Vascular complications are the main causes of morbidity and mortality in diabetes mellitus. ECs play a pivotal role in the regulation of vascular tone, as well as in the maintenance of vascular integrity, blood fluidity and homeostasis. EC injury is the initial step leading to irreversible structural abnormalities, followed by progressive microvascular occlusion in the eye and kidney as well as intimal proliferation in large vessels [33–35]. The exact cause of EC injury is still unclear.
In this study, PAECs were used as an in vitro model to study human vascular disease associated with EC injury since there is a similarity between human and porcine tissue . ECs of both micro and macro vascular origin present similar pathological features in diabetic complications. Thus ECs from porcine aorta (macro vessel) injured by high glucose could mimic EC injury associated with uncontrolled hyperglycemia in diabetes. High glucose injury in our model agrees with previous observations of ECs grown under hyperglycemic conditions showing decreased proliferation and fibrinolytic potential and increased programmed cell death [37, 38]. It was previously reported that normal human umbilical vein ECs showed increased proliferation when cultured in medium with high glucose (30 mM) for ten to twelve days , a longer period compared to the seven day treatment in our study. Cell proliferation increased similarly when umbilical ECs were obtained from pregnant diabetic women . PAECs treated with high glucose for seven days in the present study may represent early forms of injury and showed reduced live cell number indicating decreased cell proliferation. Some variation exists in the response of ECs to high glucose conditions. Cell conditions such as variation between animals, subtle differences in medium, CO2 levels, humidity, and other unidentified factors may be responsible for this variation.
Depletion and abnormalities in HS and HSPG have been found in the kidney, skin and aortic intima of diabetic patients with nephropathy [13–15, 41]. The degradation of HSPG may play a role in EC injury leading to diabetic vascular complications. Heparanase cleaves HS chains at specific sites and may be responsible for HSPG degradation contributing to EC injury. Heparanase has been found in the kidney and urine of diabetic patients . In order to determine if heparanase as well as high glucose damage ECs, PAECs were treated with heparinase I.
Several heparanases have been purified and characterized from platelets, placenta, and Chinese Hamster Ovary (CHO) cells including connective tissue activating peptide III (CTAP-III), Hpa I, Hpa II and CHO cell heparanases . Heparinase I, from Flacobacterium heparinum (Cytophagia heparinia), the commercially available heparanase, was chosen for PAEC treatment, and cleaves HS . Heparinase I did not cause injury when PAECs were cultured in M199 with serum for six to ten days, but showed a dose effect when cultured in serum-free medium for two days. These findings suggest that a serum constituent inhibits heparanase activity. A recently discovered cell surface protein, HS/heparin-interacting protein (HIP), was shown to prevent heparanase access to its substrate HS by competing with the same binding recognition site as in the HS chain [44, 45]. Thus in our experiments, serum may contain HIP so that heparanase was active only in serum free medium.
Our findings showing cell injury with both high glucose and heparinase I treatment suggest that high glucose may induce heparanase upregulation which degrades HS causing cell injury. This injury occurs in the presence of serum which would contain HIP that may interact with heparanase at the cell surface. This suggests that with high glucose, heparanase may be produced within the cell. Heparanase activity is optimal between pH 5.0 and 6.5, with much less activity above pH 7.0 . Glucose (30 mM) added for seven days to ECs lowers the medium pH (medium color become yellow) and may further stimulate heparanase activity.
Exogenous heparin significantly reduces proteinuria in diabetic patients and animals [47, 48]. Heparin promotes antioxidant and barrier properties of blood vessels, prevents the formation of occlusive vascular thrombi, protects against proteolytic or oxidative damage, and lowers blood pressure [26, 49, 50]. Heparin and HS, similar in chemical structure, possess common physiological and biological features important in the vasculature. Heparin modifies the synthesis and the structure of HSPG [51, 52]. In our study, addition of heparin to heparinase I treated ECs significantly increased live cell number and decreased LDH release compared to ECs treated with heparinase I alone suggesting that heparin has the ability to prevent cell injury by heparanase. A significant decrease in LDH release and a trend towards an increase in live cell number (close to control levels) seen in high glucose and heparin treated cells compared to high glucose alone also indicate the potential of heparin to protect ECs injured by high glucose.
HS and heparin have high affinity for bFGF and are part of the bFGF/bFGFR complex that affects the growth, differentiation and migration of many cell types . Thus, bFGF function is protected by HS synthesis and perturbed by its degradation. Our results showed a significant increase in live cell number and a trend towards a decrease in LDH release both in cells treated with high glucose plus bFGF and high glucose plus bFGF plus heparin when compared to high glucose alone indicating some protective effects of bFGF. However, the live cell number in controls is significantly greater than high glucose plus bFGF indicating bFGF alone dose not eliminate high glucose injury. Since high glucose produces many metabolic and biochemical abnormalities through several cellular pathways, normal bFGF function may be altered by its interaction with abnormal metabolites. Previous studies showed that in hyperglycemia, nonenzymatic glycosylation of bFGF decreased bFGF activity  and could explain our observations here. Moreover, we observed similar results when ECs were damaged by heparinase I. The protective effect of bFGF on heparinase I injury is shown by a significantly decreased LDH release and a trend towards an increase in cell number compared to heparinase I injury alone. This protective ability of bFGF is consistent with that seen in high glucose plus bFGF treated ECs.
The binding of heparin to bFGF depends on the molecular mass, degree of sulfation and the disaccharide composition. Unfractionated bovine lung heparin used here is highly sulfated and of high molecular weight and has previously been shown to protect bFGF from tryptic cleavage. This capacity was reduced by N-desulfation and N-acetylation of the bovine lung heparin . Previous studies have also suggested that heparin first needs to bind to the cell surface to fulfill the role of heparan sulfate in bFGF receptor interactions . We have observed that bovine lung heparin binds to the surface of cultured porcine endothelial cells and thus would be able to interact with bFGF .
When heparin was added to ECs treated with heparinase I and bFGF, live cell number increased and LDH release decreased significantly compared to heparinase I treatment alone and showed a more pronounced increase in live cells and decrease in LDH than addition of bFGF alone suggesting that bFGF and heparin bind together to prevent HSPG from degradation by heparinase I. These findings cause us to speculate that heparin may exert its protective effect in two steps: firstly, heparin increases EC synthesis of HS; secondly, newly synthesized HS with exogenous bFGF and heparin form the bFGF/HS or heparin/ bFGFR complex which allows bFGF to play its physiological role in cell growth, differentiation, proliferation. As well, in the case of heparanase injury, heparin in the medium may compete with HS for heparanase and thus may prevent the degradation of HS.
Insulin not only stimulates cells to utilize glucose, but also promotes DNA synthesis and cell growth. The latter effect was supported in this study when ECs, treated with insulin alone, significantly increased live cell number compared to controls (Figure 1). Insulin protection of high glucose or heparinase I treated ECs was shown in all treatment combinations including insulin alone, insulin plus heparin, insulin plus bFGF and insulin plus heparin plus bFGF. The mechanism by which insulin protects ECs from high glucose injury is not entirely understood. Other vasoprotective actions of insulin are its ability to increase NO production , act as an antioxidant and prevent atherosclerosis by reducing oxygen consumption . Our present study suggests that the combination of insulin, heparin, and bFGF may have additive effects with significantly increased live cell number and decreased LDH compared to high glucose or heparinase I alone. With high glucose injury the three combined treatments were more effective than bFGF and bFGF plus heparin and suggested increased effectiveness compared to insulin plus bFGF when medium LDH levels were considered. With heparinase I injury combined treatments were significantly more effective than bFGF with a tendency towards increased effectiveness with heparin or insulin plus bFGF when live cell number was considered.