Sympathetically mediated increases in cardiac output, not restraint of peripheral vasodilation, contribute to blood pressure maintenance during hyperinsulinemia

Abstract

Vasodilatory effects of insulin support the delivery of insulin and glucose to skeletal muscle. Concurrently, insulin exerts central effects that increase sympathetic nervous system activity (SNA), which is required for the acute maintenance of blood pressure (BP). Indeed, in a cohort of young healthy adults, herein we show that intravenous infusion of insulin increases muscle SNA while BP is maintained. We next tested the hypothesis that sympathoexcitation evoked by hyperinsulinemia restrains insulin-stimulated peripheral vasodilation and contributes to sustaining BP. To address this, a separate cohort of participants were subjected to 5-s pulses of neck suction (NS) to simulate carotid hypertension and elicit a reflex-mediated reduction in SNA. NS was conducted before and 60 min following intravenous infusion of insulin. Insulin infusion caused an increase in leg vascular conductance and cardiac output (CO; P < 0.050), with maintenance of BP (P = 0.540). As expected, following NS, decreases in BP were greater in the presence of hyperinsulinemia compared with control (P = 0.045). However, the effect of NS on leg vascular conductance did not differ between insulin and control conditions (P = 0.898). Instead, the greater decreases in BP following NS in the setting of insulin infusion paralleled with greater decreases in CO (P = 0.009). These findings support the idea that during hyperinsulinemia, SNA-mediated increase in CO, rather than restraint of leg vascular conductance, is the principal contributor to the maintenance of BP. Demonstration in isolated arteries that insulin suppresses α-adrenergic vasoconstriction suggests that the observed lack of restraint of leg vascular conductance may be attributed to sympatholytic actions of insulin.

NEW & NOTEWORTHY We examined the role of sympathetic activation in restraining vasodilatory responses to hyperinsulinemia and sustaining blood pressure in healthy adults. Data are reported from two separate experimental protocols in humans and one experimental protocol in isolated arteries from mice. Contrary to our hypothesis, the present findings support the idea that during hyperinsulinemia, a sympathetically mediated increase in cardiac output, rather than restraint of peripheral vasodilation, is the principal contributor to the maintenance of systemic blood pressure.

INTRODUCTION

Beyond its metabolic effects, insulin has vasodilatory actions that increase delivery of insulin and glucose to target tissues, including skeletal muscle (4–7, 15, 32, 40, 57, 61, 69). In isolated arteries, the vasomotor effects of insulin are dependent on the balance between endothelial-derived nitric oxide (NO)-mediated vasodilation and endothelin-1-mediated vasoconstriction, with the net effect typically being vasodilation in the healthy state (16, 17, 31). However, in vivo, endothelin-1 is not the sole opposing force to insulin-stimulated NO-mediated vasodilation. Indeed, the hemodynamic effects of insulin reflect an integrated response of neurohumoral and endothelial-derived vasoactive signals (47, 48). That is, insulin also has centrally mediated effects that activate the sympathetic nervous system, which can limit insulin-induced vasodilation via α-adrenergic receptor-mediated vasoconstriction in healthy young adults (1, 51, 62). In support of this, patients with regional sympathectomy exhibit a more rapid NO-mediated vasodilation in response to insulin in the denervated limb than in the innervated limb (55). Accordingly, via central and peripheral mechanisms, insulin mediates several opposing hemodynamic actions that may serve to concurrently increase skeletal muscle blood flow (and subsequent delivery of insulin and glucose) and induce vasoconstriction in select vascular beds to maintain blood pressure.

However, the extent to which sympathetic nerve activity (SNA) limits insulin-stimulated skeletal muscle blood flow remains largely unknown. Herein our approaches for studying the role of SNA in modulating vasodilatory responses to insulin involved direct measures of SNA using the gold standard technique of microneurography (59), as well as a transient “silencing” of SNA via application of neck suction (NS), which selectively loads the carotid baroreceptors (i.e., simulated carotid hypertension). Neck suction is a method commonly used in human physiology research to study carotid baroreflex control of blood pressure (11) and was employed here, for the first time, as a tool to examine the role of SNA in modulating insulin-stimulated leg blood flow. We hypothesized that sympathoexcitation evoked by hyperinsulinemia would restrain insulin-stimulated leg blood flow and conductance, contributing to the maintenance of blood pressure. More specifically, we hypothesized that during systemic insulin infusion, reducing SNA via neck suction would produce greater peripheral vasodilation and reduction in blood pressure relative to neck suction under basal (nonhyperinsulinemic) conditions. Furthermore, to examine the role of insulin stimulation in modulating α-adrenergic receptor-mediated vasoconstriction, a follow-up proof-of-concept experiment was performed in mouse isolated arteries.

METHODS

Human participants.

All participants were young adults (<40 yr of age), healthy, nonobese [body mass index (BMI) <30 kg/m²], normoglycemic (fasting glucose <100 mg/dL), and nonsmokers without chronic diseases and taking no medications known to affect endocrine, cardiovascular, or autonomic function. Women were not pregnant (confirmed by negative pregnancy test on the morning of the study visit) and were studied in the self-reported early follicular phase of the menstrual cycle (days 1–7) or placebo phase of oral contraceptive use. Participants were asked to refrain from alcohol, caffeine, and exercise for 24 h and fast for 12 h before the study visit following recently published guidelines (36). Written informed consent was obtained from all participants, and all experiments and procedures were approved by the Institutional Review Board at the Mayo Clinic (protocol 1, no. 13-001237) and the University of Missouri (protocol 2, no. 2013126) and conformed to the Declaration of Helsinki.

Experimental protocol 1 in humans: insulin-stimulated muscle SNA.

A subset of data from protocol 1 were from previously published work (2, 37); however, the analyses and results included here are specific to the unique hypothesis raised. Participants were admitted to the Clinical Research and Trials Unit at the Mayo Clinic, and the study day began with instrumentation at 0800. Participants rested supine during instrumentation. An intravenous catheter was placed in the dominant arm for insulin and glucose infusion, and an arterial catheter (20 gauge, 5 cm) was placed in the nondominant arm using aseptic technique under local anesthesia (2% lidocaine). Beat-to-beat heart rate and blood pressure were continuously monitored using a lead II electrocardiogram (GE Datex-Ohmeda Cardiocap/5; GE Healthcare, Chicago, IL) and brachial arterial catheter (TruWave Pressure Transducer; Edwards Lifescience, Irvine, CA), respectively.

Standard microneurography procedures and techniques (24) were used to record multiunit postganglionic muscle SNA from the peroneal nerve, posterior to the fibular head. The peroneal nerve was localized with transcutaneous stimulation and two-dimensional ultrasonography (12). A muscle sympathetic fascicle was identified when taps on the muscle belly or passive muscle stretch evoked mechanoreceptive impulses and no afferent neural response was evoked by skin stimuli. The recorded signal was amplified 100,000-fold, band-pass filtered (700–2,000 Hz), rectified, and integrated (resistance-capacitance integrator circuit, time constant 0.1 s) using a nerve traffic analyzer (662C-3; Bioengineering of University of Iowa).

Following a period of quiet rest, insulin (Novolin R) was infused intravenously at a steady rate of 1.0 mU·kg fat-free mass⁻¹·min⁻¹, an infusion rate that produces circulating levels of insulin that mimic the postprandial state. Fat-free mass was determined via DEXA. Blood glucose was determined every ~5 min at bedside using the glucose oxidase method (GM9 Glucose Analyzer, Analox Instruments, Stourbridge, UK; 2900D Biochemistry Analyzer, Yellow Springs Instruments, Yellow Springs, OH) and maintained at baseline levels, achieved by variable infusion rates of a dextrose solution. Plasma was also obtained and stored at −80°C for analysis of insulin, epinephrine, and norepinephrine. Analyses were performed by the Immunochemistry Core Laboratory of the Clinical Research and Trials Unit and the Department of Laboratory Medicine and Pathology at the Mayo Clinic. Plasma insulin was assessed using a two-site immunoenzymatic assay performed on the Dxl automated immunoassay system (Beckman Instruments, Chaska, MN). Plasma catecholamines were measured by reverse-phase high-performance liquid chromatography with electrochemical detection after extraction with activated alumina. Steady-state measures were collected at baseline and after 60 min of the insulin infusion.

Experimental protocol 2 in humans: insulin-stimulated blood flow and mitigation of muscle SNA by neck suction.

Before participation, individuals were familiarized with the study procedures including the neck suction device. On the study day, participants were admitted to the Clinical Research Center at the University of Missouri at 0800. Participants rested supine during instrumentation, which included a three-lead electrocardiogram (Bio Amp FE132; ADInstruments, Colorado Springs, CO) and noninvasive beat-to-beat blood pressure via finger photoplethysmography (Human NIBP Controller ML282; ADInstruments, Colorado Springs, CO) calibrated to upper arm blood pressure (automated sphygmomanometry). Femoral artery diameter and blood velocity were measured using Doppler ultrasound (iE33; Philips Medical Systems, Bethesda, MD). An 11-MHz linear array transducer was placed over the superficial femoral artery. Simultaneous diameter and velocity signals were obtained in duplex mode at a pulsed frequency of 5 MHz and corrected with an insonation angle of 60°. Sample volume was adjusted to encompass the entire lumen of the vessel without extending beyond the walls, and the cursor was set midvessel.

An intravenous catheter was placed in the antecubital vein of both arms, one for insulin and glucose infusion and the other for periodic blood sampling. Insulin (Humulin R) was infused intravenously to reach a steady rate of 40 mU·m⁻² body surface area·min⁻¹, an infusion rate that elicited similar circulating levels of insulin as were evoked in protocol 1. As in protocol 1, blood glucose was determined every ~5 min at bedside (YSI 2300 STAT PLUS glucose analyzer) and maintained at baseline levels by infusing a dextrose solution at a variable rate. Plasma was also obtained and stored at −80°C for analysis of insulin, which was assessed using a commercially available kit (cat no. 80-INSHU-E10.1; ALPCO, Salem, NH).

Following instrumentation, participants were fitted with a malleable neck collar that encircled the anterior two-thirds of the neck for the application of neck suction (BaroCollar; Physiology Research Instruments, Austin, TX). Appropriate neck chamber placement was ensured by first fitting the subject on the basis of observed neck size and then performing a calibration trial. During neck suction testing, 5-s pulses of −60 mmHg of suction were applied to selectively load (simulated carotid hypertension) the carotid baroreceptors. The 5-s period of neck suction provides reflex activation of carotid baroreceptors and allows for the examination of peripheral vascular responses during a baroreflex-mediated fall in SNA (11, 13, 14, 18, 29, 30, 42, 43, 45, 50, 52, 64). A variable pressure source was used to generate the changes in neck collar pressure, and these changes were delivered to the collar through large-bore two-way solenoid valves. To accurately quantify the stimulus applied, a pressure transducer was connected to a port on the collar. To minimize respiratory-related modulation of heart rate, the 5-s pulses of neck suction were delivered to the carotid sinus during controlled breathing. Specifically, participants were instructed to breathe along to a metronome set at ~15 breaths/min, consistent with work published previously (23, 26, 39, 49). Respiration was assessed by use of a Piezo Respiratory Belt Transducer (MLT1132; ADInstruments, Colorado Springs, CO).

Following instrumentation and a period of quiet rest, baseline blood samples and cardiovascular measurements were collected. During the recording period, five trials of neck suction (sustained 5-s pulses at −60 mmHg) were delivered, each separated by ~2 min, and results from the five trials were averaged (11, 42). This approach has been shown previously to transiently reduce muscle SNA (11, 13, 14, 18, 29, 30, 42, 45, 50). Following baseline neck suction, insulin infusion began. Sixty minutes following the start of the insulin infusion, five trials of neck suction (sustained 5-s pulses at −60 mmHg) were repeated, again separated by ~2 min.

Analysis in protocols 1 and 2.

Data from both protocols were collected using the PowerLab data acquisition system (analog-to-digital converter; ADInstruments, Colorado Springs, CO) with a sampling rate of 1,000 Hz. Stroke volume was estimated from the arterial blood pressure waveform using the Modelflow method through LabChart (LabChart; ADInstruments, Sydney, NSW, Australia), which incorporates age and sex. Cardiac output and total peripheral resistance were calculated. In protocol 1, sympathetic neurograms were analyzed using a semiautomated program (Ensemble-C; Elucimed, Ltd.) by a single investigator (D. W. Jacob), and muscle SNA was expressed as burst frequency (bursts/min), burst incidence (bursts/100 heartbeats), and normalized burst area [arbitrary units (AU)/min]. In protocol 2, ultrasound recordings were analyzed off-line using specialized edge detection software (Cardiovascular Suite; Quipu srl, Pisa, Italy). Blood flow was calculated from continuous diameter and mean blood velocity recordings using the following equation: 3.14 × [diameter (cm)/2]² × mean blood velocity (cm/s) × 60 and reported as milliliters per minute. Vascular conductance was calculated as blood flow (mL/min) ÷ mean arterial blood pressure (mmHg) × 100 and reported as milliliters per minute per 100 mmHg. Hemodynamic variables were analyzed second by second and data time-aligned for averages across trials. Main outcome variables include a relative (%) change in hemodynamic parameters following acute neck suction, where a 10-s average immediately before the onset of NS was equal to 100%. Data are from the initiation of neck suction and include the 5-s stimulus and 5-s poststimulus recovery.

Experimental protocol 3 in isolated arteries from mice: insulin-induced modulation of α-adrenergic receptor-mediated vasoconstriction.

On the basis of findings from protocol 2 in humans, we designed a follow-up proof-of-concept experiment in isolated arteries from mice to test the hypothesis that vascular exposure to insulin can suppress α-adrenergic receptor-mediated vasoconstriction. Furthermore, we tested whether this phenomenon is endothelium dependent and reliant on the phosphatidylinositol 3-kinase (PI3K)/Akt arm of the insulin signaling pathway.

All animal study procedures received prior approval by the University of Missouri Institutional Animal Care and Use Committee. The University of Missouri is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. For this experiment, isolated aortic rings were used from 20 wild-type [28.4 ± 0.6 g body wt, mean ± SE; C57BL/6J background bred at the University of Missouri] and 4 diabetic db/db (48.3 ± 0.7 g body wt, no. 00697; The Jackson Laboratory, Bar Harbor, ME) 8-wk-old male mice. Aortas were harvested and cleaned of perivascular adipose tissue in ice-cold physiological saline solution (pH 7.40) and cut into four 2-mm segments, such that experiments could be performed in duplicate within each experimental condition. The ability to obtain multiple rings from a single aorta represents a strength of this model. Aortic rings were then mounted in wire myograph organ bath chambers (620M; Danish Myo Technology, Hinnerup, Denmark) containing warmed physiological saline solution gassed with 95% O₂-5% CO₂ and maintained at 37°C as previously described (21, 66). Aortic rings were treated with 80 mM KCl to ensure viability. Next, aortas were incubated with or without insulin (100 nM, Humulin R; Eli Lilly, Indianapolis, IN) for 30 min before stimulation with increasing concentrations of phenylephrine (an α₁-adrenergic receptor agonist; 10⁻⁹ to 10⁻⁵ M) while insulin (or lack thereof) remained in the bath. Additional mechanistic experiments were completed in aortic rings that underwent endothelium denudation, aortic rings treated with an NO synthase inhibitor [N^ω-nitro-l-arginine methyl ester (l-NAME), 300 µM, no. N5751; Millipore-Sigma, St. Louis, MO], and aortic rings treated with a PI3K inhibitor (wortmannin, 100 nM, no. 19545-26-7; Millipore-Sigma). Aortic rings were mechanically denuded by gently rubbing the intima of the blood vessel against the myograph pins. Endothelial denudation was confirmed by the lack of a relaxation response to acetylcholine (10⁻⁵ M) at the end of the phenylephrine dose-response curve. Vasoconstriction responses to phenylephrine were normalized to KCl-induced constriction. Area under the curve (AUC) for each phenylephrine dose-response curve was also calculated using the trapezoidal rule. It should be acknowledged that aortic rings were selected as model of choice for this proof-of-concept experiment and that vessels were exposed to a high range of phenylephrine concentrations. To most appropriately establish the role of insulin in mitigating SNA-induced peripheral vasoconstriction, experiments should be recapitulated in isolated resistance arteries as well as in vivo under more physiological conditions (e.g., using sympathoexcitatory maneuvers).

Statistical analysis.

In protocols 1 and 2, the effect of insulin (i.e., baseline vs. hyperinsulinemia) and neck suction on main outcome variables was assessed using a one-way or two-way repeated measures analysis of variance (ANOVA), as appropriate, and pairwise multiple comparisons were made using Student–Newman–Keuls method. No sex differences were detected on primary outcome variables in protocols 1 and 2; therefore, data from male and female participants were pooled for analysis. In protocol 3, a two-way repeated measures ANOVA was used to assess the influence of insulin on the vasoconstrictor responses to increasing phenylephrine concentrations under each different condition. Within condition, comparisons for AUC were made using a paired two-tailed Student’s t test. An α of P < 0.05 was considered statistically significant. Data are reported as means ± SE.

RESULTS

Hyperinsulinemia robustly increases muscle SNA without impacting mean arterial blood pressure.

Twenty healthy men (n = 13) and women (n = 7) completed protocol 1. Participant characteristics are reported in Table 1. Plasma insulin and blood glucose concentrations during systemic infusion of insulin are presented in Fig. 1A and demonstrate successful achievement of hyperinsulinemia (P < 0.001) with maintenance of fasted blood glucose levels (P = 0.491). Resting heart rate increased in response to insulin infusion (P = 0.005, Fig. 1B), whereas mean arterial blood pressure remained unchanged (P = 0.685, Fig. 1C). Data from Modelflow showed an increase in cardiac output (P = 0.006, Fig. 1D) and reduction in total peripheral resistance (P = 0.009, Fig. 1E) during insulin infusion. No change in stroke volume was observed (87 ± 3 to 88 ± 3 mL/beat; P = 0.473). Muscle SNA burst frequency (P = 0.002, Fig. 1F), burst incidence (29.9 ± 1.9 to 37.3 ± 3.3 bursts/100 heartbeats, P = 0.007), and normalized burst area (4,121 ± 227 to 5,028 ± 372 AU/min, P = 0.002) all increased from baseline in response to insulin infusion. Similarly, insulin infusion increased plasma catecholamine levels (norepinephrine, 164 ± 15 to 208 ± 16 pg/mL, P < 0.001; epinephrine, 47 ± 5 to 61 ± 6 pg/mL, P = 0.015; n = 18 participants).

Table 1.

Participant demographics

	Protocol 1	Protocol 2
Men, women, n	13, 7	5, 5
Age, yr	28 ± 1	29 ± 2
Height, cm	178 ± 3	170 ± 2
Weight, kg	78 ± 3	69 ± 3
Body mass index, kg/m2	25 ± 1	24 ± 1

Values are means ± SE; n = no. of participants.

Fig. 1.

Hyperinsulinemia robustly increases muscle sympathetic nerve activity without impacting mean arterial blood pressure. In protocol 1, steady-state measures were collected at baseline (BL) and after 60 min of the insulin infusion (INS). Plasma insulin and blood glucose concentrations (A), heart rate (B), mean blood pressure (BP; C), cardiac output (CO; D), total peripheral resistance (TPR; E), and muscle sympathetic nerve activity (MSNA; F). Representative original traces from a male participant (22 yr old, body mass index 25.3 kg/m²; G). Data are reported as means ± SE (n = 20 participants). In A and F, error bars are within symbols. *Significance (P < 0.05) from BL.

Sympathetically mediated increases in cardiac output, rather than restraint of peripheral vasodilation, contribute to the maintenance of blood pressure during hyperinsulinemia.

Ten healthy men (n = 5) and women (n = 5) completed protocol 2. Participant characteristics are reported in Table 1. Plasma insulin and blood glucose concentrations during systemic infusion of insulin are presented in Fig. 2A, and as in protocol 1, support the successful achievement of hyperinsulinemia (P < 0.001) with maintenance of fasted blood glucose levels (P = 0.718). Consistent with protocol 1, insulin infusion increased resting heart rate (P < 0.001, Fig. 2B), whereas mean arterial blood pressure was maintained (P = 0.540, Fig. 2C). Also in alignment with protocol 1, insulin infusion increased cardiac output (P = 0.002, Fig. 2D) and tended to reduce total peripheral resistance (P = 0.068, Fig. 2E), with no change in stroke volume (92 ± 6 to 93 ± 6 mL/beat, P = 0.659). Femoral blood flow (Fig. 2F) and vascular conductance (Fig. 2G) increased from baseline in response to insulin infusion (P = 0.005 and P = 0.016, respectively).

Fig. 2.

Sympathetically mediated increases in cardiac output, rather than restraint of peripheral vasodilation, contribute to the maintenance of blood pressure during hyperinsulinemia. In protocol 2, cardiovascular measures were continuously collected in response to 5-s pulses of neck suction (NS) to simulate carotid hypertension and elicit a reflex-mediated reduction in sympathetic activation. NS was conducted before [i.e., baseline (BL)] and 60 min following the start of the insulin infusion (INS). During the recording period, five trials of NS (sustained 5-s pulses at −60 mmHg) were delivered, each separated by a 2-min washout. Results from the five trials were averaged. Plasma insulin and blood glucose concentrations (A), heart rate (B), mean blood pressure (BP; C), cardiac output (D), total peripheral resistance (TPR; E), femoral blood flow (F), and vascular conductance (VC; G). For B–G, panels at left display the effect of insulin infusion, and panels at right display the relative effect of NS over a 10-s period at BL (○) vs. during INS (yellow circles). Data are reported as means ± SE (n = 10 participants). In A, error bars are within symbols. *Significance (P < 0.05) from BL.

Neck suction elicited a reduction in mean arterial blood pressure that was greater in the presence of hyperinsulinemia compared with control (main effect of neck suction, P < 0.001; interaction of insulin and neck suction, P = 0.045, Fig. 2C). Neck suction also resulted in larger decreases in heart rate (main effect of neck suction, P < 0.001; interaction of insulin and NS; P = 0.026, Fig. 2B) and cardiac output (main effect of neck suction, P < 0.001; interaction of insulin and neck suction, P = 0.009, Fig. 2D) during hyperinsulinemia compared with control. The effect of neck suction on stroke volume did not differ between insulin and control conditions (main effect of neck suction, P < 0.001; interaction of insulin and neck suction, P = 0.563). Furthermore, the effect of neck suction on total peripheral resistance, femoral blood flow, and femoral vascular conductance did not differ between insulin and control conditions (main effect of neck suction, P < 0.001, P < 0.001, and P < 0.001, respectively; interaction of insulin and neck suction, P = 0.166, P = 0.602, and P = 0.898, respectively, Fig. 2, E–G).

Insulin buffers α-adrenergic vasoconstriction, and this effect is endothelium dependent and reliant on an intact insulin signaling pathway.

In naïve (i.e., untreated) aortic rings, insulin exposure blunted phenylephrine-induced vasoconstriction (AUC, P = 0.047, Fig. 3A). This insulin-suppressing effect on phenylephrine-induced constriction was lost in aortic rings devoid of endothelium (AUC, P = 0.633, Fig. 3B), in aortic rings treated with NO synthase inhibitor l-NAME (AUC, P = 0.212, Fig. 3C), in aortic rings treated with PI3K inhibitor wortmannin (AUC, P = 0.702, Fig. 3D), and in aortic rings from diabetic db/db mice (AUC, P = 0.412, Fig. 3E).

Fig. 3.

Insulin buffers α-adrenergic vasoconstriction, and this effect is endothelium dependent and reliant on an intact insulin signaling pathway. In protocol 3, aortic rings were incubated with (+) or without (−) insulin (INS, 100 nM) for 30 min before stimulation with increasing concentrations of phenylephrine (Phe) while insulin (or lack thereof) remained in the bath. Experiments were performed in control (i.e., untreated) aortic rings (n = 5 mice; A), aortic rings devoid of endothelium (n = 5 mice; B), aortic rings treated with NO synthase inhibitor N^ω-nitro-l-arginine methyl ester (l-NAME, n = 5 mice; C), aortic rings treated with phosphatidylinositol 3-kinase (PI3K) inhibitor wortmannin (n = 5 mice; D), and aortic rings from diabetic db/db mice (n = 4 mice; E). Data are reported as means ± SE. a.u., arbitrary units; AUC, area under the curve. *Significant interaction (P = 0.018); #significance (P = 0.047) from Without INS (−).

DISCUSSION

Herein we sought to investigate the role of sympathetic activation in restraining vasodilatory responses to hyperinsulinemia and consequently sustaining blood pressure in healthy adults. Blood pressure is a highly regulated variable by which acute, beat-to-beat changes are achieved via sympathetic innervation of the peripheral vasculature as well as combined sympathetic/parasympathetic activity at the level of the heart. As such, the maintenance of blood pressure in the presence of insulin must result from changes in total peripheral resistance and/or cardiac output (i.e., heart rate and stroke volume). Contrary to our hypothesis, the present findings support the idea that during hyperinsulinemia, neurally mediated increase in cardiac output, rather than restraint of peripheral vasodilation, is the principal contributor to the maintenance of systemic blood pressure.

During the postprandial state, insulin secretion causes vasodilator actions, which are of paramount importance for nutrient delivery to metabolically active tissues, particularly skeletal muscle, and ensuing glycemic control (4–7, 15, 32, 40, 57, 61, 69). Mechanistically, at the endothelial cell level, insulin binds the insulin receptor (IR) and activates downstream IR substrate 1/2/PI3K signaling, ultimately leading to phosphorylation of endothelial NO synthase, increased NO production, and subsequent vasodilation (31, 57, 68). However, in the face of such vasodilator effects of insulin, countercurrent mechanisms are required for preservation of blood pressure during hyperinsulinemia. Indeed, as Loring Rowell (54) once proclaimed in the context of exercise-induced hyperemia, “skeletal muscle is a ‘sleeping giant’ whose blood flow must be under tonic vasoconstrictor constraint if hypotension is to be averted.” Here we reasoned that this principle would be germane to the setting of hyperinsulinemia. That is, we posited that sympathoexcitation caused by hyperinsulinemia likely limits vasodilatory responses in skeletal muscle via α-adrenergic receptor-mediated vasoconstriction, thus contributing to the maintenance of blood pressure. Congruently with this notion, we first corroborated in protocol 1 that systemic infusion of insulin robustly increases muscle SNA while blood pressure remains stable. Next, we examined the degree to which silencing of SNA during insulin stimulation would favor peripheral vasodilation. Specifically, in protocol 2 we posited that during systemic insulin infusion, acute reductions in SNA via neck suction would produce greater peripheral vasodilation and a reduction in blood pressure relative to neck suction under the basal, noninsulinized control state.

We found that following neck suction, the reduction in blood pressure was greater during systemic insulin infusion compared with that in the absence of insulin, supporting a role for sympathetic activation in sustaining blood pressure during acute hyperinsulinemia. However, in contrast to our hypothesis, the effect of neck suction on femoral blood flow and vascular conductance did not differ between insulin and control conditions, nor was there an observable effect on total peripheral resistance. Instead, we report that the greater decreases in blood pressure following neck suction, while in the setting of high insulin levels, were accompanied by greater decreases in cardiac output. This finding, combined with the observation that cardiac output was markedly increased during insulin infusion in both protocols, supports the view that sympathetically mediated increases in cardiac output (primarily via an increase in heart rate), as opposed to restraint of peripheral vasodilation, are critical to the preservation of blood pressure during hyperinsulinemia. Insulin-mediated increases in heart rate are consistent with prior work in humans (1, 9, 10, 63) and are associated with a relative cardiac sympathetic predominance in the setting of lower vagal influence (8, 44, 60). From a translational standpoint, this increase in cardiac sympathetic tone is necessary to maintain blood pressure after a meal (53, 58). For example, patients with autonomic failure exhibit a significant fall in blood pressure after a meal that is attenuated by indomethacin (a β-adrenergic receptor agonist; 53).

Because hyperinsulinemia provokes a prominent increase in muscle SNA (e.g., Fig. 1; 1, 9, 35, 63, 67), we were surprised by the finding that skeletal muscle blood flow responses to insulin were not under a substantial SNA-mediated vasoconstrictor constraint. Next, we put forth three ideas that may reconcile discrepancies between the present and expected findings: 1) nonuniform increases in SNA in response to hyperinsulinemia, 2) sympatholytic actions of insulin, and/or 3) sympathetically mediated vasodilation in the setting of high insulin levels.

First, it is possible that during acute hyperinsulinemia, other vascular beds beyond skeletal muscle (e.g., splanchnic) may play a more critical role in sympathetically mediated control of blood pressure. In this regard, there is evidence for nonuniform increases in SNA in response to hyperinsulinemia. For example, preclinical animal models show no change in renal SNA with high insulin levels despite significant increases in lumbar SNA, and similar observations have been made in humans (3, 22, 38). Second, insulin signaling, besides exerting direct endothelium-dependent NO-mediated vasodilatory effects, may attenuate sympathetic vasoconstriction, a term known as sympatholysis. As such, it is conceivable that the lack of SNA-mediated peripheral vasoconstrictor constraint apparent in the present study in healthy, insulin-sensitive individuals could be attributed to insulin-induced “lysing” of α-adrenergic receptor-mediated vasoconstriction. To begin to explore this idea, we designed a proof-of-concept experiment (i.e., protocol 3) to determine whether, indeed, vascular exposure to insulin suppresses α-adrenergic vasoconstriction. In agreement with this postulate, we show that insulinization of isolated aortic rings blunts constriction to phenylephrine, an α₁-adrenergic receptor agonist. Notably, this buffering effect of insulin was only noted in arteries with an intact endothelium and an unrestricted insulin signaling pathway. That is, either endothelium denudation or inhibition of PI3K via wortmannin [a pharmacological model of selective insulin resistance (41, 46)] eliminated the effect of insulin. The insulin-suppressing effect on phenylephrine-induced constriction was also lost in arteries from db/db mice, a well-established murine model of insulin resistance and diabetes (19, 56, 65), as well as in arteries treated with an NO synthase inhibitor. Collectively, these findings are consistent with the idea that vascular insulin signaling can blunt sympathetic vasoconstriction (33) and that the sympatholytic factor in this context is endothelial derived and NO dependent. In line with this, previous work in other models has led to the conclusion that the endothelium is an important modulator of sympathetic vasoconstriction (25) and that NO may in fact be sympatholytic (27, 28). Consistent with the data from db/db mice, the effects of hyperinsulinemia on the mechanisms studied herein may be altered when in a diseased state (e.g., type 2 diabetes). Third, another explanation that likely coexists with that stated above is that insulin augments β-adrenergic receptor-mediated vasodilation, a concept with precedence in the literature (20, 33, 34). Accordingly, it is possible that any increase in muscle SNA potentiates insulin-stimulated vasodilation through β-adrenergic receptor activation and consequent downstream NO production (20).

In aggregate, our findings in insulin-sensitive adults do not support the role of SNA-mediated vasoconstriction as a mechanism to restrain insulin-induced skeletal muscle vasodilation. Both insulin-induced lysing of α-adrenergic vasoconstriction and insulin-induced potentiation of β-adrenergic vasodilation likely explain this finding. Importantly, because of such observed lack of SNA-mediated restraint of peripheral vasodilation, healthy individuals must heavily rely on increases in cardiac output to support blood pressure during hyperinsulinemia. Indeed, our data do support the role of SNA-mediated increases in cardiac output as the primary mechanism in place to maintain blood pressure in the face of peripheral vasodilation caused by hyperinsulinemia. Further research is needed to determine whether this cardiac output dependency for the maintenance of blood pressure during hyperinsulinemia is absent in patients with insulin resistance who display chronic sympathetic overactivity and impaired vasodilatory responses to insulin.

GRANTS

This work was supported in part by the Margaret W. Mangel Faculty Research Catalyst Fund (J.K.L. and J.P.), American Heart Association Grant 15SDG 25080095 (J.K.L.), and National Heart, Lung, and Blood Institute Grants R00-HL-130339 (J.K.L.) and R01-HL-137769 (J.P.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.K.L. and J.P. conceived and designed research; J.K.L., J.A.S., R.N.S., J.L.H., K.N.H., D.W.J., M.T.M., Z.I.G., B.D.J., T.B.C., C.M.-A., and J.P. performed experiments; J.K.L., J.A.S., R.N.S., J.L.H., K.N.H., D.W.J., and M.T.M. analyzed data; J.K.L., J.A.S., J.L.H., K.N.H., D.W.J., M.T.M., Z.I.G., B.D.J., T.B.C., T.B., C.M.-A., and J.P. interpreted results of experiments; J.K.L., R.N.S., and J.P. prepared figures; J.K.L. and J.P. drafted manuscript; J.K.L., J.A.S., R.N.S., J.L.H., K.N.H., D.W.J., M.T.M., Z.I.G., B.D.J., T.B.C., T.B., C.M.-A., and J.P. edited and revised manuscript; J.K.L., J.A.S., R.N.S., J.L.H., K.N.H., D.W.J., M.T.M., Z.I.G., B.D.J., T.B.C., T.B., C.M.-A., and J.P. approved final version of manuscript.