Tumor Necrosis Factor-α Antagonism Improves Vasodilation During Hyperinsulinemia in Metabolic Syndrome

A total of 16 patients with metabolic syndrome (National Cholesterol Education Program Adult Treatment Panel [NCEP ATP] III criteria) and 13 healthy control subjects, approximately matched for sex and age, were recruited for this study. In all patients, waist circumference was >102 cm in male subjects and >88 cm in female subjects, thus indicating central obesity. Studies consisted of infusion of drugs into the brachial artery and measurement of forearm blood flow responses by means of strain-gauge plethysmography. In Study 1, control subjects and 10 patients with metabolic syndrome received infusion of regular insulin (0.2 mU · kg⁻¹ · min⁻¹); after 45 min of insulin infusion, dose-response curves to graded doses of acetylcholine (ACh) (release of endogenous NO) and sodium nitroprusside (SNP) (exogenous NO donor) were obtained. Thereafter, while keeping insulin infusion constant, patients with metabolic syndrome received infusion of infliximab (200 μg/min for 45 min) and dose-response curves to ACh and SNP were repeated. In Study 2, to assess whether the effect of infliximab on vascular response to ACh might relate to reduction of oxidative stress, six additional patients with metabolic syndrome underwent a first dose-response curve to ACh during hyperinsulinemia alone. Vitamin C was then given at 25 mg/min for 45 min and a second dose-response curve to ACh was obtained. Finally, infliximab infusion was superimposed and a third dose-response curve to ACh was obtained during concurrent administration of vitamin C and infliximab. Statistical analyses were performed by t test and ANOVA, as appropriate. Data are reported as mean ± SEM and P < 0.05 was considered statistically significant.

Patients had higher blood pressure (P < 0.001), plasma cholesterol (P < 0.05), triglycerides (P < 0.05), and glucose (P < 0.01) than control subjects. Baseline insulin was lower in control subjects than in patients (6.2 ± 1.5 vs. 11.3 ± 1.7 μU/ml, respectively; P = 0.03); following insulin administration, venous insulin concentration in the perfused forearm rose to 171 ± 37 in control subjects versus to 224 ± 32 μU/ml in patients (P = 0.44).

The vasodilator response to ACh was blunted in patients compared with control subjects (12.7 ± 1.4 vs. 26.5 ± 3.1 ml · min⁻¹ · dl⁻¹ at the highest dose of ACh, respectively; P < 0.001). Similarly, the vasodilator response to SNP was lower in patients than in control subjects (12.9 ± 1.1 vs. 16.3 ± 0.6 ml · min⁻¹ · dl⁻¹, respectively; P < 0.001). In patients with metabolic syndrome participating in Study 1, infliximab improved the vasodilator response to both ACh (Fig. 1A) and SNP (Fig. 1B). In patients participating in Study 2, vascular response to ACh was significantly enhanced by administration of vitamin C compared with hyperinsulinemia alone (Fig. 1C); under those conditions, however, infusion of infliximab in conjunction with vitamin C did not modify vascular responsiveness to ACh versus vitamin C alone (Fig. 1D).

This study provides the novel finding that TNF-α neutralization improves NO-dependent vasodilation during hyperinsulinemia, thereby suggesting that TNF-α activation is involved in vascular dysfunction of metabolic syndrome.

Overexpression of TNF-α has previously been reported not only in obese adipose tissue (1) and in the skeletal muscle of insulin-resistant animals and humans (2), but also in vascular smooth muscle cells (VSMCs) (3). The decreased responsiveness to both endothelium-dependent and -independent stimuli seen during hyperinsulinemia in our patients, taken in conjunction with the favorable effect of infliximab on responses to both ACh and SNP, suggests that TNF-α activation in metabolic syndrome affects NO-dependent vasodilation through mechanisms other than endothelial dysfunction. This phenomenon might be determined by impaired insulin signaling within VSMCs, thus affecting the facilitatory action physiologically exerted by insulin on vasorelaxation. This hypothesis stems from studies showing that insulin may impact VSMCs’ vasodilator capacity through multiple mechanisms. In particular, impairment of insulin inactivation of the small GTPase RhoA and its target ρ-kinase, leading to decreased vasodilation, has been reported in VSMCs from insulin- resistant animals (4). Because this defect localizes at the same level of PI3-kinase activation (5) at which TNF-α upregulation affects insulin signaling (6), the vascular effect of TNF-α antagonism observed in our patients is consonant with restoration of the facilitatory role of insulin on VSMCs relaxation.

Recent experimental evidence indicates that TNF-α overexpression increases oxidative stress (7) and that TNF-α–induced reactive oxygen species play a causal role in insulin resistance (8). We tested the possibility that increased oxidative stress could also be involved in human TNF-α vasculopathy by use of vitamin C. This antioxidant improved the vasodilator response to ACh during hyperinsulinemia in metabolic syndrome, whereas infliximab infusion on top of vitamin C did not exert additive effects. This suggests that increased oxidative stress is indeed involved in mediating the effects of TNF-α on vasodilation during hyperinsulinemia, a view supported by previous results showing that reactive oxygen species could affect vascular reactivity through changes in receptor function or activity of signaling pathways (9,10).

The beneficial action of infliximab demonstrated in our study suggests a novel mechanism by which TNF-α activation might be involved, via increased oxidative stress, in vascular dysfunction of patients with obesity-related metabolic syndrome; it remains to be elucidated whether interventions targeting cytokines may become an effective strategy for prevention in these patients.