Cyclooxygenase Inhibitors Affect Recovery of Ischemic-injured Jejunum

Abstract

he cyclooxygenase (COX) enzyme produces prostaglandins. There are 3 isoforms of COX (including COX-1 and COX-2). As prostaglandins are involved in intestinal repair, the objective of this study was to examine the effects of COX inhibitors on healing of jejunal mucosa following ischemia, with the following specific aims. Aim 1: Assess effects of in vivo administration of flunixin (non-selective inhibitor) and etodolac (COX-2 preferential inhibitor) on barrier function of ischemic-injured equine jejunum after 18hrs recovery. Aim 2: Assess effects of in vitro treatment with a selective COX-2 inhibitor, deracoxib on the permeability of ischemic equine jejunum. Aim 3: Assess effects of ischemia and flunixin on absorption of lipopolysaccharide (LPS) in vitro. Aim 4: Examine the role of COX-2 in jejunal recovery using COX-2 knockout mice. Ussing chambers were used to assess the barrier function of ischemic and normal jejunal mucosa when exposed to the different inhibitors mentioned above. Transepithelial electrical resistance (TER), a sensitive index of barrier function, was calculated and the permeability of the mucosa assessed using mannitol. The amount of epithelial denudation following ischemia was determined and compared between treatments. Levels of expression of the COX-1 and ?2 enzymes before and after ischemia were assessed by Western blot. For aim 3, permeability to lipopolysaccharide (LPS) was measured and the passage of fluorescent-labeled LPS into mucosa assessed with histology. Statistical significance level for all tests was chosen at p < 0.05. Both flunixin and etodolac treatment retarded in vivo recovery of jejunal barrier function after ischemia. These drugs act on the paracellular space as treatments did not alter epithelial denudation or restitution. Normal equine jejunal mucosa expressed both COX-1 and COX-2 and ischemia upregulated both isoforms. The selective COX-2 inhibitor deracoxib also adversely affected mucosal recovery in vitro but to a lesser extent; deracoxib did not significantly increase permeability to mannitol. The increased permeability of mucosa to mannitol after flunixin treatment did not reflect increased LPS absorption despite a trend toward this finding. Ischemia alone increased the absorption of LPS. Although the precise route of LPS across the mucosa was not determined, it was postulated to be mostly through the defects from epithelial cell loss. Attempts to clarify the role of COX-2 in mucosal recovery from ischemia using COX-2 knockout mice produced conflicting results. COX-1 inhibition increased permeability to mannitol in both COX-2 knockout and wild-type mice. There was a trend towards increased permeability due to COX-2 inhibition in wild-type mice. For knockout mice, the COX-2 inhibitor also unexpectedly increased permeability. COX-1 protein was present in all mice and COX-2 was present in normal mucosa of wild-type mice. Both isoforms were not upregulated by ischemia. Untreated horses recovered baseline levels of TER and permeability by 18hrs after ischemia. Both flunixin and etodolac did not allow sufficient prostaglandin production for recovery. In vitro treatment with the selective COX-2 inhibitor deracoxib also adversely affected recovery but to a lesser extent than the other drugs tested. The increased permeability due to flunixin did not exacerbate LPS absorption in this model. The clinical significance of the effects of COX inhibitors on permeability will require further in vivo testing. COX-1 and '2 are both present in normal jejunal mucosa from both horses and mice, thus it is possible that the action of both enzymes is required for complete recovery from ischemia.

Description

Keywords

cyclooxygenase, NSAID, intestine, equine, ischemia

Citation

Degree

PhD

Discipline

Physiology

Collections