Preclinical models for the study of Barrett's carcinogenesis

Inhibition of BMP2 and BMP4 Represses Barrett's Esophagus While Enhancing the Regeneration of Squamous Epithelium in Preclinical Models

BACKGROUND & AIMS

Barrett's esophagus is considered to be a metaplastic lesion that predisposes for esophageal adenocarcinoma. Development of Barrett's esophagus is considered to be driven by sonic hedgehog mediated bone morphogenetic protein (BMP) signaling. We aimed to investigate in preclinical in vivo models whether targeting canonical BMP signaling could be an effective treatment for Barrett's esophagus.

METHODS AND RESULTS

Selective inhibition of BMP2 and BMP4 within an in vivo organoid model of Barrett's esophagus inhibited development of columnar Barrett's cells, while favoring expansion of squamous cells. Silencing of noggin, a natural antagonist of BMP2, BMP4, and BMP7, in a conditional knockout mouse model induced expansion of a Barrett's-like neo-columnar epithelium from multi-lineage glands. Conversely, in this model specific inhibition of BMP2 and BMP4 led to the development of a neo-squamous lineage. In an ablation model, inhibition of BMP2 and BMP4 resulted in the regeneration of neo-squamous epithelium after the cryoablation of columnar epithelium at the squamocolumnar junction. Through lineage tracing the generation of the neo-squamous mucosa was found to originate from K5+ progenitor squamous cells.

CONCLUSIONS

Here we demonstrate that specific inhibitors of BMP2 and BMP4 attenuate the development of Barrett's columnar epithelium, providing a novel potential strategy for the treatment of Barrett's esophagus and the prevention of esophageal adenocarcinoma.

Novel In Vivo Mouse Cryoablation Model to Explore Unique Therapeutic Approaches for Premalignant Columnar Lesions

Patients with epithelial metaplasias have an increased risk of developing malignancies. In Barrett’s esophagus, neo-columnar epithelium develops proximal to the squamous-columnar junction (SCJ) in the esophagus as the result of prolonged exposure to bile and acid reflux. Patients require lifetime periodic surveillance, due to lack of effective eradication therapies. The shortage of innovative treatment options is mostly attributable to the paucity of adequate in vivo models of neo-columnar epithelium regeneration. This protocol describes the generation of a cryoablation model to study regeneration of neo-epithelia at the SCJ. Cryoablation of the columnar and squamous mucosa at the SCJ was achieved through local application of liquid N₂O in wild-type and reporter mice in combination with acid suppression. Acid suppression alone, showed restoration of the SCJ with normal histological features of both the neo-columnar and neo-squamous epithelium within 14 days. As a proof of principle, mice were treated with mNoggin, an inhibitor of bone morphogenetic proteins (BMPs), which are involved in the development of columnar epithelia. Local application of mNoggin to the ablated area at the SCJ significantly reduced the development of the neo-columnar mucosa. Although this model does not faithfully recapitulate the exact characteristics of Barrett’s esophagus, it is a well-suited tool to study the mechanisms of therapeutic inhibition of neo-columnar regeneration. It therefore represents an efficient and easy platform to test novel pharmacological therapies for treatment of neo-epithelial lesions at the SCJ.

Chronic High-Fat Diet Induces Early Barrett's Esophagus in Mice through Lipidome Remodeling

Esophageal adenocarcinoma (EAC) incidence has been rapidly increasing, potentially associated with the prevalence of the risk factors gastroesophageal reflux disease (GERD), obesity, high-fat diet (HFD), and the precursor condition Barrett’s esophagus (BE). EAC development occurs over several years, with stepwise changes of the squamous esophageal epithelium, through cardiac metaplasia, to BE, and then EAC. To establish the roles of GERD and HFD in initiating BE, we developed a dietary intervention model in C57/BL6 mice using experimental HFD and GERD (0.2% deoxycholic acid, DCA, in drinking water), and then analyzed the gastroesophageal junction tissue lipidome and microbiome to reveal potential mechanisms. Chronic (9 months) HFD alone induced esophageal inflammation and metaplasia, the first steps in BE/EAC pathogenesis. While 0.2% deoxycholic acid (DCA) alone had no effect on esophageal morphology, it synergized with HFD to increase inflammation severity and metaplasia length, potentially via increased microbiome diversity. Furthermore, we identify a tissue lipid signature for inflammation and metaplasia, which is characterized by elevated very-long-chain ceramides and reduced lysophospholipids. In summary, we report a non-transgenic mouse model, and a tissue lipid signature for early BE. Validation of the lipid signature in human patient cohorts could pave the way for specific dietary strategies to reduce the risk of BE in high-risk individuals.