Immunohistochemical and functional correlations of renal cyclooxygenase-2 in experimental diabetes

Abstract

Prostaglandins (PGs) generated by the enzyme cyclooxygenase (COX) have been implicated in the pathological renal hemodynamics and structural alterations in diabetes mellitus, but the role of individual COX isoenzymes in diabetic nephropathy remains unknown. We explored COX-1 and COX-2 expression and hemodynamic responses to the COX-1 inhibitor valeryl salicylate (VS) or the COX-2 inhibitor NS398 in moderately hyperglycemic, streptozotocin-diabetic (D) and control (C) rats. Immunoreactive COX-2 was increased in D rats compared with C rats and normalized by improved glycemic control. Acute systemic administration of NS398 induced no significant changes in mean arterial pressure and renal plasma flow in either C or D rats but reduced glomerular filtration rate in D rats, resulting in a decrease in filtration fraction. VS had no effect on renal hemodynamics in D rats. Both inhibitors decreased urinary excretion of PGE(2). However, only NS398 reduced excretion of thromboxane A(2). In conclusion, we documented an increase in renal cortical COX-2 protein expression associated with a different renal hemodynamic response to selective systemic COX-2 inhibition in D as compared with C animals, indicating a role of COX-2-derived PG in pathological renal hemodynamic changes in diabetes.

Figure 1

Expression of COX-2 and COX-1 immunoreactive proteins, shown as COX/actin ratio, in renal cortex of control (C; n = 5), moderately hyperglycemic diabetic rats (DM; n = 5), and diabetic rats on intensive insulin treatment (DM-IT; n = 4). Moderately hyperglycemic diabetic rats demonstrated an increase in cortical COX-2 expression as compared with control rats. This was reversed by improved metabolic control in diabetic rats on intensive insulin treatment. No differences in COX-1 expression were found between control and diabetic rats. ^AP < 0.05.

Figure 2

Localization of immunoreactive COX-2 in MD (a and b, arrows) and in non-MD tubular cells (c and d, arrows) of control (a and c) and diabetic rats (b and d). In control rats, COX-2 immunoreactivity was present in occasional cells. In contrast, diabetic kidneys demonstrated immunostaining of groups of cells both in MD regions and non-MD cells of TALH. COX-2 was also detected in occasional podocytes (arrowheads) both in control (a) and diabetic glomeruli (b). Findings in diabetic rats on intensive insulin treatment resembled those in control rats (e). ×400.

Figure 3

Effects of acute selective COX-2 inhibition with NS398 (Period I, 0.3 mg/kg; Period II, 1.5 mg/kg) or vehicle on GFR, RPF, FF, and RVR in control and diabetic rats. As shown in the left panels, NS398 induced no changes in renal hemodynamics in control rats. By contrast, diabetic rats responded to COX-2 inhibition with significant decreases in GFR and FF. Right panels depict disparate renal hemodynamic responses induced by COX-2 and COX-1 inhibitors in diabetic rats. Unlike NS398, administration of VS (Period I, 6 mg/kg; Period II, 30 mg/kg) did not influence GFR. Open squares, control vehicle-treated rats; filled diamonds, control rats treated with NS398; open circles, diabetic vehicle-treated rats; filled triangles, diabetic rats treated with NS398; open diamonds, diabetic rats treated with VS. ^AP < 0.05 vs. basal; ^BP < 0.001 vs. basal; ^CP < 0.05 vs. diabetic rats treated with VS.

Figure 4

Effects of COX-2 inhibitor NS398 (1.5 mg/kg) and COX-1 inhibitor VS (VS, 30 mg/kg) on urinary excretion of PGE₂ and TxB₂ in diabetic rats. Diabetic rats demonstrated increased baseline excretion of PGE₂ (P < 0.01) and TxB₂ (P < 0.05) as compared with control rats. Both inhibitors induced significant decreases in urinary PGE₂ excretion (P < 0.05). However, only treatment with NS398 was associated with a reduction of TxB₂ (P < 0.05). D-VE, diabetic rats treated with vehicle; D-NS398, diabetic rats treated with NS398; D-VS, diabetic rats treated with VS; Control, control rats administered vehicle. ^AP < 0.05 vs. baseline; ^BP < 0.05 vs. control; ^CP < 0.01 vs. control.