A study of capillary pericyte viability on extracellular matrix produced by endothelial cells in high glucose

Aims/hypothesis. Thickening of the basement membrane and selective loss of pericytes occur early in diabetic retinopathy. As we showed previously that pericyte adhesion is impaired on extracellular matrix produced by endothelial cells in high hexose concentrations, we aimed to verify if altered adhesion could influence pericyte viability and replication. Methods. Conditioned extracellular matrices were obtained by growing human umbilical vein endothelial cells in media containing 28 mmol/l D-glucose, with or without the inhibitors of protein glycation thiamine or aminoguanidine, and D-galactose or L-glucose up to 28 mmol/l. Having removed the endothelium, bovine retinal pericytes were grown on these matrices and, in separate experiments, on laminin, fibronectin or type IV collagen. Pericyte viability and replication were measured by cell counts and DNA synthesis after 7 days, cell cycle traversal after 2 days and apoptosis after 18 h, 2 days and 7 days. Results. Pericyte counts and DNA synthesis were reduced on matrices produced in high D-glucose and D-galactose, whilst matrix obtained in L-glucose reduced DNA synthesis but not counts. Both thiamine and aminoguanidine corrected reduced pericyte viability when added to high D-glucose. Cell cycle and apoptosis were not affected by growing pericytes on different conditioned matrices. Laminin, fibronectin and type IV collagen did not modify pericyte replication. Conclusions/interpretations. Reduced pericyte counts could depend on impaired initial adhesion to the extracellular matrix produced by endothelium in high hexose concentrations, rather than impaired replication or viability. Altered cell-matrix interactions might facilitate pericyte dropout in diabetic retinopathy, independently of the effects of high glucose on pericyte replication.