Search

We investigated the hypothesis that benfotiamine, a lipophilic derivative of thiamine, affects replication delay and generation of advanced glycosylation end-products (AGE) in human umbilical vein endothelial cells cultured in the presence of high glucose. Cells were grown in physiological (5.6 mM) and high (28.0 mM) concentrations of D-glucose, with and without 150 microM thiamine or benfotiamine. Cell proliferation was measured by mitochondrial dehydrogenase activity. AGE generation after 20 days was assessed fluorimetrically. Cell replication was impaired by high glucose (72.3%+/-5.1% of that in physiological glucose, p=0.001). This was corrected by the addition of either thiamine (80.6%+/-2.4%, p=0.005) or benfotiamine (87.5%+/-8.9%, p=0.006), although it not was completely normalized (p=0.001 and p=0.008, respectively) to that in physiological glucose. Increased AGE production in high glucose (159.7%+/-38.9% of fluorescence in physiological glucose, p=0.003) was reduced by thiamine (113.2%+/-16.3%, p=0.008 vs. high glucose alone) or benfotiamine (135.6%+/-49.8%, p=0.03 vs. high glucose alone) to levels similar to those observed in physiological glucose. Benfotiamine, a derivative of thiamine with better bioavailability, corrects defective replication and increased AGE generation in endothelial cells cultured in high glucose, to a similar extent as thiamine. These effects may result from normalization of accelerated glycolysis and the consequent decrease in metabolites that are extremely active in generating nonenzymatic protein glycation. The potential role of thiamine administration in the prevention or treatment of vascular complications of diabetes deserves further investigation.

This study aimed at verifying whether thiamine, a co-enzyme which decreases intracellular glycolysis metabolites by allowing pyruvate and glyceraldheyde 3-phosphate to enter the Krebs cycle and the pentose-phosphate shunt, respectively, corrects delayed replication caused by high glucose concentrations in cultured human umbilical vein (HUVEC) and bovine retinal endothelial cells (BREC). After incubation in physiological (5.6 mmol/l) or high (28.0 mmol/l) glucose with or without 150 mumol/l thiamine, cells were counted and proliferation assessed by mitochondrial dehydrogenase activity. Lactate was measured in both cell types as an index of glycolytic activity and fluorescent advanced glycosylation end-products (AGE) concentration was determined in the HUVEC lysate. Both cell counts and proliferation assays in either of the cell types confirmed the impairment to cell replication induced by high glucose. When thiamine was added to cells kept under high glucose conditions, the number of surviving cells was significantly increased and the reduced cell proliferation appeared to be corrected. Lactate assays confirmed the increased production of this metabolite by BREC and HUVEC in high glucose, which was reduced by thiamine. Fluorescent AGE determination showed that thiamine may prevent non-enzymatic glycation in HUVEC. Thiamine restores cell replication, decreases the glycolytic flux and prevents fluorescent AGE formation in endothelial cells cultured in high glucose, suggesting that abnormal levels of glycolytic metabolite(s) may damage cells.

Damage caused to the vessel wall by diverse mechanisms may lead to diabetic microangiopathy. Consequently, research work is more and more focusing on the pathophysiology of vascular cells, with particular emphasis on endothelium. This paper reviews the present knowledge on the alterations of small vessel endothelium in diabetes. The most important risk factors for diabetic microangiopathy are the duration of disease and the degree of metabolic control maintained throughout the years. However, genetic factors may also contribute. These are examined first, followed by the presumed roles played by increased protein glycation and the production of Advanced Glycosylation End Products, the "polyol pathway" and free radical generation. Endothelium is a widespread, extremely active organ which regulates complex physiologic functions and its structure and function are discussed in the second section of this review. The third part deals with how diabetes can affect endothelium and describes observations on endothelial metabolism in vitro as well as morphologic and functional alterations in the patients. Unfortunately, the mechanisms leading to progressive degeneration of the microcirculation and organ damage in diabetic patients remain largely unaccounted for.

Human umbilical vein endothelial cells (HUVEC) cultured in high glucose exhibit delayed replication and colchicine-resistant microtubules. Tubulin dysfunction and stabilization, brought about by acetylation of the NH2-terminal residues, loss of the C-terminal tyrosine and binding of microtubular-associated proteins (MAPs) may be involved in the above phenomenon. The effects of L-tyrosine on HUVEC replication in high glucose were tested and the hypothesis that non-enzymatic glycosylation might impair tubulin depolymerization was also checked by growing the cells in the presence of L-glucose, which binds to intracellular proteins but remains metabolically inactive. After 18 days in culture, the number (mean +/- SEM, n = 7) of HUVEC grown in 28.0 mmol/l D-glucose (435.7 +/- 59.1 x 10(3)) was lower than in 5.6 mmol/l D-glucose (818.3 +/- 75.2 x 10(3)), p < 0.0001. The addition of L-tyrosine 1.7 mmol/l corrected such growth inhibition (623.3 +/- 81.7 x 10(3)), p < 0.0001 vs. D-glucose 28.0 mmol/l, but the cells recovered were less numerous than in physiological glucose alone (p = 0.016). The addition of L-tyrosine to D-glucose 5.6 mmol/l (731.0 +/- 63.2 x 10(3)) did not modify the cell number significantly. HUVEC in extra L-glucose (687.4 +/- 72.0 x 10(3)) were less numerous than in 5.6 mmol/l D-glucose, p = 0.028, but more than in D-glucose 28 mmol/l, p < 0.0001, and were not modified by the addition of L-tyrosine (729.4 +/- 67.1 x 10(3)). HUVEC grown in physiologic and high glucose exhibited specific immunofluorescence for acetylated tubulin and MAPs.(ABSTRACT TRUNCATED AT 250 WORDS)

We briefly summarize current knowledge on 1) the abnormalities of von Willebrand factor (vWF) as an indicator of endothelial cell (EC) dysfunction in diabetes and 2) the modifications induced in the growth of cultured ECs by high glucose in the incubation media. A MEDLINE search (1986 through Sept. 1989) was performed to update previous relevant references on vWF and ECs in healthy and diabetic subjects. Main data in the literature and personal contributions were scrutinized. Study quality, information, and relevance to the subject were assessed. vWF is synthesized and stored mainly in ECs. Its plasma levels are increased in diabetic microangiopathy but are not influenced by circulating glucose, insulin, or growth hormone, nor do they acutely affect platelet function in diabetes. Supraphysiological concentrations of glucose inhibit the replication of cultured ECs from large vessels via different possible mechanisms but appear to stimulate pathways involved in the activation of capillary ECs. vWF is a possible marker of EC damage in diabetes, and prospective studies will ascertain its role as a predictor for the development of microangiopathy. The possible dichotomy in the response of cultured ECs from large and small vessels to high glucose in the culture media may help explain some of the lesions observed in the walls of arteries and capillaries in diabetes.

Plasma von Willebrand factor was studied for quantitative and qualitative assessment in basal conditions and after release from the endothelium, as elicited by venous stasis of the forearm, in 8 healthy subjects and 8 patients with diabetic microangiopathy. von Willebrand factor was measured as factor VIII-related antigen (VIII R:Ag) and ristocetin co-factor (VIII R:Co). Its molecular size distribution was evaluated by bidimensional immunoelectrophoresis. The patients had higher basal levels of VIII R:Ag (171 +/- 45% vs 64 +/- 11% of plasma pooled from 20 healthy donors, p less than 0.05) and VIII R:Co (144 +/- 11% vs 88 +/- 8%, p less than 0.01). After stasis, both moieties increased significantly in the 2 groups, remaining higher in the patients (VIII R:Ag = 292 +/- 74% vs 89 +/- 18%, p = 0.01; VIII R:Co = 216 +/- 25% vs 116 +/- 10%, p less than 0.01). No differences in the molecular size distribution were observed between patients and controls, nor within the 2 groups before and after stasis. It is concluded that, in diabetic microangiopathy, the endothelial cells synthesize and store increased amounts of structurally normal von Willebrand factor.

The levels of angiotensin-converting enzyme and Factor VIII-related antigen, 2 endothelial synthesized glycoproteins, were measured in basal conditions and after forearm venous stasis in 12 healthy controls and 12 patients with diabetic microangiopathy. Angiotensin-converting enzyme levels were similar in the controls and in the patients both basally (185 +/- 17 nm/ml/min (SEM) and 192 +/- 15, respectively) and after stasis (238 +/- 17 and 220 +/- 15), whereas Factor VIII-related antigen was lower in the former basally (101 +/- 13% vs 184 +/- 16%, p less than 0.001) and after stasis (140 +/- 21% and 243 +/- 21%, p less than 0.01). It is concluded that Factor VIII-related antigen is a more sensitive indicator of endothelial cell damage in diabetic microangiopathy than angiotensin-converting enzyme.

Recent advances in our knowledge of endothelial function in health and in the presence of generalized microvascular damage, such as found in diabetic microangiopathy, are discussed in this review article. In microangiopathy there are no typical morphological signs of endothelial stress, apart from ubiquitous thickening of the basement membrane and the finding of both hyper- and acellular capillaries, whereas important derangements of endothelial function may occur in the early stages of the disease, like accelerated turnover, intracellular accumulation of potentially toxic intermediate metabolites, hyperpermeability of the vessel wall, altered synthesis of some haemostatic factors, expression of angiogenetic potentialities in inappropriate situations. The possible pathogenetic mechanisms underlying such alterations are reviewed, with special emphasis on the shortcomings of the experimental models available for laboratory investigation and on the in vivo situation.

Endothelial cells are directly exposed to the metabolic derangements of diabetes mellitus and may be damaged early, if not primitively, in the pathogenesis of diabetic microangiopathy. Cultured human umbilical vein endothelial cells were subjected to equimolar concentrations of glucose or mannitol for the evaluation of 3H-thymidine uptake, cell replication, cell death and repair of standard mechanical lesions. 3H-thymidine uptake was inhibited dose-dependently and to a similar extent by both glucose and mannitol. The former reduced cell replication whereas the latter did not (p less than 0.01 at 27.8 mmol/l, p less than 0.001 at 50.0 mmol/l). Neither caused excess cell death nor interfered with lesion repair. These results suggest that supra-physiological amounts of glucose inhibit DNA synthesis with osmotic mechanisms, delay cell replication through at least partially non osmotic effects, do not cause excess cell death and do not impair local injury repair. The above effects may play a role in the pathogenesis of long-term complications of diabetes.