Renal COX-2 and prostaglandin release in diabetes: role of no synthase and aldose reductase

John Quilley and Yu-Jung Chen, Department of Pharmacology, New York Medical College, Valhalla, NY 10595, USA.

In the streptozotocin (STZ)-diabetic rat, endoperoxide-mediated renal vasoconstrictor responses to arachidonic acid (AA) are markedly enhanced and associated with increased prostaglandin (PG) release and induction of COX-2. Experiments with tempol, a SOD mimetic, indicated that superoxide (O 2 *) contributes to the induction of COX-2 and increased PG release (Li et al., 2005). O 2 * can be formed by NO synthase (NOS) activity and O 2 * can react with NO to form peroxynitrite (ONOO-) which can also induce COX-2. We have shown that renal nitrotyrosine expression, a marker for ONOO - is increased 3-fold in diabetes (p<0.01). In addition to NOS, the activity of aldose reductase (AR) is increased in diabetes and may lead to O 2* generation. To address the potential roles of NOS and AR in the induction of COX-2 and the increased release of PGs, we treated diabetic rats with either L-NAME (100mg/kg/day) or zopolrestat (50mg/kg/day), respectively, and compared these to untreated diabetic rats and treated and untreated non-diabetic rats (n=4-6/group). Diabetes was induced in male Wistar rats, 180-200g, with STZ, 70mg/kg i.v. After 3-4 weeks treatment, the right kidney was perfused to determine 6-KetoPGF1α and PGE2 release in response to AA and the left kidney was used for COX-2 protein expression.

In the diabetic rat kidney, vasoconstrictor responses to 1μg AA were greatly enhanced versus control, 102 ±43 mmHg compared to 8 ±2 mmHg, respectively (P<0.01) and release of 6-KetoPGF1α was increased by 5.3 ±0.9 ng/min compared to 2.3 ±0.3 ng/min (p<0.01). L-NAME treatment increased blood pressure in both diabetic and control rats from 126 ±3 and 130 ±3 mmHg to 140 ±3 and 150 ±9 mmHg, respectively, and enhanced vasoconstrictor responses to AA; the increase was relatively greater in non-diabetic rats, 16-fold compared to <2-fold for diabetic rats so that the difference between diabetic and control rats was eliminated. The increase in AA-stimulated 6-KetoPGF 1 a release in diabetic rats was reduced to control levels by L-NAME (3.2 ±0.5ng/min), which prevented the 2-fold increase in COX-2 expression.

In diabetic rats, renal sorbitol levels were 13.1 ±3.4 versus 2.3 ±1.0μg/g for control (p<0.01) and were reduced to 4.1 ±0.9μg/g by zopolrestat (p<0.01), showing effective inhibition of AR activity. Zopolrestat treatment did not affect renal vasoconstrictor responses to AA in either diabetic or control rats but reduced the AA-stimulated increase in PGE2 release in diabetic rats from 8.1 ±3.7ng/min to 3.8 ±1.2ng/min.

These results, coupled with those from the earlier study with tempol which prevented the induction of COX-2 and the associated increase in PG release in diabetes, are consistent with the idea that NOS and AR contribute to the generation of O2*, which, directly or indirectly, causes up-regulation of COX-2.

Li et al. J Pharmacol Exp Ther 314: 818-824, 2005.