Effect of papillotoxic agents on expression of cyclooxygenase isoforms in the rat kidney

Inhibition of renal vasodilatory prostaglandins (PGs) and secondary ischemia due to inhibition of cyclooxygenase (COX) activity has been suggested as a possible mechanism for development of analgesic-related renal papillary necrosis (RPN) in rats. Recently, it has been shown that COX exists in two related but unique isoforms, COX-1 and COX-2. It is unclear what potential roles these isoforms play in the maintenance of blood flow in the renal papilla or genesis of RPN. We evaluated the effect of 2 papillotoxic agents, including a nonsteroidal anti-inflammatory drug, indomethacin, and a chemical agent, 2-bromoethanamine hydrobromide (2-BEA), on COX-1 and COX-2 in the renal papilla as a means of assessing what changes occur in the expression of these isoforms during the development of RPN. Female Wistar rats approximately 10-17 wk old were treated with either indomethacin (75 mg/kg, single dose, or 10 mg/kg/day for 5 days) or 2-BEA (100 mg/kg/day for 4 days) to create lesions of RPN. In this study, a single 75-mg/kg dose of indomethacin did not cause light microscopic changes of RPN. However, RPN was observed in animals administered indomethacin at 10 mg/kg/day for 1 wk or 2-BEA for 5 days. The immunohistochemical analyses of kidneys showed that both COX-1 and COX-2 were present in the renal papilla of control rats. In animals treated with indomethacin (75 mg/kg), a slight to moderate decrease in both isoforms was observed in essentially normal renal papillary cells within 2 hr, that was followed by an increase in COX-2 immunoreactivity in the renal papilla, macula densa, and thick ascending limbs (both 10- and 75-mg/kg animals). This COX-2 immunoreactivity was greatest in animals with concomitant indomethacin-induced gastrointestinal injury, suggesting a possible role of inflammatory cytokines in COX-2 induction. No changes in the expression of COX isoforms in the intact papilla occurred as a result of 2-BEA; however, cells undergoing degeneration and necrosis lost immunoreactivity to both COX isoforms. The possible mechanism that leads to an initial decrease in COX immunoreactivity in indomethacin-treated animals is not known; however, a reversible ultrastructural change in the papillary cells cannot be ruled out. This decrease in COX isoforms in the renal papilla may contribute to the development of RPN through the loss of vasodilatory PGs.