TGF-beta is a common mediator for podocyte differentiation and apoptosis

Abstract

Podocytes, highly specialized cells in the kidney which regulate plasma filtration, are often injured and lost in glomerular disease. The underlying mechanisms for growth-arrest, differentiation, and depletion of podocytes during renal development and disease remain poorly understood, but may involve a factor that is typically expressed in developing/injured kidneys, transforming growth factor-beta (TGF-beta).;We used a unique system of conditionally-immortalized podocytes which can be stimulated to differentiate to investigate the roles that TGF-beta may play in the regulation of podocyte differentiation and disease. We found that these cells produce Tgf-beta2 at a low level, inducing G0/G1-arrest and differentiation through Smad3-dependent induction of the cyclin dependent kinase inhibitor p15Ink4b (Cdkn2b). However, high concentrations exogenous TGF-beta1 or TGF-beta2 alter these effects by suppressing Cdkn2b, inducing Gadd45beta, and activating proapoptotic p38 MAPK, leading to apoptosis. Thus, we identify distinct signaling profiles activated in a concentration-dependent manner by TGF-beta, suggesting a critical TGF-beta concentration threshold which switches to proapoptotic signaling profiles and apoptosis from podocyte maturation. Thus, TGF-beta may act as a common mediator for podocyte response to injury, and its concentration may ultimately specify the fate of the podocyte.;We used microarray technology to identify genes that are selectively expressed in differentiated and undifferentiated podocytes and mesangial cells, and to test whether glomerular cell-selective transcriptomes can be used for molecular classification in a mouse models. RNA samples derived from glomerular- or tubular-enriched tissue, primary mesangial cell cultures, and undifferentiated and differentiated podocytes were probed with Affymetrix Genechips(TM). Cell-selective transcripts were derived using current normalization and subtraction schemes. Podocyte-selective transcript sets, when applied to microarray data from the kidneys of a Tgf-beta-transgenic mouse model for progressive glomerulosclerosis, proved sufficient to classify animals according to phenotype. When cell type-selective transcriptomes were combined, classification by phenotype severity was achieved with 100% statistical support. These studies contribute to our understanding of how TGF-beta may regulate podocyte phenotype and provides candidate genes for downstream targets. In this way, we have increased our understanding of the mechanisms which underlie renal pathology and generated tools for future investigations.