Background. High glucose up-regulates the mesangial cell expression of p27(Kip1), an inhibitor of cyclin-dependent kinases/cyclin complexes. Previous in vitro studies using cultured mesangial cells from p27(Kip1-/-) mice demonstrated that these cells do not undergo high glucose-mediated cellular hypertrophy. Since glomerular hypertrophy is an early feature of diabetic nephropathy and may precede the development of glomerulosclerosis, interference with p27(Kip1) expression may attenuate diabetic nephropathy. However, it is unclear whether deletion of p27(Kip1) protects the kidneys of diabetic nephropathy in vivo. Methods. Type 1 diabetes mellitus was induced in p27(Kip1+/+), p27(Kip1+/-), and p27(Kip1-/-) mice by injection of streptozotocin (STZ). Mice were studied for 6 weeks. Animals injected with citrate buffer only served as controls. At the end of the experiments, urine was collected, albuminuria was determined with an enzyme-linked immunosorbent assay (ELISA), and blood glucose concentrations were measured. Kidneys were perfusion-fixed for quantitative morphologic analysis with glutaraldehyde and for immunohistochemical studies with formaldehyde. Glomerular cell number and volume were analyzed. Glomerulosclerosis, tubulointerstitial, and vascular damage indices were semiquantitatively assessed according to standard methodology. Quantitative glomerular parameters (cell numbers and volumes of endothelial, mesangial, and epithelial cells) were measured on semithin sections. Expression of transforming growth factor-beta 1 (TGF-beta 1), laminin, and collagen type IV were determined by immunohistochemical staining. Results. In contrast to animals only injected with citrate buffer, mice that received STZ developed hyperglycemia. There was no significant difference in the degree of hyperglycemia among p27(Kip1+/+), p27(Kip1+/-), and p27(Kip1-/-) mice. Diabetic p27(Kip1+/+), but not control p27(Kip1+/+) animals, developed albuminuria. Albuminuria was significantly reduced in diabetic p27(Kip1+/-) and more profoundly in p27(Kip1-/-) animals. Diabetic p27(Kip1+/+) mice revealed a significant increase in mean glomerular volume at 6 weeks. The volumes of mesangial and endothelial cells and podocytes all increased, whereas cell numbers were reduced, consistent with cell hypertrophy. Glomerular, endothelial, mesangial and podocyte hypertrophy were reduced in diabetic p27(Kip1+/-) and p27(Kip1-/-) animals. Diabetic p27(Kip1+/+) animals had significantly increased glomerulosclerosis, tubulointerstium, and vascular damage indices compared to nondiabetic p27(Kip1+/+) controls. Diabetic p27(Kip1-/-) mice exhibited significantly less structural damage than diabetic wild-type animals. Diabetic p27(Kip1+/-) animals revealed intermediate glomerulosclerosis, tubulointerstium, and vascular damage values. Immunohistological stainings demonstrated increases in TGF-beta 1, collagen type IV, and laminin expression in kidneys of diabetic p27(Kip1+/+) animals compared to nondiabetic p27(Kip1+/+) controls. Staining intensity for type IV collagen and laminin, but not for TGF-beta 1, was significantly lower in diabetic p27(Kip1-/-) mice. Conclusion. Deletion of p27(Kip1) attenuates the functional and morphologic features of diabetic nephropathy. Although deletion of p27(Kip1) abolished some parameters of diabetic glomerular hypertrophy, the significant reduction of TGF-beta 1 expression in the tubulointerstitium indicates that other protective mechanisms could be operative. The p27(Kip1) gene is haplo-insufficient because diabetic p27(Kip1)+/- mice exhibited an intermediate degree of functional and structural renal injury. Our data shows that p27(Kip1) plays an important role in diabetic nephropathy.
