José L Llergo

High glucose induces cell death of cultured human aortic smooth muscle cells through the formation of hydrogen peroxide

Alterations of the vessel structure, which is mainly determined by smooth muscle cells through cell growth and/or cell death mechanisms, are characteristic of diabetes complications. We analysed the influence of high glucose (22 mM) on cultured human aortic smooth muscle cell growth and death, as hyperglycaemia is considered one of the main factors involved in diabetic vasculopathy. Growth curves were performed over 96 h in medium containing 0.5% foetal calf serum. Cell number increased by 2–4 fold over the culture period in the presence of 5.5 mM (low) glucose, while a 20% reduction in final cell number was observed with high glucose. Under serum‐free conditions, cell number remained constant in low glucose cultures, but a 40% decrease was observed in high glucose cultures, suggesting that high glucose may induce increased cell death rather than reduced proliferation. Reduced final cell number induced by high glucose was also observed after stimulation with 5 or 10% foetal calf serum. The possible participation of oxidative stress was investigated by co‐incubating high glucose with different reactive oxygen species scavengers. Only catalase reversed the effect of high glucose. Intracellular H2O2 content, visualized with 2′,7′‐dichlorofluorescein and quantified by flow cytometry, was increased after high glucose treatment. To investigate the cell death mechanism induced by high glucose, apoptosis and necrosis were quantified. No differences were observed regarding the apoptotic index between low and high glucose cultures, but lactate dehydrogenase activity was increased in high glucose cultures. In conclusion, high glucose promotes necrotic cell death through H2O2 formation, which may participate in the development of diabetic vasculopathy.

Endothelial dysfunction and metabolic control in streptozotocin‐induced diabetic rats

The aim of this work was to study the influence of the metabolic control, estimated by the levels of glycosylated haemoglobin in total blood samples (HbA1c), in developing vascular endothelial dysfunction in streptozotocin‐induced diabetic rats. Four groups of animals with different levels of insulin treatment were established, by determining HbA1c values in 5.5 to 7.4%, 7.5 to 9.4%, 9.5 to 12% and >12%, respectively. The parameters analysed were: (1) the endothelium‐dependent relaxations to acetylcholine (ACh) in isolated aorta and mesenteric microvessels; (2) the vasodilator responses to exogenous nitric oxide (NO) in aorta; and (3) the existence of oxidative stress by studying the influence of the free radical scavenger superoxide dismutase (SOD) on the vasodilator responses to both ACh and NO. In both isolated aortic segments and mesenteric microvessels, the endothelium‐mediated concentration‐dependent relaxant responses elicited by ACh were significantly decreased when the vessels were obtained from diabetic animals but only with HbA1c values higher than 7.5%. There was a high correlation between HbA1c levels and the impairment of ACh‐induced relaxations, measured by pD2 values. The concentration‐dependent vasorelaxant responses to NO in endothelium‐denuded aortic segments were significantly reduced only in vessels from diabetic animals with HbA1c values higher than 7.5%. Again, a very high correlation was found between the HbA1c values and pD2 for NO‐evoked responses. In the presence of SOD, the responses to ACh or NO were only increased in the segments from diabetic rats with HbA1c levels higher than 7.5%, but not in those from non‐diabetic or diabetic rats with a good metabolic control (HbA1c levels <7.5%). These results suggest the existence of: (1) a close relation between the degree of endothelial dysfunction and the metabolic control of diabetes, estimated by the levels of HbA1c; and (2) an increased production of superoxide anions in the vascular wall of the diabetic rats, which is also related to the metabolic control of the disease.

Prevention of endothelial dysfunction in streptozotocin-induced diabetic rats by gliclazide treatment.

The aim of the present work was to analyze whether the oral hypoglycemic drug gliclazide affects diabetic endothelial dysfunction in streptozotocin-induced diabetic rats. Gliclazide was compared with glibenclamide, ascorbic acid, and aminoguanidine. An insulin-dependent model of diabetes was selected to exclude insulin-releasing effects of the drugs. Both in isolated aortic segments and mesenteric microvessels, endothelium-dependent relaxation evoked by acetylcholine (ACh, 1 nM to 10 microM) was significantly reduced in vessels from diabetic animals. This impairment was reversed when the segments were previously incubated with 100 U/ml superoxide dismutase. When streptozotocin-induced diabetic rats were orally treated from the time of diabetes induction with gliclazide (10 mg/kg) or ascorbic acid (250 mg/kg), ACh-induced endothelium-dependent relaxation was well preserved both in aortic segments and mesenteric microvessels. In addition, the impaired vasodilatation to exogenous nitric oxide (NO) in aortic segments was also improved in gliclazide-treated diabetic rats. On the other hand, oral treatment with glibenclamide (1 and 10 mg/kg) or aminoguanidine (250 mg/kg) did not produce significant improvements in diabetic endothelial dysfunction. We conclude that gliclazide reverses the endothelial dysfunction associated with diabetes. This effect appears to be due not to the metabolic actions of the drug but rather to its antioxidant properties, as it can be mimicked by other antioxidants. We propose that the mechanism involved is the inactivation of reactive oxygen species, which are increased in diabetes probably as a result of increased early protein glycosylation products, such as glycosylated hemoglobin (HbA(1c)). These effects of gliclazide are not shared by other oral hypoglycemic agent such as glibenclamide, or by blockade of advanced glycosylation end product (AGE) generation with aminoguanidine.

Treatment with acarbose may improve endothelial dysfunction in streptozotocin-induced diabetic rats.

We sought to determine whether a single reduction of hyperglycemia and those derivatives from nonenzymatic protein glycosylation may be effective in reducing the development of diabetic endothelial dysfunction. Therefore, we investigated how acarbose, an inhibitor of intestinal alpha-glucosidase that reduce hyperglycemia by lowering glucose absorption, may prevent the impairment of acetylcholine (ACh)-induced endothelium-dependent relaxations observed in isolated vascular segments from untreated streptozotocin-induced diabetic rats. When administered after diabetes induction, 10 mg/kg acarbose decreased modestly the enhancement of blood glucose and glycosylated hemoglobin (HbA1c) levels, but not those of advanced glycosylation end products (AGEs). This effect was linked to a partial improvement of ACh-induced responses both in conductance vessels, such as aortic segments, and resistance vasculature, like mesenteric microvessels. When acarbose was introduced after 6 weeks of untreated diabetes, blood glucose, HbA1c, and AGE levels were not affected and endothelial dysfunction remained unchanged in mesenteric microvessels, whereas a small improvement was observed in aortic segments. The addition of 100 U/ml superoxide dismutase enhanced the impaired relaxations to values similar to vessels from nondiabetic rats, indicating a main role for superoxide anions in diabetes-induced endothelial dysfunction. We conclude that hyperglycemia itself or elevated HbA1c, but not plasma AGEs, are related to enhanced oxidative stress and to the impairment of endothelium function associated to diabetes. This process can be partially prevented by reducing glucose absorption with acarbose.

Vascular smooth muscle cell hypertrophy induced by glycosylated human oxyhaemoglobin

1 Nonenzymatic protein glycosylation is a possible mechanism contributing to oxidative stress and vascular disease in diabetes. In this work, the influence of 14%‐glycosylated human oxyhaemoglobin (GHHb), compared to the non‐glycosylated protein (HHb), was studied on several growth parameters of rat cultured vascular smooth muscle cells (VSMC). A role for reactive oxygen species was also analysed. 2 Treatment of VSMC for 48 h with GHHb, but not with HHb, increased planar cell surface area in a concentration dependent manner. The threshold concentration was 10 nM, which increased cell size from 7965±176 to 9411±392 μm2. Similarly, only GHHb enhanced protein content per well in VSMC cultures. 3 The planar surface area increase induced by 10 nM GHHb was abolished by superoxide dismutase (SOD; 50–200 u ml−1), deferoxamine (100 nM–100 μM), or dimethylthiourea (1 mM), while catalase (50–200 u ml−1) or mannitol (1 mM) resulted in a partial inhibition of cell size enhancement. 4 When a known source of oxygen free radicals was administered to VSMC, the xanthine/xanthine oxidase system, the results were analogous to those produced by GHHb. Indeed, enhancements of cell size were observed, which were inhibited by SOD, deferoxamine, or catalase. 5 These results indicate that, at low concentrations, GHHb induces hypertrophy in VSMC, this effect being mediated by superoxide anions, hydrogen peroxide, and/or hydroxyl radicals. Therefore, glycosylated proteins can have a role in the development of the structural vascular alterations associated to diabetes by enhancing oxidative stress.

Effects of captopril, losartan, and nifedipine on cell hypertrophy of cultured vascular smooth muscle from hypertensive Ren-2 transgenic rats

We hypothesized that tissular renin‐angotensin system (RAS) induces vascular hypertrophy in hypertensive Ren‐2 transgenic rats (TGR; strain name TGR(mRen2)L27). This assumption was tested in cell cultures of vascular smooth muscle (VSMC) from both hypertensive TGR and control normotensive Sprague‐Dawley (SD) rats. Planar cell surface area, protein synthesis, and protein content per cell were studied, the role for locally produced angiotensin II (AII) was evaluated and the possible pharmacological interference by different drugs was analysed. By use of radioimmunoassay techniques, AII could be determined in TGR cultures (10.25±0.12 pg per 107 cells) while it could not be detected in SD ones. Under serum‐free conditions, VSMC from hypertensive TGR were hypertrophic when compared to SD VSMC, as they presented a higher protein content per cell (335±18 and 288±7 pg per cell respectively; P<0.05) and increased mean planar cell surface area, as determined by image analysis (4,074±238 and 4,764±204 μm2, respectively; P<0.05). When exogenously added to cultured SD and TGR VSMC, AII (100 pM to 1 μM) promoted protein synthesis and protein content in a concentration‐dependent manner without affecting DNA synthesis. Maximal effects were observed at 100 nM. At this concentration, AII effectively increased planar cell surface area in both SD and TGR cultures by ∼20%. Treatment of TGR cultures, in the absence of exogenous AII, with the angiotensin‐converting enzyme inhibitor captopril or the angiotensin AT1 receptors antagonist losartan (100 nM to 10 μM) reduced planar cell surface area in a concentration‐dependent manner. In addition, both captopril and losartan (10 μM), decreased protein synthesis by ∼15%. Treatment of SD VSMC, in the absence of exogenous AII, with both captopril and losartan had no effect either on planar cell surface area or protein synthesis. Treatment with the Ca2+ antagonist nifedipine (100 nM to 10 μM) reduced cell size in both SD and TGR cultures. Maximal cell reduction reached by nifedipine averaged 906±58 and 1,292±57 μm2, in SD and TGR, respectively (P<0.05). In addition, nifedipine, nitrendipine and nisoldipine (all at 10 μM) decreased protein synthesis in both cell types by 15–25%. We concluded that cultured VSMC from TGR are hypertrophic in comparison with those from SD. This cell hypertrophy can be the consequence of the expression of the transgene Ren‐2 that activates a tissular RAS and locally produces AII, which acts in a paracrine, autocrine, or intracrine manner. Cell hypertrophy in TGR cultures could be selectively reduced by RAS blockade, while nifedipine decreased cell size and protein synthesis in both hypertrophic and non hypertrophic cells.

Thapsigargin induces apoptosis in cultured human aortic smooth muscle cells.

Vascular remodeling is a key feature of many pathologic states, including atherosclerosis, or hypertension. Vascular smooth muscle cells participate in determining the vessel structure by several mechanisms such as cell migration, cell growth, or cell death (necrosis or apoptosis). Here we report that thapsigargin, an inhibitor of endoplasmic reticulum Ca2+ -adenosine triphosphatase (ATPase), is able to induce apoptosis in human vascular smooth muscle cells (HVSMCs). Apoptosis was assessed by three different methods: differential chromatin binding dye staining. cytoplasmic histone-associated DNA fragments detection by enzyme-linked immunosorbent assay (ELISA) and terminal deoxyribonucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL). When HVSMCs were treated for 1 h with thapsigargin (100 nM-10 microM), there was a concentration-dependent increase in both parameters 24 h after the thapsigargin pulse. When a time-course experiment was performed, both parameters were significantly enhanced from 3 to 6 h after the exposure to thapsigargin. We conclude that thapsigargin promotes apoptosis in HVSMCs, providing a useful tool for the study of programmed cell death in human vascular smooth muscle.

Nifedipine, losartan and captopril effects on hyperplasia of vascular smooth muscle from Ren-2 transgenic rats

Vascular smooth muscle cells from hypertensive transgenic rats for the mouse Ren-2 gene exhibited radioimmunoassayable angiotensin II and hyperplasia in comparison with cells from Sprague-Dawley rats. However, neither captopril, losartan, saralasin, nor PD123319 (all at 10 microM) modified DNA synthesis or cell number observed in 4-day growth curves with 10% fetal calf serum. Nifedipine reduced DNA synthesis in both cell types, the concentration required being significantly higher in Sprague-Dawley- (1 microM) than in transgenic-derived cultures (100 nM). The EC50 values were of 2.43 +/- 0.32 and 1.0 +/- 0.17 microM, respectively (P < 0.05). In both cell types, only 10 microM nifedipine reduced serum-induced cell proliferation, but inhibition percentage was higher in transgenic-derived cultures. In conclusion, hyperplasia of transgenic-derived vascular smooth muscle cells is not blocked by angiotensin-converting enzyme inhibitors or angiotensin receptor antagonists, but these cells are more sensitive to the antiproliferative effects of nifedipine.

Pharmacological interference of vascular smooth muscle cell hypertrophy induced by glycosylated human oxyhaemoglobin.

Nonenzymatically glycosylated human oxyhaemoglobin induces vascular smooth muscle cell hypertrophy by releasing reactive oxygen species. We analysed the ability of drugs with antihypertrophic properties for the vascular wall and/or antioxidant activity, such as captopril, losartan, and nifedipine, or gliclazide, carvedilol, and ascorbic acid, to interfere with 10 nM glycosylated human oxyhaemoglobin-induced increase in vascular smooth muscle cell size (118+/-0.5% of basal). Vascular smooth muscle cell hypertrophy was abolished concentration-dependently, with pD(2) values over a 100-fold interval: 6.4+/-0.3, 7.7+/-0.4, 7.3+/-0.4, 7.4+/-0.6, 8. 8+/-0.2, and 9.0+/-0.2 for captopril, losartan, nifedipine, ascorbic acid, carvedilol and gliclazide, respectively. Drugs with powerful antioxidant properties, especially carvedilol and gliclazide, are particularly effective in preventing glycosylated human oxyhaemoglobin-induced vascular smooth muscle cell hypertrophy.