Background: Diabetic nephropathy is characterized by glomerular hypertrophy. We have recently shown that experimental diabetes mellitus is associated with an increase in glomerular expression of the cyclin kinase inhibitor p21WAF1/CIP1 (p21). Furthermore, in vitro glucose-induced mesangial cell hypertrophy is also associated with an up-regulated expression of p21. In this study, we tested the hypothesis that p21 mediates diabetic glomerular hypertrophy in vivo.
Methods: Experimental diabetes mellitus was induced by streptozotocin in mice in which p21 was genetically deleted (p21 -/-) and in wild-type mice (p21 +/+). Kidney biopsies were obtained from diabetic and control (citrate injected) p21 +/+ and p21 -/- mice at day 60. The tissue was used for morphologic studies of glomerular size (measured by computer image-analysis system), glomerular cellularity (cell count), glomerular matrix expansion (silver stain), apoptosis (TUNEL), and expression of transforming growth factor-beta1 (TGF-beta1) by in situ hybridization.
Results: The glomerular tuft area increased 11.21% in diabetic p21 +/+ mice at day 60 compared with control (3329.98 +/- 244.05 micrometer(2) vs. 2994. 39 +/- 176.22 micrometer(2), P = 0.03), and the glomerular cell count did not change in diabetic p21 +/+ mice at day 60 compared with the control. These findings are consistent with glomerular hypertrophy. In contrast, the glomerular tuft area did not increase in diabetic p21 -/- mice at day 60 compared with the control (3544.15 +/- 826.49 vs. 3449.15 +/- 109.65, P = 0.82), nor was there an increase in glomerular cell count (41.41 +/- 13.18 vs. 46.95 +/- 3.00, P = 0.43). Diabetic p21 +/+ mice, but not p21 -/- mice, developed an increase in proteinuria at day 60 compared with the control. Tubular cell proliferation, measured by proliferating cell nuclear antigen immunostaining, was increased in both diabetic p21 +/+ (2.1-fold) and p21 -/- (7.61-fold) mice compared with controls. Glomerular cell apoptosis did not increase in diabetic mice. Although glomerular TGF-beta1 mRNA levels increased in both strains of diabetic mice at day 60, the glomerular matrix did not expand.
Conclusions: Hyperglycemia was associated with glomerular hypertrophy in p21 +/+ mice. Despite the increase in TGF-beta1 mRNA, diabetic p21 -/- mice did not develop glomerular hypertrophy, providing evidence that the cyclin kinase inhibitor p21 may be required for diabetic glomerular hypertrophy induced by TGF-beta1. The loss of p21 increases tubular but not glomerular cell proliferation in diabetic nephropathy. The absence of glomerular hypertrophy appears protective of renal function in diabetic mice.