Timothy M. Chan

Blood pressure lowering by pioglitazone. Evidence for a direct vascular effect.

To examine potential mechanisms for the blood pressure-lowering action of the thiazolidinedione compound, pioglitazone (PIO), we studied the effects of the drug on blood pressure and insulin action in vivo and on vascular tissue in vitro. In vivo, PIO lowered blood pressure in fructose-fed and chow-fed rats to an extent that could not be explained by alterations in fasting plasma insulin or free magnesium concentrations or by alterations in whole-body insulin sensitivity. In vitro, PIO caused significant blunting of the contractile responses of aortic rings to NE, arginine vasopressin (AVP), and potassium chloride; the blunting of responses to NE was maintained after removal of the endothelium. To assess the potential importance of extracellular calcium to the vasodepressor effect of PIO, we measured contractile responses to NE in the absence of calcium, and then after acute restoration of calcium in the presence of NE. PIO had no effect on the contractile response in the absence of calcium. By contrast, PIO blunted by 42% the contractile response that occurred when the extracellular calcium supply was acutely restored in the presence of NE, suggesting that the blunting was mediated by blockade of calcium uptake by vascular smooth muscle. Such an effect was confirmed in cultured a7r5 vascular smooth muscle cells, which exhibited a brisk increase in intracellular calcium in response to AVP that was blocked by PIO in a dose-dependent fashion. Our data indicate that PIO has a direct vascular effect that appears to be mediated at least in part by inhibition of agonist-mediated calcium uptake by vascular smooth muscle. The direct vascular effect may contribute to the blood pressure-lowering actions of PIO in vivo, because that effect could not be explained by alterations in whole-body insulin sensitivity.

Endothelial-Dependent Vascular Effects of Insulin and Insulin-Like Growth Factor I in the Perfused Rat Mesenteric Artery and Aortic Ring

Insulin and insulin-like growth factor I (IGF-I) exhibit vasoactivity. To examine the role of the endothelium in mediating the vascular responses to insulin and IGF-I, we exposed both isolated intact rat mesenteric arteries and rat aortic rings to these growth factors in the presence and absence of endothelium. Perfusion of rat mesenteric arteries with insulin, IGF-I, or IGF-II resulted in the potentiation of arginine vasopressin (AVP)-induced vasoconstriction. Of these growth factors, IGF-I was the most potent, with a significant effect at 0.6 nM and maximal effects at 6.0 nM, followed by IGF-II and insulin. Endothelial denudation or addition of cycloheximide prevented the growth-factor effects. Tissue cGMP levels in the mesenteric artery were minimally affected by growth factors. Insulin and IGF-I vascular effects were not inhibited by BQ123, an endothelin (ET) antagonist that blocked ET-1 enhancement of AVP response. Perfusion of mesenteric arteries with IGF-I for 1 h did not alter vessel ET-1 or ET-1 mRNA contents. Addition of indomethacin markedly inhibited the IGF-I effect on AVP contraction. Thus, the mesenteric vascular effect of insulin and IGF-I is not associated with ET-1 release but appears to link to an increased release of an endothelial-derived contracting factor or the decreased production of an endothelial-derived relaxing factor from the cyclooxygenase pathway. In contrast to their action in the mesenteric artery, insulin (exceeding 100 nM) and IGF-I (1–30 nM) attenuated AVP- and norepinephrine-induced contraction in rat aortic rings. Endothelial-denudation abolished this effect. L-Ng monomethyl arginine markedly reduced insulin and IGF-I responses in the aortic rings, suggesting involvement of endothelial nitric oxide production. Furthermore, IGF-I moderately increased tissue cGMP levels in the rings. These results suggest that the vascular effects of insulin and IGF-I are vessel-specific and mediated by the endothelium, possibly via IGF-I receptors.

Eosinophil peroxidase produces hypobromous acid in the airways of stable asthmatics

Eosinophil peroxidase and myeloperoxidase use hydrogen peroxide to produce hypobromous acid and hypochlorous acid. These powerful oxidants may damage the lungs if they are produced as part of the inflammatory response in asthma. The aim of this study was to determine if peroxidases generate hypohalous acids in the airways of individuals with stable asthma, and if they affect lung function. Sputum was induced from patients with mild to moderate asthma and from healthy controls. Eosinophil peroxidase, myeloperoxidase, chlorinated and brominated tyrosyl residues, and protein carbonyls were measured in sputum supernatants. Eosinophil peroxidase protein was significantly elevated in asthmatic subjects whereas myeloperoxidase protein was not. There was significantly more 3-bromotyrosine (Br-Tyr) in proteins from the sputum of asthmatics compared to controls (0.79 vs. 0.23 mmol Br-Tyr/mol Tyr; medians p < .0001). Levels of 3-chlorotyrosine (0.23 vs. 0.14 mmol Cl-Tyr/mol Tyr; medians p = .11) and protein carbonyls (0.347 vs. 0.339 nmol/mg protein; medians p = .56) were not significantly increased in asthmatics. Levels of 3-bromotyrosine were strongly correlated with eosinophil peroxidase protein (r = 0.79, p < .0001). There were no significant correlations between the markers of oxidative stress and lung function. We conclude that eosinophil peroxidase produces substantial amounts of hypobromous acid in the airways of stable asthmatics. Although this highly reactive oxidant is a strong candidate for exacerbating inflammatory tissue damage in the lung, its role in asthma remains uncertain.

Stimulation of tyrosine-specific protein phosphorylation in the rat liver plasma membrane by oxygen radicals.

Incorporation of 32P from [gamma-32P]ATP into endogenous proteins, added histone and the copolymers Glu 80 Tyr 20 by rat liver plasma membranes was markedly increased by several naphthoquinones, including menadione. This stimulation was most marked with Glu 80 Tyr 20, has an absolute requirement for either dithiothreitol or reduced glutathione, and was inhibited by superoxide dismutase, catalase, and desferrioxamine to varying degrees depending on the quinones used. Their effectiveness in stimulating the apparent tyrosine-specific protein phosphorylation correlated with the rates of DTT-dependent redox cycling measured by oxygen consumption. Increased protein phosphorylation was also seen with particulate fractions isolated from hepatocytes incubated with quinones. A free radical-mediated mechanism is suggested for the quinone stimulation of protein phosphorylation.

Impaired insulin receptor phosphorylation in skeletal muscle membranes of dbdb mice: The use of a novel skeletal muscle plasma membrane preparation to compare insulin binding and stimulation of receptor phosphorylation

A method has been developed to isolate skeletal muscle plasma membranes from mice in good yield without harsh extraction procedures. The method involves perfusion of mouse hindquarters with a calcium-deficient buffer containing collagenase and hyaluronidase. This is followed by gentle disruption, filtration, and differential centrifugations. The entire procedure takes about six hours and the yield is approximately 4 mg. protein from 10 g. equivalent of hindquarter muscle. The preparation contained predominantly plasma membranes based on specific activities of marker enzymes, electron microscopic data, and specific binding sites for insulin and a -adrenergic ligand. Studies using such preparations from lean, 4-5 week old and 12-20 week old db/db mice showed marked reduction in the phosphorylation of the 95 kDa subunit of the insulin receptor of the obese mice with no change in insulin binding. In addition, there was a progressive reduction in insulin sensitivity in stimulating receptor phosphorylation in the db/db mice.

Muscle Protein Turnover in the Perfused Hindquarters of Lean and Genetically Obese-Diabetic (db/db) Mice

The stability of the perfused mouse hindquarter was assessed for a period of 5 h, and the preparation shown to remain stable for metabolic studies over this time period. Muscle protein synthesis and degradation rates in lean and diabetic-obese (db/db) mice were measured using the in situ perfused hindquarter preparation. The rates of protein synthesis were 48% lower in the muscles of intact db/db mice than in the lean controls when expressed per gram TCA precipitable protein and 46% lower when expressed per gram dry weight. Adrenalectomy, which has been shown to restore the lean body mass of the db/db mice to normal, had the effect of returning protein synthesis rate in muscle of db/db mice to lean control values. Insulin at a dose of 1 mU/ml stimulated protein synthesis in lean mice only, showing that the process of protein synthesis in the db/db mice is also insensitive to insulin. Measurements of the rates of degradation of muscle protein showed no differences between lean and db/db mice. These findings suggest that the decreased lean body mass of db/db mice is the result of a defect in protein synthesis rather than due to altered degradation.

Effect of ciglitazone on glucose uptake and insulin sensitivity in skeletal muscle of the obese (obob) mouse: Distinct insulin and glucocorticoid effects☆

The oral hypoglycemic agent, ciglitazone, (5-[4-(1-methylcyclohexylmethoxy)benzyl]-thiazolidine-2,4-dione), was fed for nine days to genetically obese (ob/ob) mice aged 5 to 6 weeks. This treatment resulted in a lowering of plasma glucose and circulating insulin levels, but did not cause a fall in plasma corticosterone levels. Basal 2-deoxy-D-glucose uptake by the perfused hindquarters of ob/ob mice was unchanged by ciglitazone feeding. In the presence of 0.1 mU/mL insulin in the perfusion medium, there was a significant increase in the uptake rate of 2-deoxy-D-glucose by the skeletal muscle of ciglitazone-treated ob/ob mice, while there was no insulin effect in untreated ob/ob mice. Insulin at a concentration of 1 mU/mL caused a further stimulation of 2-deoxy-D-glucose transport. However, this response was significantly lower than the maximal stimulation in lean mice. Ciglitazone feeding did not have any effect on [5-3H]-glucose metabolism by the perfused muscle which remained subnormal, suggesting that the posttransport metabolism of glucose was limited by substrate availability. In the perfused mouse liver, net [14C]-glucose production from [14C]-lactate was unchanged by ciglitazone treatment while gluconeogenesis from [14C]-alanine was reduced. These findings show that ciglitazone produces its hypoglycemic effect by improving the insulin sensitivity in skeletal muscle, as others have reported in the adipose tissue. The presence of elevated plasma levels of corticosterone and lower levels of insulin in ciglitazone-treated ob/ob mice suggests that the adrenal glucocorticoids are responsible for the basal defects in glucose transport and the hyperinsulinemia is responsible for the insulin insensitivity.

ACTIVATION OF A MEMBRANE-ASSOCIATED PHOSPHATIDYLINOSITOL KINASE THROUGH TYROSINE-PROTEIN PHOSPHORYLATION BY NAPHTHOQUINONES AND ORTHOVANADATE

We have previously reported that several naphthoquinones stimulated tyrosine-specific protein phosphorylation in isolated rat liver membranes. Our more recent study demonstrated a similar effect by orthovanadate, which concomitantly stimulated phosphorylation of protein-tyrosine and phosphatidylinositol (Ptd-Ins). Results presented here show a simultaneous increase in PtdIns phosphorylation along with stimulation of tyrosine-protein phosphorylation by naphthoquinones. This PtdIns kinase resembles the type I PtdIns kinase in that it was insensitive to adenosine inhibition. The product, nevertheless, comigrated with a PtdIns-4-phosphate standard in TLC using three different solvent systems. Stimulation of PtdIns phosphorylation by vanadate or naphthoquinones could be achieved in the following preparations: intact rat liver membranes, Triton X-100-solubilized membranes, solubilized membranes partially purified by Sephacryl chromatography, solubilized membranes purified by wheat germ agglutinin chromatography. The naphthoquinone or vanadate-activated PtdIns kinase activity could be isolated by antiphosphotyrosine antibody-agarose affinity chromatography. The relative potencies of a series of ring-substituted naphthoquinones in the stimulation of tyrosine-protein phosphorylation, PtdIns kinase activity, dithiothreitol-dependent oxygen consumption, and cytochrome c reduction were highly correlated. We conclude that oxidant(s) produced by redox cycling of naphthoquinones stimulated an adenosine-insensitive PtdIns kinase through tyrosine phosphorylation of the enzyme.

Stimulation of tyrosine-specific protein phosphorylation and phosphatidylinositol phosphorylation by orthovanadate in rat liver plasma membrane☆

Orthovanadate stimulated the incorporation of 32P from [gamma-32P]ATP by Triton X-100-solubilized rat liver plasma membrane into endogenous, trichloroacetic acid-precipitable materials as well as added (Glu4:Tyr1) copolymers. Extraction of incubation mixture with chloroform-methanol-HCl revealed that the increase in 32P incorporation by vanadate was predominantly into endogenous phospholipids. [32P]Phosphatidylinositol 4-phosphate (PtdIns-4-P) was identified by thin-layer chromatography as the major phosphorylated product of vanadate stimulation, which also resulted in elevated 32P, predominantly in P-Tyr in endogenous membrane proteins. Vanadate effects on protein tyrosine and phosphatidylinositol phosphorylation were concomitant and exhibited similar sensitivity. These effects of vanadate were enhanced by the presence of either dithiothreitol or NAD(P)H. Phosphatidylinositol phosphorylation could also be stimulated by a substrate of and inhibited by a synthetic inhibitory copolymer of tyrosine kinase. These results suggest that vanadate, an oxygen radical producer, stimulates a tyrosine kinase-PtdIns kinase coupled system much like those described for a number of growth factors and oncogene encoded products.

Peripheral Insulin Insensitivity in the Hyperglycemic Athymic Nude Mouse: Similarity to Noninsulin-Dependent Diabetes Mellitus

Abstract Previous studies in the homozygous athymic nude mouse (USC colony) have indicated a diabetic state characterized by spontaneous hyperglycemia, abnormal glucose tolerance, and normal or relatively decreased plasma insulin levels. Pancreatic islet cell population assessed by morphometric and immunohistochemical studies demonstrated normal insulin-secreting cells in the hyperglycemic nude mouse. To further elucidate the pathogenesis of the hyperglycemic state in the athymic nude mouse, we have studied the effects of insulin on the transport of glucose in skeletal muscle by measuring basal and insulin-stimulated uptake of a nonmetabolizable glucose analogue, 2-deoxy-d-glucose by using the perfused hindquarter preparation. Although basal 2-deoxy-d-glucose uptake by peripheral skeletal muscle was similar in the hyperglycemic and control mice, the insulin-stimulated uptake of 2-deoxy-d-glucose was significantly decreased in the athymic nude mouse at both 0.1 milliunits/ml and supraphysiologic concentrations of insulin (1 milliunit/ml) when compared with control mice (P < 0.05 and P < 0.001, respectively). This form of peripheral insulin insensitivity with normal pancreatic β cell reserve, in addition to the lean body mass of the diabetic animal, mimics in part the peripheral insulin insensitivity seen in non-insulin-dependent diabetes mellitus.