal intestinal metaplasia, is of particular interest, because retinoid receptors are also heterodimeric partners of multiple nuclear receptors. We recently published a report demonstrating the over-expression of the nuclear bile acid receptor FXR in Barrett’s oesophagus and its possible role in the regulation of apoptosis. Newly obtained results from our laboratory indicate that also other nuclear receptors, such as peroxisome proliferatoractivated receptors alpha and gamma (PPARa and c), are up-regulated in Barrett’s oesophagus. Both FXR and PPARs are transcription factors which are abundantly expressed in the lower digestive tract, and are activated by bile, fatty acids and retinoids (by virtue of their heterodimeric form with the retinoid X receptor, RXR). The observations by Chang et al. and our findings, suggest that the retinoid signalling cascade is involved in the pathogenesis of Barrett’s oesophagus and could be mediated by nuclear receptor signalling mechanisms. Biopsy samples were obtained at endoscopy from normal oesophagus, Barrett’s oesophagus and oesophageal adenocarcinoma (six patients per group). Expression of PPARa and PPARc was quantified by real time PCR (primers F 59-GCAAACTTGGA CCTGAAC-39; R 59-CCCATTTCCATACG CTAC-39 and F 59-CCAAGCTGCTCCAG AAAA-39; R 59-AGCGGGTGAAGACTCAT GT-39, respectively) using 18S ribosomal RNA for normalisation. Immortalised cells derived from Barrett’s oesophageal epithelium (CP-18821) were cultured as previously described and treated with agonists or antagonists of PPARa (10 mmol/l WY14463 and 20 mmol/l MK886, respectively) and of PPARc (20 mmol/l rosiglitazone and 30 mmol/l PD068235, respectively). Apoptosis in Barrett’s derived cells was assessed by flow cytometry after labelling with 7-aminoactinomycin D. Results were compared using the Student t test and a p value ,0.05 was considered statistically significant. The expression of both PPARa and PPARc was significantly higher in Barrett’s oesophagus compared to normal mucosa and adenocarcinoma of the oesophagus (fig. 1). This is in agreement with several studies showing that PPARs are present in intestinal epithelial cells, where they are involved in the regulation of inflammatory processes and apoptosis. Treatment with the PPAR antagonists MK886 and PD068235 significantly increased the proportion of apoptotic CP-18821 cells (fig. 2), suggesting that PPAR signalling may contribute to the regulation of apoptotic processes in Barrett’s oesophagus. Treatment with PPAR agonists WY-14463 or rosiglitazone did not affect apoptosis. These new data show that PPARa and PPARc are up-regulated in Barrett’s oesophagus and suggest that they contribute to the regulation of apoptosis. Since these receptors constitute heterodimeric units with RXR, our data extend those of Chang et al., supporting the hypothesis that retinoids, bile acids and fatty acids play an active role in the pathogenesis of Barrett’s oesophagus.