Diabetic retinopathy is the main cause of blindness in Western countries yet the molecular basis of its pathogenesis remains poorly understood. In the present study, we have examined the effects of hyperglycaemia and the electrophilic agent diethylmaleate (DEM) on L-cystine transport, intracellular glutathione (GSH) levels and activation of p42/p44 and p38 mitogen-activated protein kinase (MAPK) in bovine retinal endothelial cells and pericytes.

Retinal endothelial cell (BREC) and retinal pericytes (BRP) were isolated from ~20 bovine eyes obtained from an abattoir and cultured in Minimal Eagle’s Medium containing either 10 % horse or fetal calf serum, respectively. Cells were characterized as endothelial cells or pericytes using antibodies against von Willebrand factor or α-actin, respectively. Confluent cell monolayers were exposed to normal D-glucose (5.6 mM) and elevated D-glucose (15 or 25 mM) for defined time periods (0-60 min and 24-72 h). D-Mannitol (9.4 mM + 5.6 mM D-glucose) served as an osmotic control. Intracellular GSH levels were determined using the glutathione-S-transferase (GST) catalysis reaction. Influx of L-[¹⁴C]cystine (50 µM) was measured over 2 min in cells exposed to normal or elevated D-glucose or DEM. Changes in the phosphorylation of p42/44MAPK and p38MAPK were investigated by western blot analysis.

Intracellular GSH levels in BREC (43 ± 4 nmol (mg protein)^-1, mean ± S.E.M., n = 3) and BRP (15 ± 1 nmol (mg protein)^-1, n = 4) were not altered significantly following treatment with elevated D-glucose for 24 h. Exposure of BREC to 15 mM D-glucose for up to 24 h had no significant effect on basal L-cystine transport. Diethylmaleate (100 µM, 24 h), an agent known to conjugate intracellular GSH, resulted in adaptive increases in L-cystine transport (413 ± 58 vs. 661 ± 76 pmol (µg protein)^-1 min^-1, n = 3-4, Student’s unpaired t test, P < 0.05) and GSH (43 ± 4 vs. 60 ± 4 nmol (mg protein)^-1, n = 5, P < 0.05). Exposure of cells to 15 mM D-glucose or D-mannitol for 5 min to 1 h had no significant effect on phosphorylation of p42/p44MAPK in either BREC or BRP, whereas exposure of BRP to 25 mM D-glucose for 48 h resulted in activation of p38MAPK.

Hyperglycaemia had negligible effects on L-cystine transport and GSH levels suggesting that, under our experimental conditions, elevated D-glucose caused minimal oxidative stress. In contrast, DEM caused a classical adaptive increase in L-cystine transport (most likely via anionic amino acid transporter xCT, see Sato et al. 1999) and GSH levels. Further studies are required to establish whether activation of p38MAPK in response to prolonged exposure to elevated D-glucose modulates L-cystine transport activity and intracellular GSH levels.

This work was supported by K. C. Wong Scholarship.