Insulin Action Is Associated With Endothelial Function in Hypertension and Type 2 Diabetes
Abstract—A primary defect in the vascular action of insulin may be a key intermediate mechanism that links endothelial dysfunction with reduced insulin-mediated cellular glucose uptake in metabolic and cardiovascular disorders. The present study was designed to characterize more fully the relations between insulin action and endothelial function in male patients with essential hypertension (H, n=9) or type 2 diabetes (D, n=9) along with healthy control subjects (C) matched for age, body mass index, and lipid profile. They attended for measurement of whole-body insulin sensitivity (MCR) by the hyperinsulinemic clamp technique (day 1) and forearm vasoreactivity in response to intra-arterial infusions of insulin/glucose (day 2) and NG-monomethyl-l-arginine (L-NMMA) and norepinephrine (day 3) by bilateral venous-occlusion plethysmography. Results expressed as mean±SE MCR (mL/kg per minute) were 7.22±0.99 (C), 6.32±0.78 (H), and 5.06±0.53 (D). Insulin/glucose-mediated vasodilation (IGMV) was 17.1±5.6% (C), 17.2±5.5% (H), and 12.3±6.4% (D). L-NMMA vasoconstriction (LNV) was 37.9±5.1% (C), 37.5±2.3% (H), and 33.6±2.8% (D). There were no significant differences among groups for these parameters. Pooled correlation analyses revealed associations between MCR and IGMV (r=0.46, P<0.05), MCR and LNV (r=0.44, P<0.05), and IGMV and LNV (r=0.52, P<0.01). This study supports functional coupling between insulin action (both metabolic and vascular) and basal endothelial nitric oxide production in humans.
Abnormally reduced insulin-stimulated glucose uptake (insulin resistance) and vascular endothelial dysfunction are both features of metabolic syndromes and cardiovascular disorders. However, it is unclear whether this association is causal or a reflection of a common underlying pathophysiological mechanism. In addition to its metabolic and mitogenic actions, insulin is a directly acting vasodilator, an action that is both endothelium dependent1 and enhanced by d-glucose2 and l-arginine3 uptake. It is therefore tempting to speculate that the vascular effect of insulin may be a key intermediate mechanism that links endothelial function with insulin-mediated cellular glucose uptake. Indeed, in previous studies involving young healthy volunteers, we have demonstrated associations between insulin sensitivity and both basal endothelial nitric oxide (NO) production4 and insulin-mediated vasodilation.5 Others have shown associations between endothelial function and insulin-mediated vasodilation6 and, recently, between insulin action (both metabolic and vascular) and microvascular function.7 However, interpretation of results from studies in older volunteers and patient groups is less clear; for example, in type 2 diabetes and hypertension there are data that report blunting of insulin-mediated vasodilation that parallels changes in insulin sensitivity,8 9 although contrasting data report no defect in the vascular action of insulin.10 11 However, it is possible that these conflicting results are due to other factors such as age, obesity, and dyslipidemia, which have been shown to be important determinants of both insulin action and vascular endothelial function.12 13 14 In the present study, we hypothesized that insulin-resistant subjects (patients with essential hypertension and type 2 diabetes) would display blunting of both insulin-mediated vasodilation and basal endothelial NO production compared with subjects carefully controlled for potentially confounding factors such as age, body mass index (BMI), and lipid profile.
Twenty-seven male volunteers (9 patients with uncomplicated diet-controlled type 2 diabetes, 9 with essential hypertension either receiving no medication or off drugs for ≥3 weeks, and 9 healthy subjects recruited by advertisement) participated in this study, which was approved by the Ethics Committee of the West Glasgow Hospitals University NHS Trust. All subjects gave informed consent. No subjects were taking medication during the study, and all abstained from alcohol, tobacco, and strenuous physical activity for 24 hours and from food and caffeine-containing drinks overnight before the 3 study days, which were ≥1 week apart. Subject characteristics are displayed in the Table⇓.
Measurement of Whole-Body Insulin Sensitivity
Subjects attended for measurement of whole-body insulin sensitivity with the use of a 150-minute hyperinsulinemic isoglycemic clamp. A primed continuous infusion of soluble insulin (1.5 mU/kg per minute) was administered along with a variable rate infusion of 20% glucose adjusted manually to maintain serum glucose at fasting levels on the basis of arterialized samples withdrawn every 5 minutes from an ipsilateral right dorsal hand vein (heated-air box at 55°C). Insulin sensitivity was calculated during the last 40 minutes of the study and expressed as metabolic clearance ratio of glucose (MCR, mL/kg per minute) to correct for differences in fasting glucose levels.
Forearm Vascular Reactivity
On 2 further study days, subjects attended for measurement of forearm blood flow (FBF) by bilateral venous-occlusion strain-gauge plethysmography. A 27-gauge unmounted steel needle was inserted under local anesthesia into the brachial artery of the nondominant arm for infusions. Blood flow was recorded in both forearms during 3-minute periods of wrist cuff inflation at 8-minute intervals; each measurement was the mean of 5 sequential recordings. On one day, after baseline readings had been obtained, subjects received an intra-arterial infusion of insulin (5 mU/min) and d-glucose (75 μmol/min) for 72 minutes, as previously described.2 FBF ratio was measured every 8 minutes; the mean of the final 3 readings (56 to 72 minutes) was used as a summary measure for insulin/glucose-mediated vasodilation. On a separate day, again after baseline readings, subjects received an intra-arterial infusion of norepinephrine (50 ng/min to assess endothelium-independent vasoconstriction) for 24 minutes followed by a 16-minute washout period and then NG-monomethyl-l-arginine (L-NMMA 1 mg/min; surrogate measurement of basal endothelial NO production) for a further 24 minutes. Again, FBF ratio was measured every 8 minutes throughout, and the final (steady-state) measurement for each infusate was used in subsequent analyses.
To compare results among the 3 groups, summary measures were calculated where appropriate, and unpaired t tests were used with correction for multiple comparisons where appropriate. Data from all 3 groups were pooled for univariate correlation analysis after ensuring (by formal comparison of group regression lines) that the associations were not spurious as the result of clustering of patient data by group.
The procedures were carried out without complications and were well tolerated by all subjects. All results are summarized as mean±SEM unless otherwise stated.
MCR (mL/kg per minute) was 7.22±0.99 (control subjects), 6.32±0.78 (hypertensive subjects), and 5.06±0.53 (type 2 diabetes mellitus) (Figure 1a⇓): Insulin levels achieved during the clamp studies were correspondingly 847±50, 904±65, and 940±50 pmol/L. These small differences were not significantly different.
Local Insulin/Glucose-Mediated Vasodilation
Percentage change from baseline in FBF ratio was 17.1±5.6% (control subjects), 17.2±5.5% (hypertensive subjects), and 12.3±6.4% (diabetic subjects) (Figure 1b⇑). There were no significant differences among groups. The time course of vasodilation is illustrated in Figure 2⇓. Although there was a trend for blunting of the vascular action of insulin in the diabetic group, there was no statistically significant difference with respect to the control group (area under the curve, P=0.34).
Insulin Sensitivity Versus Insulin Vasodilation
When data from the 3 groups were pooled and plotted (Figure 3⇓), there was no evidence of group clustering. There was a significant positive association (r=0.46, P<0.05) between the vascular and metabolic effects of insulin.
Vasoreactivity to L-NMMA and Norepinephrine
Percent vasoconstriction to L-NMMA was 37.9±5.1% (control subjects), 37.5±2.3% (hypertensive subjects), and 33.6±2.8% (diabetic subjects) (Figure 1c⇑). There were no significant differences among groups. Percent vasoconstriction to norepinephrine was 20.9±6.2% (control subjects), 21.9±5.6% (hypertensive subjects), and 29.4±5.6% (diabetic subjects). Again, there were no significant differences among groups.
Insulin Sensitivity Versus L-NMMA Vasoconstriction
Pooled correlation analysis (n=27) revealed that MCR was significantly positively related to percent L-NMMA vasoconstriction (r=0.44, P<0.05) but not to percent norepinephrine vasoconstriction (r=−0.35) (Figure 4⇓).
Insulin Vasodilation Versus L-NMMA Vasoconstriction
Pooled correlation analysis (n=27) revealed that insulin/glucose-mediated vasodilation was significantly positively related to percent L-NMMA vasoconstriction (r=0.52, P<0.01) but not to percent norepinephrine vasoconstriction (r=−0.15) (Figure 5⇓).
Insulin Sensitivity and L-NMMA Vasoconstriction Versus BMI and HDL-Cholesterol
As expected, insulin sensitivity was associated with BMI (r=−0.67, P<0.001) and HDL-cholesterol (r=0.44, P<0.05). However, there was no significant association between percent L-NMMA vasoconstriction and either BMI (r=0.22) or HDL-cholesterol (r=−0.05).
In this study, we have shown for the first time clear associations between (1) the metabolic and vascular effects of insulin, (2) insulin sensitivity and endothelial function, and (3) insulin vasodilation and endothelial function in normal subjects and patients with type 2 diabetes or hypertension. Of additional interest was the observation that the time course of insulin-mediated vasodilation (Figure 2⇑) was similar in these healthy volunteers (56 years of age) to that obtained in younger men (26 years of age) who underwent a similar protocol,5 although the magnitude of vasodilation was reduced by half, consistent with the expected effect of aging. Importantly, there was no evidence of any difference in the relations of the above variables in the 3 groups of subjects. Although there is no indication of causality from correlation analyses, the similarity of the associations among the patient and control groups suggests that functional coupling exists independent from the influence of either blood pressure or ambient glucose level. Indeed, this consistency suggests that there is a clear physiological mechanism that underlies the common association of insulin resistance and endothelial dysfunction in metabolic and cardiovascular disorders. Although no significant differences were observed for these variables among the 3 groups, it must be acknowledged that the small numbers in each group render these comparisons prone to type II statistical error.
Much recent debate has focused on whether endothelial dysfunction might be a primary determinant of insulin action,15 16 with the proposal that the association between the two represents a defect in delivery of glucose to metabolically active tissue caused by reduced blood flow. However, it is also possible that a common antecedent could produce parallel defects in insulin action and endothelial function: An adverse lipid environment is the strongest candidate for this role,14 17 although it has also been suggested that inflammatory mediators may be important18 as well as inherent corticosteroid sensitivity.19
However, we believe that our data strongly support the notion that insulin plays a key role in the maintenance of vascular endothelial function and that inherited or acquired defects in insulin action lead to relative endothelial dysfunction as well as insulin resistance. For example, BMI and HDL-cholesterol were correlated with insulin sensitivity but not with basal endothelial function, as assessed by vasoconstriction to L-NMMA. In contrast, insulin-mediated vasodilation was correlated with both insulin sensitivity and basal endothelial function, leading us to speculate that the vascular action of insulin may be a marker of a key intermediate mechanism linking endothelial function with insulin-mediated glucose uptake. Thus we suggest that insulin is unlikely to be a physiologically important vasodilator but plays a necessary role in the tonic stimulation of endothelial NO production, helping to maintain endothelial health and integrity. This idea is supported by several recent lines of evidence. First, endothelial cells have insulin receptors; in vitro, insulin stimulates NO release from cultured endothelial cells.20 Second, defective endothelium-dependent vasorelaxation has been demonstrated in mice lacking insulin receptor substrate-1.21 Third, Jiang and colleagues22 reported that there were multiple defects in components of the insulin signaling pathway in vascular tissue of the insulin resistant Zucker rat. It was not possible in this study to identify whether the abnormalities were present in endothelial tissue or in vascular smooth muscle, but these data are in keeping with the proposal that primary defects in the insulin signaling pathway, for whatever reason, lead to reduced glucose uptake in skeletal muscle and adipose tissue as well as altered endothelial NO synthesis. Thus primary defects in the insulin signaling pathway may manifest themselves not only as relative metabolic defects but also as “vascular insulin resistance,” resulting in relative endothelial dysfunction and subsequent predisposition to atherothrombotic cardiovascular disease. The results of tissue-specific endothelial receptor knockout experiments in mice are awaited with interest.
This work was supported by a British Heart Foundation Junior Research Fellowship awarded to Dr Cleland (FS/96024). Dr Connell is supported by an MRC Program Grant (9317119).
- Received September 16, 1999.
- Revision received October 19, 1999.
- Accepted November 10, 1999.
Steinberg HO, Brechtel G, Johnson A, Fineberg N, Baron AD. Insulin-mediated skeletal muscle vasodilation is nitric oxide dependent: a novel action of insulin to increase nitric oxide release. J Clin Invest. 1994;94:1172–1179.
Petrie J, Ueda S, Webb DJ, Elliott HL, Connell JMC. Endothelial nitric oxide production and insulin sensitivity: a physiological link with implications for pathogenesis of cardiovascular disease. Circulation. 1996;93:1331–1333.
Cleland SJ, Petrie JR, Ueda S, Elliott HL, Connell JMC. Insulin-mediated vasodilation and glucose uptake are functionally linked in humans. Hypertension. 1999;33(suppl 2):554–558.
Serne EH, Stehouwer CDA, Ter Maaten JC, Ter Wee PM, Rauwerda JA, Donker AJM, Gans ROB. Microvascular function relates to insulin sensitivity and blood pressure in normal subjects. Circulation. 1999;99:896–902.
Laakso M, Edelman SV, Brechtel G, Baron AD. Impaired insulin-mediated skeletal muscle blood flow in patients with NIDDM. Diabetes. 1992;41:1076–1083.
Raitakari M, Nuutila P, Ruotsalainen U, Laine H, Teras M, Iida H, Makimattila S, Utriainen T, Oikonen V, Sipila H, Haaparanta M, Solin O, Wegelius U, Knuuti J, Yki-Jarvinen H. Evidence for dissociation of insulin stimulation of blood flow and glucose uptake in human skeletal muscle. Diabetes. 1996;45:1471–1477.
Laakso M, Edelman SV, Brechtel G, Baron AD. Decreased effect of insulin to stimulate skeletal muscle blood flow in obese man: a novel mechanism for insulin resistance. J Clin Invest. 1990;85:1844–1852.
Baron AD, Steinberg HO. Endothelial function, insulin sensitivity, and hypertension. Circulation. 1997;96:725–726.
Yudkin JS, Stehouwer CDA, Emeis JJ. C-reactive protein in healthy subjects: associations with obesity, insulin resistance and endothelial dysfunction: a potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol. 1999;19:972–978.
Abe H, Yamada N, Kamata K, Kuwaki T, Shimada M, Osuga J, Shionoiri F, Yahagi N, Kadowaki T, Tamemoto H, Ishibashi S, Yazaki Y, Makuuchi M. Hypertension, hypertriglyceridemia, and impaired endothelium-dependent vascular relaxation in mice lacking insulin receptor substrate-1. J Clin Invest. 1998;101:1784–1788.