Effect of high glucose on glycosaminoglycans in cultured retinal endothelial cells and rat retina

Abstract

Introduction: The endothelial glycocalyx regulates vascular permeability, inflammation, and coagulation, and acts as a mechanosensor. The loss of glycocalyx can cause endothelial injury and contribute to several microvascular complications and, therefore, may promote diabetic retinopathy. Studies have shown a partial loss of retinal glycocalyx in diabetes, but with few molecular details of the changes in glycosaminoglycan (GAG) composition. Therefore, the purpose of our study was to investigate the effect of hyperglycemia on GAGs of the retinal endothelial glycocalyx.

Methods: GAGs were isolated from rat retinal microvascular endothelial cells (RRMECs), media, and retinas, followed by liquid chromatography-mass spectrometry assays. Quantitative real-time polymerase chain reaction was used to study mRNA transcripts of the enzymes involved in GAG biosynthesis.

Results and conclusions: Hyperglycemia significantly increased the shedding of heparan sulfate (HS), chondroitin sulfate (CS), and hyaluronic acid (HA). There were no changes to the levels of HS in RRMEC monolayers grown in high-glucose media, but the levels of CS and HA decreased dramatically. Similarly, while HA decreased in the retinas of diabetic rats, the total GAG and CS levels increased. Hyperglycemia in RRMECs caused a significant increase in the mRNA levels of the enzymes involved in GAG biosynthesis (including EXTL-1,2,3, EXT-1,2, ChSY-1,3, and HAS-2,3), with these increases potentially being compensatory responses to overall glycocalyx loss. Both RRMECs and retinas of diabetic rats exhibited glucose-induced alterations in the disaccharide compositions and sulfation of HS and CS, with the changes in sulfation including N,6-O-sulfation on HS and 4-O-sulfation on CS.

Keywords: diabetic retinopathy; endothelial glycocalyx; glycosaminoglycans; hyperglycemia.

Fig. 1

Effects of hyperglycemia on glycosaminoglycans in cultured retinal endothelial cells. Cells treated with high glucose for 6 days exhibited a) a significant reduction in total amount of GAGs (n = 5), b) no changes in HS (n = 5), and a loss of c) CS (n = 5) and d) HA (n = 4) in comparison to cells grown under normal glucose media. Mannitol treatment caused a significant increase in total GAGs, HS, CS, and HA levels. Disaccharide composition of e) HS and f) CS in RRMECs (n = 5) under normal glucose, high-glucose, and mannitol media. Data were analyzing using 1-way ANOVA and are represented as mean ± SEM. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, ns = non-significant.

Fig. 2

Effects of hyperglycemia on shedding of glycosaminoglycans from RRMECs. The change in media concentration of a) total amount of GAGs (n = 5), b) HS (n = 5), c) CS (n = 5), and d) HA (n = 5) is shown for all groups compared with normal glucose conditions (X-normal glucose). The percentage change in disaccharide composition of e) HS and f) CS in culture media (n = 5) for all groups is compared with normal glucose conditions. Data were analyzed using 1-way ANOVA and are represented as mean ± SEM.

Fig. 3

Effects of hyperglycemia on enzymes responsible for the synthesis of GAGs. RRMECs grown in high-glucose media expressed a significant increase in mRNA transcripts of a) EXTL1, b) EXTL2, c) EXTL3, d) EXT1, e) EXT2, f) ChSY1, h) ChSY3, j) HAS2, and k) HAS3, no change in g) ChPF, and a significant decrease in mRNA levels of i) HAS1. Mannitol significantly decreased the mRNA levels of e) ChPF and i) HAS3. Data were analyzed using 1-way ANOVA and are represented as mean ± SEM (n = 5–6).

Fig. 4

a) Body weight and b) blood glucose concentrations of nondiabetic and STZ-induced diabetic rats. Data are presented as means ± SEM (n = 9–10).

Fig. 5

Effects of diabetes on GAGs in retina. Eight weeks of diabetes caused a significant increase in a) the total amount of GAGs and c) CS in the retina. However, b) diabetes had no effect on retinal HS levels but significantly decreased the level of HA measured by d) LC–MS and e) ELISA. Disaccharide composition of f) HS and g) CS in the retina. Data were analyzed using a 2-tailed Student’s t-test and are represented as mean ± SEM; n = 9–10 for the LC–MS analysis and n = 3–4 for ELISA.

Fig. 6

Proportion of glycosaminoglycans in RRMECs and rat retinas. The dominant GAG in a) RRMECs and b) rat retina was CS, followed by HS and HA. a) Hyperglycemic conditions in RRMECs increased the amount of HS and decreased that of CS, while the proportion was not affected in b) diabetic rat retinas. Data were analyzed using 2-way ANOVA followed by Tukey’s multiple comparison test and are represented as mean ± SEM; n = 5 for RRMECs and n = 9–10 for retina analysis.

Fig. 7

Summary of hyperglycemia-induced changes observed on the expression of GAGs in RRMECs and rat retina. Hyperglycemia causes a significant increase in the shedding of HS, CS, and HA and a loss of CS and HA from RRMECs. Similarly, HA expression is significantly decreased in the retinas of diabetic rats compared with those of controls; however, CS levels are significantly increased. Hyperglycemia causes a significant increase in mRNA levels of several enzymes responsible for GAG synthesis, including EXTL-1,2,3, EXT-1,2, ChSY-1,3, and HAS-2,3 in RRMECs, pointing to shedding as a major mechanism for loss of endothelial glycocalyx. Both retinal endothelial cells cultured in vitro under high-glucose conditions and the retinas of diabetic rats exhibited altered sulfation of HS and CS, including N,6-O-sulfation on HS and 4-O-sulfation on CS. Underlined text is the observations that were same in both our in vitro and in vivo models.