Endothelial Nitric Oxide Production and Insulin Sensitivity
A Physiological Link With Implications for Pathogenesis of Cardiovascular Disease
Background Insulin sensitivity varies up to threefold in apparently healthy individuals, but the mechanism for this is unknown. We have examined the hypothesis that vascular endothelial nitric oxide production and insulin sensitivity are directly related in humans.
Methods and Results Nineteen healthy male subjects were studied on 3 separate days 1 week apart during which time they underwent measurement of insulin sensitivity by the euglycemic hyperinsulinemic clamp technique (soluble insulin 1.5 mU·kg−1·min−1) and measurement of in vivo basal and stimulated endothelial nitric oxide production by forearm venous occlusion plethysmography. There was a correlation between insulin sensitivity and forearm vasoconstrictor responses to NG-monomethyl-l-arginine, the substrate inhibitor of nitric oxide synthase (r=.52, P<.05). No correlations were observed between insulin sensitivity and noradrenaline, acetylcholine, or sodium nitroprusside responses.
Conclusions Endothelial nitric oxide synthesis and insulin sensitivity are positively related in healthy humans, which suggests a direct physiological link.
The association between hyperinsulinemia and essential hypertension has been recognized for over 30 years.1 The elevation in circulating insulin levels appears to occur as a pancreatic β-cell compensatory response to decreased sensitivity to insulin-stimulated glucose uptake in peripheral tissues, particularly skeletal muscle,2 but the mechanisms that underlie the development of insulin resistance in hypertension and other common conditions such as obesity and non–insulin-dependent diabetes mellitus remain unknown.
The elucidation of these mechanisms has been complicated by the threefold variation in insulin sensitivity observed within groups of apparently healthy nonobese individuals.3 Insulin is an arterial vasodilator in skeletal-muscle vascular beds, and there is evidence that insulin-mediated vasodilation is reduced in states of insulin resistance.4 It has been suggested recently that the vascular effects of insulin may be dependent on endothelial nitric oxide synthesis/release,5 and decreased basal endothelial nitric oxide synthesis has been reported in essential hypertension.6
Nineteen healthy, normotensive, male volunteers aged 21 to 35 years with normal glucose tolerance participated in these studies with the approval of the West Ethical Committee of the West Glasgow Hospitals University NHS Trust. No subjects were taking medication, and all subjects abstained from alcohol and tobacco for 24 hours and from food and caffeine-containing drinks overnight before study days. At a screening visit, physical health was confirmed by history and physical examination, supine blood pressure was measured in triplicate (Dinamap Critikon, Johnson and Johnson Professional Products Ltd), and a self-reported index of moderate and vigorous physical activity level was determined.7 Characteristics (mean±SD) of the subjects were body mass index (BMI) 24.6±3.0 kg/m2, fasting glucose level 5.1±0.3 mmol/L, fasting cholesterol level 4.40±0.74 mmol/L, and blood pressure 130±10/69±7 mm Hg. Mean (±SD) age was 27.1±5.3 years; three subjects were smokers; six had a family history of cardiovascular disease or diabetes; and median alcohol consumption was 16 U (interquartile range, 6 to 20 U).
Hyperinsulinemic Euglycemic Clamp
On the first study day, subjects underwent assessment of sensitivity to insulin-mediated glucose uptake by use of a 3-hour primed infusion of 1.5 mU·kg−1·min−1 soluble insulin (Human Actrapid, NovoNordisk A/S) along with a variable-rate infusion of 20% dextrose (Baxter Healthcare) adjusted to achieve euglycemia at 5.2 mmol/L on the basis of arterialized samples withdrawn every 5 minutes from an ipsilateral right dorsal hand vein (heated-air box at 55°C, University of Nottingham Department of Physiology and Pharmacology). Whole-body insulin sensitivity (M) was calculated from the glucose infusion rate during the final 40 minutes.8
Forearm Venous Occlusion Plethysmography
On 2 additional study days, measurements of basal and stimulated endothelial nitric oxide production were obtained from the subjects. Forearm blood flow was measured by use of venous occlusion plethysmography with mercury-in–silicone elastomer (Silastic) strain gauges that had been electrically calibrated (Hokanson set, PMS instruments).9 A 27-gauge unmounted steel needle (Cooper’s Needleworks) was inserted under local anesthesia into the brachial artery of the nondominant arm for drug infusion. Temperature was maintained at 24°C to 26°C. Local incremental doses of drugs (see “Results”) were dissolved in 0.9% saline and infused intra-arterially at a constant rate of 1 mL/min. A 30-minute washout period was allowed between drug infusions. On day 1, subjects received acetylcholine (Miochol, Cibavision), an endothelium-dependent stimulator of nitric oxide synthase, and sodium nitroprusside (Roche), an endothelium-independent donor of nitric oxide. On day 2, subjects received ascending doses of noradrenaline (Levophed, Sanofi-Winthrop), a control vasoconstrictor, and NG-monomethyl-l-arginine (L-NMMA; Clinalfa AG), a substrate inhibitor of nitric oxide synthase. Blood flow was measured in both forearms, and each value was the mean of five sequential measurements. Percent change from basal values in the ratio of blood flow between infused and noninfused arms was calculated with blood flow in the noninfused arm as a concurrent control.9 Because serial measurements were made, each subject’s mean response to all administered doses of each drug was calculated as a summary measure.10
Data were initially examined by use of simple regression (Pearson); a multiple regression analysis was performed to examine potential confounders (Minitab statistical package, Minitab Inc).
Insulin sensitivity (M) ranged from 6.2 to 11.6 mg·kg−1·min−1. All 19 subjects completed acetylcholine and sodium nitroprusside infusions; 15 completed noradrenaline and L-NMMA infusions (arterial cannulation could not be repeated in 2 subjects, and 2 withdrew for personal reasons). Grouped (mean±SE) vasodilator responses to ascending doses of acetylcholine (20, 40, 80, and 160 pmol/min) were 131±32%, 224±54%, 329±71%, and 426±61%, respectively; for sodium nitroprusside (3, 10, and 30 pmol/min), these responses were 118±11%, 232±24%, and 527±99%, respectively. Grouped (mean±SE) vasoconstrictor responses to ascending doses (15, 30, 150, and 300 pmol/min) of noradrenaline were 9±1%, 14±2%, 29±4%, and 38±4%, respectively; and for L-NMMA (1, 2, and 4 μmol/min), responses were 13±2%, 22±2%, and 30±3%, respectively. For each individual across all doses, mean vasodilator responses (percent change in forearm blood-flow ratio) ranged from 52% to 669% for acetylcholine and from 102% to 754% for sodium nitroprusside. For each individual, mean vasoconstrictor responses ranged between 8% and 41% for noradrenaline and between 11% and 29% for L-NMMA. Individual measurements of insulin sensitivity (M) were positively related to the mean L-NMMA responses (r=.52, P<.05; Fig). However, no relationships were observed between M values and noradrenaline responses (r=.14, P=NS; Fig), acetylcholine responses (r=.15, P=NS), or sodium nitroprusside responses (r=.31, P=NS). Similar results were obtained when the area under the curve was used as an alternative summary measure of drug response.10 The results of a multiple regression analysis, in which one subject identified by the statistical package as an unusual observation was excluded and which took into account age, BMI, mean arterial pressure, plasma glucose level, serum cholesterol level, alcohol intake, family history, and physical activity level, are shown in the Table⇓.
Our results in healthy young men demonstrate, for the first time, a positive relationship between basal vascular endothelial nitric oxide production and insulin sensitivity. Under physiological circumstances, insulin causes arterial vasodilation in skeletal-muscle vascular beds. By increasing its own delivery and that of glucose to insulin-sensitive tissues, it may amplify its own action in promoting glucose uptake, and it has been reported previously4 that insulin-mediated vasodilation is impaired in insulin-resistant states. Our data, along with recent evidence that endothelial nitric oxide may mediate the vascular effects of insulin,5 suggest a direct physiological link between vascular endothelial function and insulin sensitivity, ie, individuals who are relatively insensitive to insulin-mediated glucose uptake also appear to have a corresponding decrease in basal endothelial nitric oxide production.
We observed no relationship between noradrenaline responses and insulin sensitivity and suggest that this argues against a nonspecific decrease in vascular reactivity in subjects who are less sensitive to insulin-mediated glucose uptake. Furthermore, the absence of relationships with either the acetylcholine or sodium nitroprusside responses suggests that decreased insulin sensitivity is not associated with a reduced ability of the vascular endothelium to synthesize nitric oxide when stimulated or with a reduction in the sensitivity of vascular smooth muscle to nitric oxide released from the endothelium.
Insulin sensitivity has been shown to be related to age, BMI, and maximal aerobic capacity in normal subjects but may vary up to threefold in subjects who are apparently similar in these respects.3 The present study group was as homogeneous as possible for these variables, with only a twofold variation in insulin sensitivity. The relationship between the L-NMMA response and the measured insulin sensitivity was stronger than the relationships between all of the other variables and insulin sensitivity, accounting for 43% of the variance, and was not explained by a known confounding variable such as obesity. The relationship identified in the present study is compatible with three hypotheses: (1) decreased basal endothelial nitric oxide production results in decreased insulin sensitivity, possibly due to reduced insulin-mediated vasodilation in skeletal muscle; (2) decreased insulin sensitivity results in decreased basal production of nitric oxide by the vascular endothelium; and (3) decreased basal endothelial nitric oxide production and impaired insulin sensitivity are manifestations of a common genetic or environmental antecedent. The present data do not allow us to decide between these possibilities. Further study is required to determine whether a similar relationship is observed between endothelial function and insulin sensitivity in pathophysiological states such as essential hypertension and to determine the mechanistic explanation for the observed relationship.
This study was supported by a BHF junior research fellowship awarded to Dr Petrie.
- Received December 12, 1995.
- Revision received January 24, 1996.
- Accepted January 29, 1996.
- Copyright © 1996 by American Heart Association
Reaven G. Role of insulin in human disease. Diabetes. 1988;37:1595-1607.
Scherrer U, Randin D, Vollenweider P, Vollenweider L, Nicod P. Nitric oxide accounts for insulin’s vascular effects in humans. J Clin Invest. 1994;94:2511-2515.
Sallis JF, Haskell WL, Wood PD, Fortmann SP, Rogers T, Blair S, Paffenbarger RS. Physical activity assessment methodology in the five-city project. Am J Epidemiol. 1985;121:91-106.
DeFronzo RA, Tobin J, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. 1979;237:E214-E223.
Benjamin N, Calver A, Collier J, Robinson B, Vallance P, Webb D. Measuring forearm blood-flow and interpreting the responses to drugs and mediators. Hypertension. 1995;25:918-923.
Matthews JNS, Altman DG, Campbell MJ, Royston P. Analysis of serial measures in medical research. Br Med J. 1990;300:230-235.