Podocyte damage and accumulation of advanced glycation end products (AGEs) are characteristics of diabetic nephropathy (DN). The pathophysiology of AGE-challenged podocytes, such as hypertrophy, apoptosis, and reduced cell migration, is closely related to the induction of the cell cycle inhibitor p27(Kip1) and to the inhibition of neuropilin 1 (NRP1). We have previously demonstrated that treatment with erythropoietin is associated with protective effects for podocytes in vitro. db/db mice with overt DN aged 15-16 wk were treated with either placebo, epoetin-β, or continuous erythropoietin receptor activator (CERA) for 2 wk. db/db mice compared with nondiabetic db/m control mice revealed the expected increases in body weight, blood glucose, albumin-to-creatinine ratio, and AGE accumulation. Whereas there were no differences in body weight, hyperglycemia and AGEs were observed among diabetic mice that received epoetin-β compared with CERA and placebo treatment, indicating that epoetin-β/CERA treatment does not interfere with the development of diabetes in this model. However, the albumin-to-creatinine ratio was significantly lower in db/db mice treated with epoetin-β or CERA. Furthermore, kidney weights in db/db mice were increased compared with db/m control mice, indicating renal hypertrophy, whereas the increase in renal weight in epoetin-β- or CERA-treated db/db mice was significantly lower than in placebo-treated control mice. Induction of p27(Kip1) and suppression of NRP1 were significantly reduced in the epoetin-β treatment group versus the CERA treatment group. Furthermore, erythropoietin treatment diminished the diabetes-induced podocyte loss. Together, independently from hematopoetic effects, epoetin-β or CERA treatment was associated with protective changes in DN, especially that NRP1 and p27(Kip1) expressions as well as numbers of podocytes returned to normal levels. Our data show, for the first time, that medication of overt DN with erythropoietin for a short time can ameliorate albuminuria and podocyte loss.

Epithelial-to-mesenchymal transition (EMT) is an important mechanism of renal tubulo-interstitial fibrosis in diabetic nephropathy (DN). Inducers of EMT, among others, are transforming growth factor-β1(TGF-β1) as well as extracellular collagens. In renal cells of diabetic mice and in kidneys of patients with DN, the expression of collagen VIII (gene: Col8α1/α2) is enhanced and characteristic features of DN in streptozotocin (STZ)-induced diabetic Col8α1/α2 knockout-(KO) mice are attenuated compared with diabetic wild-type mice. This study aimed to investigate whether collagen type VIII may influence the induction of EMT. DN was induced in wild-type and Col8α1/α2-KO mice using the established and widely accepted low-dose STZ model [treatment for 5 consecutive days (50 mg/kg)]. Healthy and diabetic mice were analyzed for changes in renal function and the expression of EMT-related genes and proteins. Renal morphology, fibrosis, and various EMT markers were studied in kidneys using immunohistological and molecular biological methods. Knockout of Col8α1/α2 attenuated albuminuria, extracellular matrix production, as well as fibrosis. Furthermore, the kidneys of diabetic Col8α1/α2-KO mice showed a marked reduction in interstitial myofibroblasts, and in tubular cells the inhibition of the expression of epithelial markers as well as the expression of typical mesenchymal markers was reduced. The present study demonstrates that in contrast to diabetic wild-type mice EMT-like changes were attenuated in diabetic Col8α1/α2-KO mice, which indicates that either collagen VIII may be one of the major inducers of EMT-like changes in kidneys of diabetic wild-type mice or/possibly the lack of Col8α1/α2 disrupts TGF-β1-induced EMT-like changes.

While females are less affected by non-diabetic kidney diseases compared to males, available data on sex differences in diabetic nephropathy (DN) are controversial. Although there is evidence for an imbalance of sex hormones in diabetes and hormone-dependent mechanisms in transforming growth factor β1 (TGF-β1) signaling, causes and consequences are still incompletely understood. Here we investigated the influence of sex hormones and sex-specific gene signatures in diabetes- and TGF-β1-induced renal damage using various complementary approaches (a db/db diabetes mouse model, ex vivo experiments on murine renal tissue, and experiments with a proximal tubular cell line TKPTS). Our results show that: (i) diabetes affects sex hormone concentrations and renal expression of their receptors in a sex-specific manner; (ii) sex, sex hormones and diabetic conditions influence differences in expression of TGF-β1, its receptor and bone morphogenetic protein 7 (BMP7); (iii) the sex and sex hormones, in combination with variable TGF-β1 doses, determine the net outcome in TGF-β1-induced expression of connective tissue growth factor (CTGF), a profibrotic cytokine. Altogether, these results suggest complex crosstalk between sex hormones, sex-dependent expression pattern and profibrotic signals for the precise course of DN development. Our data may help to better understand previous contradictory findings regarding sex differences in DN.

The activation of the receptor for advanced glycation end products (RAGE) is involved in the development of diabetic nephropathy. Analysis of protein phosphatase-1 indicated that advanced glycation end products did not affect its expression, but increased its phosphatase activity. Using differential display analysis we previously demonstrated that stimulation of RAGE in podocytes modulates the expression of numerous genes, among others nuclear inhibitor of protein phosphatase-1 (NIPP1). Here we found that silencing of NIPP1 induced podocyte hypertrophy, cell cycle arrest, and significantly increased protein phosphatase-1 activity. NIPP1 downregulation was associated with increased p27(Kip1) protein expression. Reporter assays revealed a transcriptional activation of nuclear factor-κB in podocytes after suppression of NIPP1. The protein level of NIPP1 was also significantly reduced in podocytes of diabetic mice. Blocking the RAGE in vivo by a soluble analog elevated the NIPP1 protein in podocytes of diabetic mice. Thus, activation of the RAGE by advanced glycation end products or other ligands suppresses NIPP1 expression in diabetic nephropathy, contributes to podocyte hypertrophy, and glomerular inflammation.

Extracellular matrix deposition during tubulointerstitial fibrosis (TIF), a central pathological process in patients with diabetic nephropathy (DN), is driven by locally activated, disease-relevant myofibroblasts. Myofibroblasts can arise from various cellular sources, e.g., tubular epithelial cells via a process named epithelial-to-mesenchymal transition (EMT). Transforming growth factor beta 1 (TGF-β1) and its downstream Smad signaling play a critical role in both TIF and EMT. Whereas Smad3 is one central mediator, the role of the other prominently expressed variant, Smad2, is not completely understood. In this study, we sought to analyze the role of renal Smad2 in the development of TIF and EMT during streptozotocin-induced DN by using a fibroblast-specific protein 1 (FSP1)-promotor-driven SMAD2 knockout mouse model with decreased tubular, endothelial, and interstitial Smad2 expression. In contrast to wild-type diabetic mice, diabetic SMAD2 knockout mice showed the following features: (1) significantly reduced DN and TIF (shown by KIM1 expression; periodic acid Schiff staining; collagen I and III, fibronectin, and connective tissue growth factor deposition); (2) significantly reduced tubular EMT-like changes (e.g., altered Snail1, E-cadherin, matrix metalloproteinase 2, and vimentin deposition); and (3) significantly decreased expression of myofibroblast markers (α-smooth muscle actin, FSP1). As one mechanism for the protection against diabetes-induced TIF and EMT, decreased Smad3 protein levels and, as a possible consequence, reduced TGF-β1 levels were observed in diabetic SMAD2 knockout mice. Our findings thus support the important role of Smad2 for pro-fibrotic TGF-β/Smad3 signaling in experimental DN.