Background. We previously reported that renal cortical cyclooxygenase (COX-2) expression increased following subtotal nephrectomy, and chronic treatment with a selective COX-2 inhibitor, SC58236, reduced proteinuria and retarded the development of glomerulosclerosis. The present studies were designed to examine the effects of COX-2 inhibition in a model of diabetic nephropathy. Methods. Rats were divided into three groups: control, diabetic (streptozotocin-induced diabetic animals with superimposed DOCA/salt hypertension; right nephrectomy and DOCA treatment), and treated (administration of the selective COX-2 inhibitor, SC58236, to a subset of diabetic/DOCA/salt rats). Insulin was administered to maintain blood glucose in the 200 to 300 mg/dL range. Results. Systolic blood pressure in the two diabetic groups was elevated within one week and remained elevated until sacrifice at six weeks (control, 108 +/- 2 mm Hg; diabetic, 158 +/- 4 mm Hg; treated, 156 +/- 7 mm Hg). When measured at six weeks, immunoreactive COX-2 expression in the renal cortex of the diabetic rats was 2.5 +/- 0.3-fold of control animals (N = 7). Immunohistochemical localization indicated increased expression in macula densa and surrounding cortical thick ascending limb of Henle (cTALH). The COX-2 inhibitor decreased COX-2 expression in diabetic rats to 1.3 +/- 0.1-fold control. In addition, SC58236 decreased expression of PAI-1 (diabetic vs. treated, 3.2 +/- 0.5 vs. 1.7 +/- 0.2-fold control, N 7, P < 0.05), vascular endothelial growth factor (VEGF; 2.0 +/- 0.2 vs. 1.2 +/- 0.2; N = 7, P < 0.05), fibronectin (2.4 +/- 0.3 to 1.3 +/- 0.1; N 7, P < 0.05) and transforming growth factor-P (TGF-beta; 2.1 +/- 0.2 vs. 1.3 +/- 0.2; N = 7, P < 0.05). Proteinuria at six weeks was decreased in the SC58236-treated rats (149 8 vs. 92 +/- 8 mg/24 h; N = 7, P < 0.01). The mesangial sclerosis index, defined as increases in extracellular matrix within the mesangial space, was determined at six weeks; the control group had an index of 0.06 +/- 0.01, the diabetic group was 2.7 +/- 0.04 and the treated group was 0.6 +/- 0.03 (P < 0.0001 compared to the diabetic group). Conclusions. These results suggest that in an experimental model of diabetes and hypertension, inhibition of COX-2 expression decreases potential mediators of glomerular and tubulointerstitial injury and also decreases biochemical, functional and structural markers of renal injury.
