The Role Of Bile Acid In The Pathogenesis Of Barrett Esophagus

Background and purpose:Barrett’s esophagus (BE), a kind of precancerous lesion, is a pathological phenomenon which normal esophageal squamous epithelium is replaced by columnar epithelium containing goblet cells. The development of BE is associated with chronic inflammatory stimulation to esophagus. Cell transformation induced by BE is an important pathogenic factor for esophageal adenocarcinoma (EAC).It has been proved that an important factor which leads to BE is chronic gastroesophageal reflux disease (GERD). BE is found by endoscopy in about 5-10% patients with GERD. Studies suggest duodenal gastroesophageal reflux plays a key role in the pathogenesis of BE. Long-term stimulation of reflux fluid can induce the ectopic expression of bile acid susceptible receptors in esophageal mucosa, promote the development of BE and further induce the progression to EAC which is also named Barrett’s adenocarcinoma (BA). Patients with BE have a higher risk to develop EAC and in last decades, the incidence of BA also increased significantly. The pathogenesis of BE is still not fully understood. Studies of investigating the molecular mechanisms of metaplasia and tumor through in vitro cell culture models cannot explain the process of in vivo esophageal squamous epithelium progressing to BE and EAC due to the complexity of human immune system.Previous studies suggest the most important factors which cause esophageal injury and inflammation are gastric acid and pepsins. Long-term exposure to gastric acid and pepsins induces the injury of esophageal mucosa and the development of columnar metaplasia. Furthermore, the normal squamous epithelium can be replaced by columnar epithelium. Recent studies find bile acid salt in reflux fluid has important effects on the injury of esophageal mucosa and metaplasia. Some clinical studies suggest the reflux of bile acid salt is more common in patients with BE and the concentration of bile acid salt in esophagus of BE is significantly higher than general GERD. Moreover, in vitro and in vivo studies have confirmed bile acid salt has a promoting effect on the development of BE. Bile acid plays its role through binding to its receptor. Bile acid receptor FXR is associated with the development and progression of tumors and is also found to be highly expressed in BE tissues, which implicates bile acid and its receptor play an important role in BE.The metaplasia of BE is often accompanied by the ectopic expression of intestine-specific genes. One of the initial inducible genes in the intestinal metaplasia of esophageal mucosa is CDX-2, which is an intestine-specifically expressed nuclear transcription factors. The increased expression of the gene is associated with gastrointestinal mucosa epithelial metaplasia, dysplasia and tumors. In vitro studies also find CDX-2 has differentiation-inducible function. Furthermore, the interaction of CDX-2 and Muc2 promoter can activate the transcription of Muc2 and the differentiation of goblet cells. The expression of CDX-2 and Muc2 can occur in the area of esophageal squamous epithelium under the chronic stimulation of reflux fluid, which suggests the metaplasia of esophageal mucosa has been initiated by those pathogenic factors. The carcinogenic effect of bile acid salt is associated with the chronic inflammation of esophageal mucosa induced by it. Clinical studies find the expression of proinflammatory genes including COX-2, PPAR-g and proinflammatory cytokines in epithelial cells of BE is upregulated and the mucosa inflammation of BE aggravates. Moreover, the increased level of PGE2 in the tissues is closely related to the incidence of BE. COX-2 is a key rate-limiting enzyme in the synthesis of PGs. The high expression of COX2 can upregulate the level of PGE2 in esophageal mucosa tissues and worsen the inflammation of mucosa. In addition, chronic inflammation may induce BE by promoting oxidative stress response mediated by the accelerated release of reactive oxygen species (ROS). Various inflammatory mediators can induce the release of ROS and lead to the persistent chronic oxidative stress injuries of esophageal mucosa. With the aggravation of esophageal inflammation, the incidences of BE and EAC increase accordingly.The BE metaplasia and development of tumors are also accompanied by the expression of a set of signal proteins and transcription factors such as Bmp4, Tff2, Krt19, etc. Bmp4 is a biomarker of BE epithelium. Bile acid can promote the intestinal metaplasia of esophageal squamous epithelium by regulating the expression of CDX-2 mediated through Bmp4 signal pathway. Tff2 has a protective function for esophageal mucosa and has a potential to facilitate the repair of damaged mucosa. In inflammatory lesions, the expression of Tff2 is upregulated. Krt19 is a skeletal protein in epithelial cells and expresses in tumors originating from epithelial tissues. Krt19 can be used as a biomarker for the micro metastasis of epithelial tumors. Testing the expression of Bmp4, Tff2 and Krtl9 facilitate our understanding to the pathophysiological mechanisms of BE and EAC.The present study used human BE tissue samples and esophageal lesion samples of animal models to investigate the functions of bile acid in BE. Our study provides experimental evidence for further studying the relationship between bile acid as well as its receptor and the development of BE.Materials and Methods1. Materials1.1 Human BE Tissue SamplesThe samples were collected from 20 patients who were diagnosed with BE by endoscopy and pathology in Qianfo Hill Hospital of Shandong Province from 2010 to 2014. All procedures were consented by the patients. The samples collected from adjacent normal esophageal mucosa were used as controls. The tissue samples were taken from the mucosa every centimeter above the gastroesophageal junction and the sampling sites were distributed in four quadrants.1.2 Grouping and Processing of Wister RatsTwo weeks old male Wister rats (weighing about 250g-280g) were purchased from the Experimental Animal Center of Shandong University and were divided into operation group and non-operation group. The rats in the operation group were further divided into three groups. Group A included 30 rats which received total gastrectomy, proximal duodenum closure and end-to-side anastomosis of the esophagus and duodenum to form simple bile acid reflux. Group B included 30 rats which received gastroesophageal junction longitudinal incision and side-to-side anastomosis of the esophagus and duodenum to form mixed reflux of gastric acid and bile acid. Group C was control group in which 10 rats received simple open and close abdomen operation.The non-operation group also included three groups:Group D included 30 rats which were fed with drinking water containing bile acid (0.3% DCA, pH 7.0); Group E included 30 rats which were fed with acid drinking water (HCl 0.01mol/L, pH 2.0) containing pepsin(0.5mg/L); Group F was control group and included 10 rats which were fed with normal fodder and clean water.All rats were raised for 6 months and 6 months late the survival rats were sacrificed for further studies.2. Methods2.1 Preparation of Human BE Tissue Samples2.1.1 The expression of FXR was tested using immunohistochemical assay. Western blot and HPLC were used to detect FXR expression in BE tissue.2.1.2 The expression of CDX-2 and Muc2 was tested using immunohistochemical staining.2.1.3 The mRNA level of COX-2 gene in human BE tissue and normal esophageal mucosa tissue was measured using Real-time PCR.2.1.4 The correlation analysis of the expression of FXR and CDX-2 and COX-2 was performed.2.2 Preparation of esophageal tissue samples of model animals2.2.1 Histological ExaminationThe esophagus was removed after the rats were killed. The esophagus was cut along its longitudinal axis and the gross pathological changes were observed. The esophagus was divided into three parts longitudinally to take pathological and biochemical studies. HE stained sections of esophageal tissues were examined by two experienced pathologists. The pathological changes of the esophageal tissues included hyperkeratosis, epithelial hyperplasia, inflammation, ulcer, BE and EAC.2.2.2 Immunohistochemical assay was used to detect the expression of FXR in the esophageal tissues. Immunohistochemical staining was used to detect the expression of CDX-2 and Muc2 in the esophageal tissues.2.2.3 ELISA was used to detect the levels of plasma IL-6 and TNF-a of experimental animals.2.2.4 Semi-quantitative RT-PCR was used to detect the expression of Cdx2, Muc2, Bmp4,Kit19 and Tff2.Results1. Results of Human BE Tissue Experiments1.1 The expression of FXR increased significantly in human BE tissue. The present study included 20 cases of BE patients. Positive staining of FXR was found in BE tissue by immunohistochemical method and primarily displayed nuclear staining. Surrounding connective tissues were also stained. Positive staining of FXR was rare in normal esophageal mucosa tissues. Western blot showed FXR was strongly expressed in BE tissue and weakly or not expressed in normal esophageal mucosa tissue. The difference was statistically significant.1.2 Immunohistochemical staining showed that, the expression of Cdx-2 and Muc2 was weakly to strongly positive in human BE tissue and both genes were rarely expressed in normal esophageal mucosa tissue. Compared with the normal esophageal mucosa tissue, the mRNA level of CDX-2 in BE tissue increased significantly.1.3 Real-time PCR revealed the mRNA level of COX-2 in BE tissue increased significantly compared with the normal esophageal mucosa tissue.1.4 Correlation analysis suggested, there was a positive correlation between the expression of FXR and CDX-2, FXR and COX-2, and CDX-2 and COX-2.2. Results of Animal Experiments2.1 In the operation groups, compared with the control group, two types of operations were associated with the increased incidences of esophageal inflammation, BE and EAC. The differences had a statistical significance. However, the incidences of above conditions were not significantly different between two operation groups (Group A and Group B). Notably, compared with simple reflux, mixed reflux did not increase the incidence of BE (c2=0.0565,p>0.05). In the non-operation groups, compared with the control group, Group D (fed with bile acid) and Group E (fed with pepsin) were both associated with a increased incidences of inflammation and BE, and the incidences of BE in Group D and Group E were significantly different (c2=4.83, p<0.05).2.2 Similar with human tissue samples, FXR, CDX-2 and Muc2 were highly expressed in esophageal tissues of BE rats and low or not expressed in the controls.2.3 In the operation groups, simple bile acid reflux led to an 80% incidence of esophageal inflammation and mixed reflux led to a 90.5% incidence accordingly. There was no a significant difference(c2=0.04, p>0.05). However, in the non-operation groups, the incidences of esophageal inflammation in Group D and Group E were significantly different (c2=8.96, p<0.01). Moreover, the levels of IL-6 and TNF-a in two reflux operation groups (Group A and Group B) and Group D increased significantly.2.4 The levels of CDX-2 and Muc2 detected by Semi-quantitative RT-PCR in two reflux operation groups (Group A and Group B) and Group D increased significantly and the result was consistent with that obtained by immunohistochemistry. As the biomarkers of cell differentiation or proliferation, the levels of Bmp4, Kit 19 and Tff2 also increased significantly in above groups (p<0.05).ConclusionIn this study, we found the expression of FXR in BE tissue significantly increased compared with normal esophageal mucosa tissue. Moreover, the expression of CDX-2, Muc2 and COX-2 in BE tissue also increased and was positively correlated with the expression of FXR. We successfully established two reflux models and proved that bile acid but gastric acid was the key factor associated with BE. It is supposed that the inflammation induced by bile acid and its receptor may be associated with the development of BE, which is likely mediated by the overexpression of some key genes such as cytokines (IL-6 and TNF-a) as well as genes involving in cell differentiation or proliferation.