Role of the arterial baroreflex in the sympathetic response to hyperinsulinemia in adult humans

Abstract

Muscle sympathetic nerve activity (MSNA) increases during hyperinsulinemia, primarily attributed to central nervous system effects. Whether peripheral vasodilation induced by insulin further contributes to increased MSNA via arterial baroreflex-mediated mechanisms requires further investigation. Accordingly, we examined baroreflex modulation of the MSNA response to hyperinsulinemia. We hypothesized that rescuing peripheral resistance with coinfusion of the vasoconstrictor phenylephrine would attenuate the MSNA response to hyperinsulinemia. We further hypothesized that the insulin-mediated increase in MSNA would be recapitulated with another vasodilator (sodium nitroprusside, SNP). In 33 young healthy adults (28 M/5F), MSNA (microneurography) and arterial blood pressure (BP, Finometer/brachial catheter) were measured, and total peripheral resistance (TPR, ModelFlow) and baroreflex sensitivity were calculated at rest and during intravenous infusion of insulin (n = 20) or SNP (n = 13). A subset of participants receiving insulin (n = 7) was coinfused with phenylephrine. Insulin infusion decreased TPR (P = 0.01) and increased MSNA (P < 0.01), with no effect on arterial baroreflex sensitivity or BP (P > 0.05). Coinfusion with phenylephrine returned TPR and MSNA to baseline, with no effect on arterial baroreflex sensitivity (P > 0.05). Similar to insulin, SNP decreased TPR (P < 0.02) and increased MSNA (P < 0.01), with no effect on arterial baroreflex sensitivity (P > 0.12). Acute hyperinsulinemia shifts the baroreflex stimulus-response curve to higher MSNA without changing sensitivity, likely due to insulin’s peripheral vasodilatory effects. Results show that peripheral vasodilation induced by insulin contributes to increased MSNA during hyperinsulinemia.

NEW & NOTEWORTHY We hypothesized that elevation in muscle sympathetic nervous system activity (MSNA) during hyperinsulinemia is mediated by its peripheral vasodilator effect on the arterial baroreflex. Using three separate protocols in humans, we observed increases in both MSNA and cardiac output during hyperinsulinemia, which we attributed to the baroreflex response to peripheral vasodilation induced by insulin. Results show that peripheral vasodilation induced by insulin contributes to increased MSNA during hyperinsulinemia.

INTRODUCTION

Ingestion of nutrients stimulates release of insulin by pancreatic β cells, which is critical for peripheral glucose uptake. Beyond its metabolic actions, insulin exerts known effects on the peripheral vasculature, causing vasodilation and thereby facilitating glucose delivery to target organs (1–7). In addition, increases in circulating insulin promote a rise in activity of the sympathetic nervous system (8–14). Insulin-mediated increases in sympathetic nervous system activity directed toward the skeletal muscle (muscle sympathetic nerve activity, MSNA) occurs gradually (8, 13, 15, 16). This gradual increase in MSNA during hyperinsulinemia has been attributed to the time necessary for saturation of receptors (17, 18) and transfer of insulin across the blood-brain barrier (19–21). Consistent with this, canine studies have shown an approximate 30-min delay between elevations in circulating levels of insulin and a rise in insulin within the cerebrospinal fluid (22, 23).

Once in the brain, insulin acts within the arcuate nucleus to increase activity of the sympathetic nervous system via the paraventricular nucleus of the hypothalamus and rostral ventrolateral medulla (24–29). Increases in insulin within the brain of rats also enhance the gain of the arterial baroreflex control of sympathetic nervous system activity (i.e., baroreflex sensitivity) (30, 31)—an important mechanism in the maintenance of arterial blood pressure. To our knowledge, only one study has attempted to translate these findings to humans. Young et al. (14) studied eight men in which insulin was elevated by systemic intravenous infusion and found an increase in the gain of the arterial baroreflex control of MSNA within 30 min. Consistent with work in rodents, the authors attributed these hyperinsulinemia-mediated changes in baroreflex sensitivity to the direct effect of insulin within central neural control pathways (14).

In addition to the direct central effects of insulin on baroreflex function, the peripheral vasodilator effect of insulin has been postulated to contribute to baroreflex-mediated increases in MSNA during acute hyperinsulinemia (15, 32–36). However, whether peripheral vasodilation induced by insulin further contributes to increased MSNA via arterial baroreflex-mediated mechanisms has not been directly examined. Accordingly, we aimed to assess whether the elevation in MSNA during hyperinsulinemia is mediated by its peripheral vasodilator effect on the arterial baroreflex. We first sought to confirm, based on prior work in humans (14), that acute hyperinsulinemia increases the gain of the arterial baroreflex control of MSNA in healthy young adults (i.e., arterial baroreflex sensitivity). For comprehensiveness, we assessed arterial baroreflex control of MSNA with both a high and low dose of insulin infusion. To determine whether the effect of hyperinsulinemia on the arterial baroreflex is primarily peripheral in nature, we conducted two additional protocols to test the following hypotheses: 1) that any effect of systemic insulin on the arterial baroreflex would be lost with coinfusion of a peripheral vasoconstrictor (i.e., phenylephrine); and 2) that the effect of hyperinsulinemia on the arterial baroreflex could be replicated with another peripheral vasodilator (i.e., sodium nitroprusside).

METHODS

Participants

All participants (n = 33, 28 M/5 F) were young adults (18–45 yr of age), healthy (no acute or chronic conditions), nonobese [body mass index (BMI) <30 kg/m²], and nonsmokers taking no medications known to affect endocrine, cardiovascular, or autonomic function. All women (n = 5) were premenopausal and had a confirmed negative pregnancy test on the morning of the study visit. Menstruating women were studied in the self-reported early follicular phase of the menstrual cycle (days 1–7) or placebo phase of oral contraceptive use. All participants were asked to refrain from alcohol, caffeine, and exercise for 24 h and fast (no food or drink, other than water) for a minimum of 4 h before the study visit based on recently published guidelines (37). Written informed consent was obtained from all participants before participation and all experiments and procedures were approved by the Institutional Review Boards at the University of Missouri (Protocol Nos. 2013126 and 2020286) and the Mayo Clinic (Protocol Nos. 13–001237 and 15–002862) and conformed to the Declaration of Helsinki, except registration in a database. Data from a subset of participants in protocol 1 (38), protocol 2 (5), and protocol 3 (39) were included in previous publications testing unrelated hypotheses. Based on data supporting a lack of sex-related differences in insulin-mediated increases in peripheral blood flow (38) and sympathetic activity (40) in healthy, normal-weight males and females, data from both sexes were pooled.

Protocol 1: Effect of Hyperinsulinemia on the Arterial Baroreflex

Participants (n = 13, 10 M/3 F) were admitted to the Clinical Research Center at the University of Missouri between 0800 and 0830. Participants rested supine during instrumentation, which included a 3-lead electrocardiogram (Lead II; Bio Amp FE132, ADInstruments) and noninvasive beat-to-beat blood pressure via finger photoplethysmography (Finapres, Finapres Medical Systems) calibrated to upper arm blood pressure (automated sphygmomanometry). Multiunit postganglionic MSNA was recorded from the peroneal nerve, posterior to the fibular head, using a tungsten microelectrode using standard microneurography procedures and techniques (41, 42). The microneurography electrode was introduced across the skin and placed into the peroneal nerve under ultrasound guidance (43). A reference electrode was positioned subcutaneously at ∼4 cm from the recording electrode. 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 band-pass filtered (700 to 2,000 Hz) and integrated (time constant: 0.1 s).

An intravenous catheter was placed in the antecubital vein of both arms, one for insulin and dextrose infusion and the other for periodic blood sampling. Following a baseline period, two priming infusions of insulin (160 to 80 mU/m² body surface area/min; Humulin R U-100) were given over 10 min, as introduced by DeFronzo and colleagues (44), to achieve an initial overshoot in plasma insulin concentration (high dose) lasting for ∼20 min. The priming infusions were followed by a constant 60-min infusion rate at 40 mU/m² body surface area/min (low dose), mimicking postprandial insulin levels (44). During insulin infusion, blood glucose was determined every five 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 until analysis for insulin using a commercially available kit (ALPCO Cat. No. 80-INSHU-E10.1, Salem, NH). Steady-state measures were collected at baseline (before insulin infusion) and after 15 min (high dose) and 60 min (low dose) of intravenous insulin infusion.

Protocol 2: Effect of Peripheral Vasoconstriction on the Arterial Baroreflex during Hyperinsulinemia

Participants (n = 7, 5 M/2 F) 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 that included a 3-lead electrocardiogram (Lead II, GE Datex-Ohmeda Cardiocap/5, GE Healthcare, Chicago, IL). An intravenous catheter was placed in the dominant arm for insulin and glucose infusion, followed by placement of an arterial catheter (20-gauge, 5 cm using aseptic technique under local anesthesia, 2% lidocaine) in the nondominant arm for blood sampling and blood pressure monitoring. Multiunit postganglionic MSNA was recorded as outlined in Protocol 1: Effect of Hyperinsulinemia on the Arterial Baroreflex. The recorded signal was amplified, band-pass filtered (700 to 2,000 Hz), and integrated (time constant: 0.1 s).

Participants completed 30 min of resting baseline followed by a 90-min hyperinsulinemic-euglycemic infusion. Intravenous insulin was infused at a constant rate of 1.0 mU/kg fat-free mass/min (low dose), mimicking postprandial insulin levels. As in protocol 1, dextrose was infused intravenously at a variable rate to maintain blood glucose at baseline levels. Blood glucose was measured every 5–10 min at bedside (GM9 Glucose Analyzer, Analox Instruments, Stourbridge, United Kingdom; 2900 D Biochemistry Analyzer, Yellow Springs Instruments, Yellow Springs, OH). Plasma was also obtained and stored at −80°C until analysis for insulin. Plasma insulin was assessed using a two-site immunoenzymatic assay performed on the Dxl automated immunoassay system (Beckman Instruments, Chaska, MN). All 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.

Steady-state measures were collected at baseline and during the last 30 min of the insulin infusion. At this time phenylephrine (α₁-adrenergic agonist) was infused intravenously following a quiet rest at a low dose (0.2–0.4 µg/kg/min; average dose = 0.29 ± 0.04 µg/kg/min) selected to achieve a small (∼1 mmHg) increase in blood pressure and subsequent baroreceptor loading (45). Data were analyzed immediately following the start of phenylephrine infusion and data were analyzed continuously from that point for ∼4 min. Following data collection, phenylephrine infusion was stopped and the MSNA signal was allowed to return to prephenylephrine infusion levels, ensuring that any fall in MSNA was not the result of signal decay.

Protocol 3: Effect of Peripheral Vasodilation on the Arterial Baroreflex

Participants (n = 13, 13 M/0 F) were admitted to the Clinical Research and Trials Unit at the Mayo Clinic and the study day began with instrumentation at 0800. Participants were semirecumbent and were instrumented with a 3-lead electrocardiogram to measure heart rate (Lead II, GE Datex-Ohmeda Cardiocap/5, GE Healthcare, Chicago, IL). An arterial catheter was placed in the nondominant arm for blood pressure monitoring (20-gauge, 5 cm using aseptic technique under local anesthesia, 2% lidocaine) and an intravenous catheter was placed in the dominant arm for systemic intravenous infusion of sodium nitroprusside (nitric oxide donor). Resting MSNA was recorded using the technique of microneurography, as outlined Protocol 1: Effect of Hyperinsulinemia on the Arterial Baroreflex. Sodium nitroprusside was infused intravenously after a quiet rest. All individuals started at a dose of 0.8 µg/kg/min, which was increased by 0.1 µg/kg/min every 2–3 min until a ∼1 mmHg fall in blood pressure and a reflex increase in MSNA was observed (average dose = 0.9 ± 0.2 µg/kg/min).

Analysis

Data were collected using PowerLab data acquisition system (analog to digital converter; ADInstruments, Inc.) with a sampling rate of 10,000 Hz for MSNA and 1,000 Hz for all other variables and stored for offline analysis. Data are reported as an average across an approximate 4-min quiet resting period before (baseline) and during insulin infusion (hyperinsulinemia). Stroke volume was estimated from the arterial blood pressure waveform using the Modelflow method through LabChart (LabChart, ADInstruments), which incorporates age and sex. Cardiac output and total peripheral resistance (TPR) were calculated.

Sympathetic neurograms were analyzed by a single investigator (N.J.M.) using a semiautomated program (Ensemble-R, Elucimed LTD) and MSNA was expressed as burst frequency (bursts/min) and burst incidence (bursts/100 heartbeats). For measures of arterial baroreflex sensitivity, diastolic blood pressures were assigned 3-mmHg bins and for each bin, the corresponding MSNA burst incidence (bursts/100 heartbeats) was determined (46). Arterial baroreflex sensitivity was quantified by plotting MSNA burst incidence (bursts/100 heartbeats) against mean diastolic pressure for each mmHg bin. Each data point was weighted according to the number of times the particular value occurred, based on the technique described by Kienbaum et al (46) and recently published guidelines (47). The value of the slope determined via linear regression analysis provided the arterial baroreflex sensitivity for each participant. The midpoint (T50) for each individual’s curve was calculated, representing the diastolic pressure (mmHg) at which there was a 50% likelihood of an MSNA burst occurring.

Cardiac baroreflex sensitivity was quantified from systolic blood pressure and R-R interval signals, which were required to rise or fall monotonically in the same direction for at least three consecutive beats (48, 49). Values belonging to the identified sequences were formed into xy-pairs and a regression curve was fitted using an automated program (Ensemble-R, Elucimed LTD), with the slope of the curve equaling cardiac baroreflex sensitivity (ms/mmHg). Analysis was performed separately for ascending (up-up) and descending (down-down) sequences, and results were pooled. Those individuals where no sequences could be identified were excluded from the analysis.

Statistical analyses were conducted using GraphPad Prism 9 for Windows (GraphPad Software, LLC). Data from protocols 1 and 3 were analyzed using Student’s paired t test. Data from protocol 2 were analyzed using a one-way repeated-measures analysis of variance (ANOVA) to determine the main effect of condition on main outcome variables. When significant, pairwise comparisons were done using the Holm–Sidak method. Normality was assessed using the Shapiro–Wilk test. Individual responses and means ± standard error of the mean (SEM) are presented. An α of P ≤ 0.05 was considered statistically significant. Based on data from Young et al. (14) showing a 2.30 ± 0.62 bursts/100 heartbeats/mmHg increase in arterial baroreflex gain during hyperinsulinemia in healthy young adults, we determined that we would need paired data from seven individuals to detect a significant effect with the power of 0.80 and α of 0.05.

RESULTS

Protocol 1: Effect of Hyperinsulinemia on the Arterial Baroreflex

Thirteen healthy men (n = 10) and women (n = 3) completed protocol 1. Participants were young (27 ± 2 yr of age), nonobese (height: 181 ± 4 cm; weight: 80 ± 4 kg; body surface area: 2.0 ± 0.1 m²; BMI: 24 ± 1 kg/m²) with normal blood pressures (systolic blood pressure: 114 ± 2 mmHg; diastolic blood pressure: 70 ± 3 mmHg). Plasma insulin and blood glucose concentrations during high-dose systemic infusion of insulin demonstrate successful achievement of hyperinsulinemia (6 ± 1 to 67 ± 4 μIU/mL, P < 0.01) while maintaining blood glucose near baseline levels (77 ± 2 to 72 ± 3 mg/dL, P = 0.10). Resting heart rate (59 ± 2 to 61 ± 2 beats/min, P = 0.13) and mean arterial blood pressure remained unchanged (P = 0.24, Fig. 1A). Data from ModelFlow showed cardiac output (P = 0.17) and TPR (P = 0.63) were preserved during insulin infusion (Fig. 1, B and C). MSNA increased from baseline during hyperinsulinemia when reported as burst incidence (31 ± 3 to 41 ± 3 bursts/100 heartbeats; P < 0.01) and burst frequency (bursts/min, P < 0.01; Fig. 1D). Arterial (P = 0.53; Fig. 1E) and cardiac (n = 10; 24 ± 4 to 27 ± 5 beats/min/mmHg, P = 0.47) baroreflex sensitivity were unaffected by acute, high-dose insulin. Notably, T50 was increased from baseline during acute hyperinsulinemia (P < 0.01; Fig. 1F).

Figure 1.

Effect of high-dose intravenous insulin infusion on the arterial baroreflex (Protocol 1). Individual data, as well as means ± standard error of the mean from n = 13 human participants (10 M/3 F) are reported for MAP (A), cardiac output (B), TPR (C), MSNA (D), baroreflex sensitivity (E), and T50 (F). Solid lines (male participants), dashed lines (female participants). Representative original traces are reported from a male participant (28 yr old) (G). Data were analyzed using Student’s paired t test (Normality: Shapiro–Wilk test). *P < 0.05 vs. baseline. MAP, mean arterial pressure; MSNA, muscle sympathetic nerve activity; TPR, total peripheral resistance.

Plasma insulin and blood glucose concentrations at 60 min of low-dose systemic infusion of insulin demonstrate successful achievement of hyperinsulinemia (6 ± 1 to 44 ± 4 μIU/mL, P < 0.01) while maintaining blood glucose near baseline levels (77 ± 2 to 72 ± 4 mg/dL, P = 0.36). Resting heart rate tended to increase (59 ± 2 to 63 ± 2 beats/min, P = 0.065) and mean arterial blood pressure remained unchanged from baseline (P = 0.39, Fig. 2A). Data from ModelFlow showed an increase in cardiac output (P < 0.01) and reduction in TPR (P = 0.02) during low-dose insulin infusion (Fig. 2, B and C). MSNA (n = 10, 7 M/3 F) tended to increase from baseline during hyperinsluinemia when reported as burst frequency (bursts/min, P = 0.058; Fig. 2D); the increase in MSNA burst incidence did not reach statistical significance (30 ± 3 to 35 ± 4 bursts/100 heartbeats; P = 0.12). Arterial baroreflex sensitivity (P = 0.42, Fig. 2E), T50 (P = 0.37, Fig. 2F), and cardiac baroreflex sensitivity (n = 10; 25 ± 3 to 22 ± 3 beats/min/mmHg, P = 0.32) were unaffected by 60 min of low-dose systemic infusion of insulin.

Figure 2.

Effect of low-dose intravenous insulin infusion on the arterial baroreflex (Protocol 1). Individual data, as well as means ± standard error of the mean from n = 13 (10 M/3 F) are reported, unless otherwise noted (MSNA, Baroreflex gain, T50: n = 10). Solid lines (male participants), dashed lines (female participants). Representative original traces are reported from a male participant (28 yr old). Data were analyzed using Student’s paired t test (Normality: Shapiro–Wilk test). *P < 0.05 vs. baseline. #P = 0.058 vs. baseline. MAP, mean arterial pressure; MSNA, muscle sympathetic nerve activity; TPR, total peripheral resistance.

Protocol 2: Effect of Peripheral Vasoconstriction on the Arterial Baroreflex during Hyperinsulinemia

Seven healthy men (n = 5) and women (n = 2) completed protocol 2. Participants were young (29 ± 3 yr of age), nonobese (height: 179 ± 5 cm; weight: 85 ± 5 kg; body surface area: 2.0 ± 0.1 m²; BMI: 26 ± 1 kg/m²) with normal blood pressures (systolic blood pressure: 121 ± 5 mmHg; diastolic blood pressure: 69 ± 4 mmHg). Plasma insulin and blood glucose concentrations during systemic infusion of insulin demonstrate successful achievement of hyperinsulinemia (8 ± 1 to 51 ± 4 μIU/mL, P < 0.01) while maintaining blood glucose at baseline levels (99 ± 2 to 99 ± 2 mg/dL, P = 0.87).

Resting heart rate increased in response to insulin infusion (64 ± 2 to 69 ± 2 beats/min, P = 0.02), whereas mean arterial blood pressure remained unchanged (P = 0.58, Fig. 3A). Data from ModelFlow showed an increase in cardiac output (P < 0.01) and reduction in TPR (P = 0.01) during insulin infusion (Fig. 3, B and C). MSNA increased from baseline during hyperinsulinemia when reported as burst incidence (40 ± 3 to 45 ± 4 bursts/100 heartbeats; P < 0.01) and burst frequency (bursts/min, P < 0.01; Fig. 3D).

Figure 3.

Effect of peripheral vasoconstriction on the arterial baroreflex during hyperinsulinemia (Protocol 2). Individual data, as well as means ± standard error of the mean from n = 7 human participants (5 M/2 F) are reported for MAP (A), cardiac output (B), TPR (C), MSNA (D), baroreflex sensitivity (E), and T50 (F). Solid lines (male participants), dashed lines (female participants). Representative original traces are reported from a male participant (22 yr old) (G). Data were analyzed using a one-way repeated-measures ANOVA (Normality: Shapiro–Wilk test; Pairwise comparisons: Holm–Sidak method). *P < 0.05 vs. baseline. MAP, mean arterial pressure; MSNA, muscle sympathetic nerve activity; TPR, total peripheral resistance.

Phenylephrine infusion during hyperinsulinemia elicited a reduction in heart rate that was no longer different from baseline (64 ± 2 to 67 ± 2 beats/min, P = 0.12). Data from ModelFlow showed that phenylephrine infusion during hyperinsulinemia resulted in a return to baseline for cardiac output (P = 0.15) and TPR (P = 0.41) (Fig. 3, B and C). MSNA decreased back to baseline levels during phenylephrine infusion when reported as burst incidence (40 ± 3 to 40 ± 4 bursts/100 heartbeats; P = 0.78) and burst frequency (bursts/min, P = 0.58; Fig. 3D). Arterial baroreflex sensitivity (P = 0.92, Fig. 3E), T50 (P = 0.74; Fig. 3F), and cardiac baroreflex sensitivity (n = 4; 17 ± 5 to 13 ± 3 to 14 ± 3 beats/min/mmHg, P = 0.47) were unaffected by hyperinsulinemia and/or phenylephrine infusion.

Protocol 3: Effect of Peripheral Vasodilation on the Arterial Baroreflex

Thirteen healthy men completed protocol 3. Participants were young (29 ± 2 yr of age), nonobese (height: 181 ± 2 cm; weight: 82 ± 3 kg; body surface area: 2.0 ± 0.1 m²; BMI: 25 ± 1 kg/m²) with normal blood pressures (systolic blood pressure: 117 ± 2 mmHg; diastolic blood pressure: 72 ± 2 mmHg). Resting heart rate increased in response to sodium nitroprusside infusion (62 ± 2 to 77 ± 3 beats/min, P < 0.01), whereas mean arterial blood pressure was reduced to 4 mmHg (P < 0.01, Fig. 4A) compared with baseline. Cardiac output tended to increase (P = 0.078; Fig. 4B) while TPR was reduced (P = 0.02; Fig. 4C) from baseline during sodium nitroprusside infusion. MSNA increased from baseline during infusion of sodium nitroprusside when reported as burst incidence (39 ± 2 to 49 ± 3 bursts/100 heartbeats; P < 0.01) and burst frequency (bursts/min, P < 0.01; Fig. 4D). Arterial baroreflex sensitivity and T50 were unaffected by infusion of sodium nitroprusside (P = 0.12 and P = 0.46, respectively; Fig. 4, E and F). Cardiac baroreflex sensitivity (n = 10; 19 ± 2 to 11 ± 2 beats/min/mmHg, P < 0.01) was attenuated during sodium nitroprusside infusion.

Figure 4.

Effect of peripheral vasodilation on the arterial baroreflex (Protocol 3). Individual data, as well as means ± standard error of the mean from n = 13 human participants (13 M/0 F) are reported for MAP (A), cardiac output (B), TPR (C), MSNA (D), baroreflex sensitivity (E), and T50 (F). Representative original traces are reported from a male participant (35 yr old) (G). Data were analyzed using Student’s paired t test (Normality: Shapiro–Wilk test). *P < 0.05 vs. baseline. MAP, mean arterial pressure; MSNA, muscle sympathetic nerve activity; TPR, total peripheral resistance.

DISCUSSION

Insulin-mediated increases in MSNA (8–14) have been primarily attributed to insulin’s effects within the central nervous system, including direct action within the arcuate nucleus (24–29). Increases in insulin within the brain also enhance the gain of the arterial baroreflex control of sympathetic nerve activity in rats (30, 31) and similar increases in baroreflex gain during hyperinsulinemia have been observed in adult men (14). Contrary to prior data, and in disagreement with our primary hypothesis, present findings do not demonstrate that acute hyperinsulinemia increases arterial baroreflex gain. Rather, we observe increases in both MSNA and cardiac output during hyperinsulinemia, which are likely attributed to the baroreflex response to peripheral vasodilation induced by insulin. Indeed, in support of this idea and consistent with our secondary hypotheses, we show that infusion of the peripheral vasoconstrictor phenylephrine abrogates insulin-mediated increases in MSNA and cardiac output, with no effect on baroreflex sensitivity. Similar to insulin, sodium nitroprusside—a potent vasodilator—increased MSNA and tended to increase cardiac output with no effect on arterial baroreflex sensitivity.

Effect of Hyperinsulinemia on the Arterial Baroreflex

In the present investigation, we aimed to assess whether the elevation in MSNA during hyperinsulinemia is mediated by its peripheral vasodilator effect on the arterial baroreflex. We first sought to confirm, based on prior work in humans (14), that acute hyperinsulinemia increases the gain of the arterial baroreflex control of MSNA in healthy young adults. To extend the findings of previous work, we assessed arterial baroreflex control of MSNA with both a high and low dose of systemic insulin, taking advantage of the step-wise hyperinsulinemic-euglycemic infusion protocol introduced by DeFronzo and colleagues (44). Briefly, two priming infusions of insulin were applied to achieve an initial overshoot in plasma insulin concentration which lasted for ∼20 min (44). In this way, we were able to examine the effect of acute, high levels of circulating insulin on baroreflex control of MSNA seemingly independent of insulin’s central effects. Indeed, prior work supports an approximate 30-min delay in the transfer of insulin across the blood-brain barrier (19–21). Using this experimental approach, we report that acute, high-dose insulin induces an increase in MSNA, which is sufficient to maintain TPR and preserve blood pressure. This increase in MSNA is achieved without a change in baroreflex gain and rather we observe a shift in the baroreflex centering point (i.e., T50) to higher blood pressures (Fig. 5A). Present observations of early resetting of the arterial baroreflex working range during hyperinsulinemia in humans are in agreement with early work from Faigus and Berne (33). The authors similarly observed a stable slope of the stimulus-response curve for diastolic blood pressure versus MSNA, but an acute resetting of the baroreflex working range (33) akin to that observed during other physiological stressors such as exercise (50) and/or hypoxia (51). Whether this baroreflex resetting is centrally or peripherally mediated (52) remains to be elucidated. Recent data suggest that brain insulin may not be well reflected by cerebrospinal fluid measurements and physiologically relevant doses of insulin may enter the brain before 30 min (53).

Figure 5.

Impact of acute hyperinsulinemia and peripheral vasodilation on arterial baroreflex control of sympathetic activity. Baroreflex curves were generated with operating and centering (T50) points using group-averaged data from protocol 1 (A: high-dose insulin, B: low-dose insulin), protocol 2 [C: low-dose insulin + phenylephrine (PE)], and protocol 3 [D: sodium nitroprusside (SNP)].

Despite baroreflex resetting with acute high-dose insulin, baroreflex resetting was not observed following a prolonged, lower dose of insulin more consistent with what may be seen following a meal (Fig. 5B). In the absence of baroreflex resetting, we observed a number of systemic hemodynamic changes. Although mean arterial blood pressure was preserved, this was no longer maintained by MSNA alone, thereby resulting in a significant fall in TPR. In this scenario, an increase in cardiac output was necessary to maintain blood pressure, driven by increases in heart rate independent of a change in cardiac baroreflex sensitivity. These observations are consistent with previous work from our group (in a separate cohort of study participants), which found that increases in cardiac output are required to maintain blood pressure during hyperinsulinemia in both men and women (54). We speculate this rise in cardiac output is necessary to preserve blood pressure, in part, because insulin blunts sympathetic vasoconstriction (38).

In the present investigation, we show that the maintenance of blood pressure during hyperinsulinemia is not due to an increase in sensitivity of the arterial baroreflex and/or changes in cardiac baroreflex sensitivity. These results are in contrast to prior data in both animals (30, 31) and humans (14), which show that insulin enhances the gain of the arterial baroreflex control of sympathetic nerve activity. Previous authors attributed these hyperinsulinemia-mediated changes in baroreflex sensitivity in humans to the direct effect of insulin within central neural control pathways (14). Discrepancies between studies may be due to sex of the study participants, dose of insulin, or health of study participants. For example, Young et al. (14) showed that men with the highest insulin sensitivity exhibited the greatest increase in MSNA baroreflex gain during hyperinsulinemia. Thus, in the setting of high insulin sensitivity, it is possible that an increase in baroreflex gain during hyperinsulinemia is essential to maintain blood pressure in the presence of reduced sympathetic vascular transduction (i.e., the transduction of MSNA into vascular tone)—at least in men (55). Notably, although an inverse relationship between arterial baroreflex sensitivity and neurovascular transduction is observed in men, such relationships are not seen in women (55). Nevertheless, all female participants in the current study were tested during the early follicular (low hormone) phase of the menstrual cycle; thus it is unlikely that differences in circulating estrogen/progesterone have contributed to discrepancies between results. Furthermore, in a subanalysis of the present data, primary conclusions were maintained when female participants were excluded.

Effect of Peripheral Vasoconstriction on the Arterial Baroreflex during Hyperinsulinemia

In a second cohort of healthy young men and women, we again administered insulin systemically at a low dose shown previously to elicit peripheral vasodilation and an increase in leg blood flow (54). Congruent with findings from protocol 1, we observed an increase in MSNA during hyperinsulinemia, which was not sufficient to preserve TPR—thus requiring an increase in cardiac output to preserve mean arterial pressure. This increase in MSNA and maintenance of blood pressure during hyperinsulinemia occurred while arterial baroreflex sensitivity and T50 were maintained at baseline levels (Fig. 5C). These data confirm and extend results from protocol 1 using only a single, steady-state dose of insulin (i.e., in the absence of the two priming infusions). Given a lack of increase in arterial baroreflex gain, we hypothesized that the elevation in MSNA during hyperinsulinemia was mediated by its peripheral vasodilator effect on the arterial baroreflex. We reasoned that rescuing peripheral resistance with coinfusion of the vasoconstrictor phenylephrine during hyperinsulinemia would attenuate MSNA and cardiac output responses. Consistent with this hypothesis, we demonstrate that coinfusion with phenylephrine returned MSNA, cardiac output, and TPR to baseline levels, with no effect on arterial or cardiac baroreflex sensitivity (Fig. 5C).

A role of insulin-stimulated vasodilation and response of the arterial baroreflex in insulin-mediated increases in MSNA was discounted previously on the basis that in healthy adults, blood pressure is unchanged during systemic insulin (8). In addition, increases in MSNA during hyperinsulinemia are observed before increases in leg blood flow (36). Based on these data, the authors concluded that insulin stimulates MSNA more rapidly than vasodilation, and thus, the ability of insulin to stimulate MSNA is dissociated from its acute hemodynamic action (36). Important to these conclusions is the assumption that the peripheral vasculature retains its sensitivity to sympathetic activation during insulin exposure; however, we have recently shown that vasoconstriction in response to sympathetic activation is attenuated during hyperinsulinemia (38). In the context of present findings, we propose the peripheral vasodilator effect of insulin contributes, at least in part, to baroreflex-mediated increases in MSNA and the consequent maintenance of blood pressure. Consistent with this, a hyperinsulinemic-euglycemic infusion increased lumbar sympathetic activity in male rats and this increase was attenuated following barodenervation (sinoaortic denervation) (34). Collectively these findings support the notion that a peripheral mechanism contributes, at least in part, to increases in MSNA during hyperinsulinemia.

Effect of Peripheral Vasodilation on the Arterial Baroreflex

In the third cohort of healthy individuals, we sought to recapitulate insulin-mediated increases in MSNA with another vasodilator (sodium nitroprusside). Consistent with our hypothesis, and similar to insulin, sodium nitroprusside increased MSNA and decreased TPR, with no effect on arterial baroreflex sensitivity and/or T50 (Fig. 5D). The similarity of findings between sodium nitroprusside and hyperinsulinemia lends further support to the idea that the elevation in MSNA (and cardiac output) during hyperinsulinemia could be mediated by its peripheral vasodilator effect on the arterial baroreflex. In relation to this, Anderson et al. (8) have argued against a peripheral baroreflex mechanism in insulin-mediated increases in MSNA. The authors ground their argument based on previous data showing that insulin-mediated increase in MSNA per unit reduction in blood pressure was greater (46%) than what others had observed with another peripheral vasodilator (i.e., sodium nitroprusside, 22%) (56). Nevertheless, a greater relative increase in MSNA per unit blood pressure during hyperinsulinemia could also be explained by the ability of insulin to blunt norepinephrine-mediated vasoconstriction (38, 54, 57), given that sodium nitroprusside does not exhibit the same sympatholytic effect (58). Collectively, these findings further support a role for peripheral vasodilation in insulin-mediated increases in MSNA. Of note, cardiac baroreflex sensitivity was reduced with nitroprusside infusion; however, comparisons between the heart rate responses with insulin and sodium nitroprusside should be interpreted with caution due to the independent effect of nitric oxide donors on the sinoatrial node (59, 60).

Perspectives

The MSNA response to acute hyperinsulinemia is attenuated in obese, insulin-resistant adults compared with normal-weight controls (16, 61–64). Furthermore, it is well-established that adults with insulin resistance exhibit impaired insulin-stimulated peripheral vasodilation (6, 7, 65, 66). Based on present findings, one may speculate that obese insulin-resistant individuals do not exhibit further elevation in MSNA during hyperinsulinemia due, in part, to impairments in the peripheral vascular response to hyperinsulinemia (i.e., reduced insulin-induced vasodilation and/or augmented α-adrenergic-mediated constriction) (16, 62, 67) and thus a blunted peripheral vasodilator effect on the arterial baroreflex. Although not tested directly, Straznicky et al, (64) observed no significant associations between changes in calf blood flow following glucose ingestion and changes in sympathetic nervous system activity in insulin-resistant adults. Given that insulin blunts sympathetically mediated vasoconstriction (38), and this may be lost with insulin resistance (54), future studies directly testing these questions in a cohort of insulin-resistant individuals are warranted.

Conclusions

Present findings reveal that acute hyperinsulinemia increases MSNA without a change in the sensitivity of the arterial baroreflex in healthy men and women. This insulin-mediated increase in MSNA resembles that observed with intravenous infusion of another peripheral vasodilator (i.e., sodium nitroprusside) and is lost when peripheral vasodilation is counteracted with infusion of phenylephrine. Together, our results show peripheral vasodilation induced by insulin contributes to increased MSNA during hyperinsulinemia.

GRANTS

This work was supported in part by the Margaret W. Mangel Faculty Research Catalyst Fund (to J.K.L. and J.P.) and the National Heart, Lung, and Blood Institute Grants R00 HL130339 (to J.K.L.), R01 HL153523 (to J.K.L.), R01 HL142770 (to C.M.-A.), and R01 HL137769 (to J.P.).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

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