Barrett's esophagus (BE) represents a substantial health care burden because it has a high frequency of progression to dysplasia and Barrett's-associated adenocarcinoma (BAA). A major reason for this problem is that we currently have limited approaches to prevent development of BAA, and of BE itself. If we could generate direct evidence that specific molecular pathways have a causative or preventive role in these events it would be of obvious benefit. Cyclooxygenase (COX)-2 and inducible nitric oxide synthase (iNOS), the enzymes responsible for the high-output production of nitric oxide (NO) and prostaglandins, respectively, are both implicated in dysregulation of epithelial cell growth and in GI carcinogenesis. Recent evidence suggests that arginase, the endogenous competitive inhibitor of iNOS, has important biological effects by blocking NO synthesis, but also has direct effects via diversion of L-arginine to other pathways, such as polyamine synthesis. Our data in human patients shows frequent and abundant expression of iNOS and COX-2 in BE and BAA, and we have recently determined that the arginase II enzyme (argll) is significantly downregulated in BE compared with the proximal esophagus in BE patients. We hypothesize that iNOS and COX-2 play a causal role in development of BE and in progression to BAA, while argll protects against these events. The PI proposes to use the R21 mechanism to apply his expertise about these enzymes in the GI mucosa to new studies in animal models of BE.
In Aim 1, we will employ a surgical model of gastroduodenal-esophageal reflux created by esophagojejunostomy, which has been shown to produce BE in rats and has been recently applied to mice. We will develop this model in our lab and determine if COX-2 or iNOS deletion protects against, and if argll deletion enhances, development of BE and carcinoma. We will compare macroscopic and histologic evidence of BE and cancer, gene expression of COX-2, iNOS, and arginases, histologic evidence of apoptosis and proliferation in: A. wild-type (WT) vs. COX-2-/- mice, B. WT vs. iNOS-/- mice, and C. WT vs. argll-/-mice.
In Aim 2, we will use injection of the lower esophageal sphicter (LES) with botulinum toxin (BoTx) to induce gastroesophageal reflux by inhibiting LES function. We will study macroscopic and histologic evidence of esophagitis, BE and BAA, gene expression, apoptosis and proliferation in A. wild-type (WT) vs. COX-2-/- mice, B. W'I- vs. iNOS -/-mice, and C. WT vs. argll -/- mice.
