COX-2 but not mPGES-1 contributes to renal PGE2 induction and diabetic proteinuria in mice with type-1 diabetes.

Prostaglandin E2 (PGE2) has been implicated to play a pathogenic role in diabetic nephropathy (DN) but its source remains unlcear. To elucidate whether mPGES-1, the best characterized PGE2 synthase, was involved in the development of DN, we examined the renal phenotype of mPGES-1 KO mice subjected to STZ-induced type-1 diabetes. After STZ treatment, mPGES-1 WT and KO mice presented the similar onset of diabetes as shown by similar elevation of blood glucose. Meanwhile, both genotypes of mice exhibited similar increases of urinary and renal PGE2 production. In parallel with this comparable diabetic status, the kidney injury indices including the urinary albumin excretion, kidney weight and the kidney histology (PAS staining) did not show any difference between the two genotypes. By Western-blotting and quantitative qRT-PCR, mPGES-1, mPGES-2, cPGES and 15-hydroxyprostaglandin dehydrogenase (15-PGDH) remain unaltered following six weeks of diabetes. Finally, a selective COX-2 inhibitor celecoxib (50 mg/kg/day) was applied to the STZ-treated KO mice, which resulted in significant reduction of urinary albumin excretion (KO/STZ: 141.5±38.4 vs. KO/STZ + Celebrex: 48.7±20.8 ug/24 h, p<0.05) and the blockade of renal PGE2 induction (kidney: KO/STZ: 588.7±89.2 vs. KO/STZ + Celebrex: 340.8±58.7 ug/24 h, p<0.05; urine: KO/STZ 1667.6±421.4 vs. KO/STZ + Celebrex 813.6±199.9 pg/24 h, p<0.05), without affecting the blood glucose levels and urine volume. Taken together, our data suggests that an as yet unidentified prostaglanind E synthase but not mPGES-1 may couple with COX-2 to mediate increased renal PGE2 sythsesis in DN.