Glucose and TGFbeta2 modulate the viability of cultured human retinal pericytes and their VEGF release.

PURPOSE: Determine the effects of glucose and exogenous TGFbeta2 on viability and VEGF release by human retinal pericytes (HRP). METHODS: Human retinal pericytes (HRP) were cultured in 5 mM (physiologic) or high (18 mM) glucose with or without added TGFbeta2. Viable cells were counted; TGFbeta2 and VEGF in the conditioned media (CM) were measured by ELISA. RESULTS: High glucose significantly reduced viable cell number and increased the levels of TGFbeta2 and VEGF. TGFbeta2 caused a significant dose-dependent effect on viable cell number and on the level of VEGF secreted into the CM by HRP in physiologic glucose, decreasing viable cell number, and increasing VEGF release per 1000 cells at a low concentration (0.1 ng/ml) and increasing viable cell number and decreasing VEGF release per 1000 cells at higher concentrations (1.0 and 10 ng/ml). TGFbeta2 affected neither parameter in high glucose. CONCLUSIONS: Elevated glucose decreased HRP viability and modulated changes in TGFbeta2 and VEGF release. This suggests a novel mechanism for HRP dropout in diabetic retinopathy.

Bone morphogenetic protein-4 enhances vascular endothelial growth factor secretion by human retinal pigment epithelial cells.

Retinal pigment epithelial (RPE) cells secrete vascular endothelial growth factor (VEGF), a cytokine known to promote angiogenesis. Results from RNase protection assays (RPAs) show that RPE from non-diabetic human donors and from adult retinal pigment epithelium-19 (ARPE-19) cells expressed significant bone morphogenetic protein-4 (BMP-4) message. In addition, ARPE-19 cells cultured in high glucose (25 mM), compared to those in physiological glucose (5.5 mM) released significantly more BMP-4 into the conditioned media (CM). However, the effect of BMP-4 on the release of VEGF by ARPE-19 cells has not been studied. Accordingly, ARPE-19 cells were treated with BMP-4 to determine VEGF secretion. BMP-4 and VEGF levels in the CM and cell lysates were measured by enzyme-linked immunosorbent assay (ELISA). Cells treated with exogenous BMP-4 had higher VEGF in the CM and this treatment effect was dose- and time-dependent, while cell lysates had low levels of VEGF. Addition of cycloheximide (CHX) or actinomycin-D (ACT) significantly reduced VEGF secretion from cells treated with BMP-4, suggesting that the BMP-4-induced secretion of VEGF requires new RNA and protein synthesis. Our results suggest that BMP-4 may play a role in the regulation of ocular angiogenesis associated with diabetic retinopathy (DR) by stimulating VEGF release from RPE cells.