Bound Leptin and Sympathetic Outflow in Nonobese Men

Abstract

Leptin exists in a free form and a receptor-bound form. Protein-bound rather than free leptin levels may be associated with regulation of muscle sympathetic nerve activity (MSNA). We determined MSNA and bound leptin concentrations in 25 men [age, 29 ± 6 yr, body mass index (BMI), 24 ± 3 kg/m²]. Baroreflex sensitivity was measured using phenylephrine and nitroprusside infusions. We measured bound leptin in patients with central (multiple system atrophy, n = 8; age, 59 ± 8 yr; BMI, 23 ± 2 kg/m²) and peripheral autonomic failure (pure autonomic failure, n = 4; age, 71 ± 10 yr; BMI, 25 ± 3 kg/m²). MSNA was correlated with protein-bound leptin concentrations (r² = 0.35; P < 0.01) but not with free leptin levels (r² = 0.09). MSNA at baseline was 15 ± 2 bursts × minutes⁻¹ in subjects with lower and 24 ± 3 bursts × minutes⁻¹ in subjects with higher bound leptin concentrations (P < 0.05). Blood pressure as well as baroreflex regulation of heart rate and MSNA was similar in both groups. Phenylephrine and nitroprusside responses were similar. Patients with multiple system atrophy and autonomic failure featured similar bound leptin levels. We conclude that protein-bound rather than free leptin levels are correlated with basal sympathetic outflow in normotensive, nonobese men. This relationship cannot be explained by a direct central nervous effect of protein-bound leptin. Instead, protein-bound leptin may increase sympathetic vasomotor tone indirectly via a baroreflex mechanism.

LEPTIN DECREASES FOOD intake through an increase in satiety and increases energy expenditure (1). The latter effect is mediated by the sympathetic nervous system and may increase the propensity to develop arterial hypertension (2). However, the metabolic and cardiovascular actions of leptin appear in part to be regulated independently from each other. For example, a subgroup of obese animals appears to be resistant to the antisatiety effect of leptin, whereas the cardiovascular effect of leptin is maintained (3). In Pima Indians, circulating leptin levels and resting energy expenditure increase appropriately with increasing adipose tissue mass (4). Yet, sympathetic vasomotor tone and blood pressure fail to increase with increasing obesity. One possible explanation for the paradoxical dissociation in leptin-associated responses is the existence of different leptins with specific biological properties. Indeed, leptin circulates in the blood in a free and a protein-bound form (5). The major leptin binding protein is the soluble leptin receptor (6, 7). The free and the receptor protein-bound leptin fractions are differentially regulated and may be involved in different regulatory circuits (8, 9). For example, resting energy expenditure appears to be more closely related to bound leptin levels than to free leptin (8). The aim of our study was to elucidate the interaction between protein-bound plasma leptin levels and sympathetic activity in humans. We hypothesized that protein-bound rather than free leptin levels may be associated with blood pressure control and sympathetic vasomotor tone. In animals, sympathetic activation decreases leptin release (10). We, therefore, assessed whether or not bound plasma leptin levels may differ in patients with central or peripheral autonomic failure as modeled by the diseases of multiple system atrophy and pure autonomic failure.

Subjects and Methods

Study subjects

We studied 25 normal male subjects after obtaining their informed written consent [age, 29 ± 6 yr; body mass index (BMI), 24 ± 3 kg/m²; heart rate, 61 ± 2 beats × min⁻¹; blood pressure, 118 ± 2/69 ± 2 mm Hg]. The subjects were included in the study after a review of the medical history and a comprehensive physical examination. The local institutional review board approved the experimental protocol.

Experimental protocol

The subjects abstained from caffeine-containing products and smoking for at least 3 d before testing. They did not ingest any medications and were fasting the night before the investigation. Subjects were studied in the supine position during morning hours. Two iv lines were placed in large antecubital veins, one for blood drawing and the other for drug application. Respiration and heart rate (electrocardiogram) were measured continuously (Cardioscreen, Medis GmbH, Ilmenau, Germany). Beat-by-beat blood pressure (Finapres, Ohmeda, Englewood, CA) and brachial arterial blood pressure (Dinamap, Critikon, Tampa, FL) were determined. Sympathetic activity was assessed directly by microneurography. After a stable baseline was reached, incremental infusions (0.2, 0.4, 0.8, 1.6 μg/kg·min) of sodium nitroprusside (NTP) and phenylephrine (PHE) were given to determine pharmacological baroreflex curves (11, 12).

Microneurography

Muscle sympathetic nerve activity was recorded from the right peroneal nerve. A unipolar tungsten electrode (uninsulated tip diameter, 1–5 μm; shaft diameter, 200 μm) was inserted into the muscle nerve fascicles of the peroneal nerve at the fibular head for multiunit recordings. Nerve activity was amplified with a total gain of 100,000, bandpass filtered (0.7–2 kHz), and integrated (Biomedical Engineering Department, University of Iowa, Iowa City, IA).

Data acquisition and analysis

Data were analog-to-digital converted (500 Hz). Heart rate, diastolic and systolic blood pressure, respiration, and MSNA bursts were defined off-line using a program based on PV-wave software (Visual Numerics Inc., Houston, TX).

Heart rate and blood pressure variability

Heart rate variability and blood pressure variability were calculated in the time and frequency domain (12). Fast Fourier transformation was used for spectral analysis (window, 256 sec; resolution, 0.004 Hz; after subtraction of the mean, trend removal, spline interpolation, and resampling with 4 Hz). High frequency power (0.15–0.4 Hz) and low frequency power (0.04–0.15 Hz) were calculated as mean values over the frequency band.

Free and bound leptin measurements

Free and bound leptin concentrations were measured by specific RIA systems as described previously (13). Plasma samples were taken after 30 min of baseline recording. In addition, we determined protein-bound leptin plasma levels under resting conditions in patients with multiple system atrophy (n = 8) and in patients with pure autonomic failure (n = 4).

Statistics

Group differences were tested with unpaired t test. Deviations from Gaussian distribution were tested using the Kolmogorov-Smirnov test. Nonlinear regression and linear regression analysis were used if indicated. If not otherwise indicated, results are presented as mean values ± sem. Significant differences were considered when P < 0.05.

Results

Regression analysis between leptin concentration and MSNA

We plotted muscle sympathetic nerve activity against free leptin and protein-bound leptin plasma concentrations. The resulting linear regression lines showed significant correlation between protein-bound leptin and MSNA (Fig. 1). MSNA (bursts × minutes⁻¹) was significantly correlated with protein-bound leptin concentration (r² = 0.354; P < 0.0028). Free leptin plasma levels were not correlated with MSNA in normal, nonobese subjects. The same relationship was found for the number of bursts per 100 heart beats and for the normalized area under the burst per 100 heart beats (r² = 0.315, P < 0.0053; and r² = 0.312, P < 0.0056, respectively). Heart rate, arterial blood pressure, and baroreflex sensitivity did not correlate with protein-bound leptin levels.

Sympathetic and parasympathetic baroreflex gain

The pharmacologically defined baroreflex curves were calculated separately for subjects with low protein-bound leptin plasma levels (low-bl, n = 12) and compared with subjects with high protein-bound leptin plasma levels (high-bl, n = 13). The cut-off value was set to 0.5 nmol/liter. Mean values of protein-bound leptin were 0.39 ± 0.03 for the low-bl group vs. 0.85 ± 0.08 nmol/liter for the high-bl group (P < 0.01). Free leptin values were similar between groups (low-bl = 48.8 ± 8.9 vs. high-bl = 46.5 ± 6.9 pmol/liter; not significant). Blood pressure (low-bl = 120 ± 3/70 ± 3 vs. high-bl = 116 ± 4/68 ± 2 mm Hg; P = 0.3475/0.405) and heart rate (low-bl = 59 ± 2 vs. high-bl = 64 ± 2 beats × min⁻¹; P = 0.068) at baseline were similar between groups. However, MSNA (low-bl = 15 ± 2 vs. high-bl = 24 ± 3 bursts × min⁻¹; P = 0.016) was significantly higher in subjects with high, compared with low protein-bound leptin plasma levels. Heart rate variability and blood pressure variability were similar between groups (Table 1). Sympathetic (low-bl = −3.8 ± 0.5 vs. high-bl = −4.0 ± 0.7 a.u./mm Hg) and parasympathetic (low-bl = 17.8 ± 1.8 vs. high-bl = 21.1 ± 3.9 msec/mm Hg) baroreflex gains as resulted from NTP and PHE infusions were similar. PHE and NTP sensitivities were similar in both groups (Fig. 2). No correlation between individual PHE sensitivity and muscle sympathetic nerve activity was found (P = 0.9126).

Leptin in multiple system atrophy and pure autonomic failure patients

To test the influence of autonomic failure, we compared the protein-bound leptin levels between patients with multiple system atrophy and pure autonomic failure. All patients had severe orthostatic hypotension. As expected, the plasma norepinephrine levels were very low in pure autonomic failure patients and were within the normal range in multiple system atrophy patients. However, protein-bound leptin levels were similar in both groups (Table 2).

Discussion

We report the first significant correlation between sympathetic vasomotor tone and protein-bound plasma leptin concentrations in nonobese, normotensive men. Free plasma leptin levels were not correlated with basal sympathetic activity. Recent studies showed that bound leptin concentrations are also correlated with resting energy expenditure, which is strongly influenced by the sympathetic nervous system (8, 9).

The existence of different leptins makes it difficult to compare our data with studies that did not differentiate bound and free leptin concentrations. In one study in healthy subjects, total plasma leptin concentrations were correlated with heart rate and mean arterial pressure, but not with sympathetic nerve traffic over a broad range of body mass indices (14). In nondiabetic men, basal sympathetic activity correlated with percentage body fat and with plasma leptin concentrations (15). Interestingly, 0.3 mg/kg sc leptin for 6 d did not change autonomic activity (16). Some of the controversies in the literature regarding the importance of leptin in the regulation of cardiovascular sympathetic outflow in humans may be explained by the fact that bound leptin concentrations were not considered.

Increased bound leptin concentrations could be a cause or a consequence of sympathetic activation. Indeed, in animals, sympathetic activation profoundly influences leptin secretion (10). We, therefore, compared bound leptin concentrations in multiple system atrophy patients and in pure autonomic failure patients. Both diseases are neurodegenerative conditions. However, pure autonomic failure patients lack sufficient peripheral postganglionic norepinephrine release, which is preserved in patients with multiple system atrophy (17, 18). Bound leptin concentrations were similar in both patient groups. Thus, the correlation between bound leptin concentrations and sympathetic activity is probably not explained by a sympathetically mediated increase in bound leptin release.

We hypothesized that increased sympathetic activity with increasing bound leptin concentrations is related to a direct effect on the central nervous system. Typically, an increase in central sympathetic activity is associated with an increase in heart rate and blood pressure. For example, ingestion of the sympathetic stimulant yohimbine increases sympathetic vasomotor tone, blood pressure, and heart rate (19). Yet, blood pressure tended to be decreased rather than increased in the group with higher bound leptin concentrations.

A centrally mediated increase in sympathetic activity could be masked by a decrease in the sensitivity to the released norepinephrine. We tested the response to the selective α-1 adrenoreceptor agonist PHE. PHE sensitivity was not decreased in the group with higher bound leptin concentrations. Thus, it is unlikely that a centrally mediated effect of bound leptin on the sympathetic nervous system failed to increase blood pressure.

An increase in central sympathetic tone is likely to cause impairment of the baroreceptor heart rate and muscle sympathetic nerve activity reflex (20, 21). A detailed baroreflex analysis did not show any difference between subjects with low and high protein bound leptin plasma levels. Moreover, heart rate variability and blood pressure variability were similar between the groups. Taken together, our findings strongly suggest that the correlation between sympathetic vasomotor tone and bound leptin concentrations in normal weight men is not explained by a direct central nervous effect. We speculate that the correlation results from a compensatory baroreflex mediated increase in sympathetic vasomotor tone. The cardiovascular actions of leptin like natriuresis, insulin sensitization, and endothelium-dependent dilatation may cause lower blood pressure and induce the baroreflex response (2). Therefore, blood pressure is maintained at slightly higher heart rate and MSNA (r² = 0.410; P = 0.0006).

Although protein-bound leptin may not elicit a centrally mediated increase in sympathetic vasomotor tone, previous studies showed a correlation between protein-bound leptin and resting metabolic rate. One possible explanation is that a reflex-mediated increase in sympathetic vasomotor tone also increased energy expenditure. An alternative explanation is that protein-bound leptin may have a selective effect on central sympathetic pathways involved in metabolic regulation (8, 9). Regulatory pathways that selectively influence the distribution of sympathetic outflow from the central nervous system to different organs may be important in the pathogenesis of obesity and hypertension. For example, a mismatch of sympathetic efferent outflow to the kidneys and the heart seems to be involved in the pathogenesis of obesity-associated arterial hypertension (22). Furthermore, in some obese states the satiety and weight-reducing actions of leptin may be selectively disturbed, whereas the cardiovascular action is preserved (3).

For several reasons, we propose that both protein-bound and free leptin should be determined in studies on the interaction of leptin and sympathetic regulation. First, as demonstrated in the present study, both leptins may have different effects on the sympathetic nervous system. Second, free and bound leptin concentrations appear to be regulated independently of one another (1, 7, 8, 23). Finally, free and receptor bound leptin may differ in their ability to enter the central nervous system through the blood-brain barrier. The transport of free leptin into the central nervous system appears to be saturable within a physiological range of leptin concentrations. In contrast, protein-bound leptin transport does not seem to be limited, because cerebrospinal fluid and plasma levels increase in parallel (8).

We conclude that in young nonobese men sympathetic vasomotor tone increases in parallel with increasing concentrations of protein-bound leptin. This relationship cannot be explained by a direct effect on the central nervous system. Instead, bound leptin may increase sympathetic vasomotor tone indirectly via a baroreflex mechanism. In contrast, previous studies are consistent with a centrally mediated effect of bound leptin on energy expenditure. We speculate that a central effect of bound leptin may be important in the distribution of sympathetic traffic to different organ systems. Abnormalities in binding leptin to the soluble receptor might contribute to the paradoxical dissociation of cardiovascular and metabolic leptin effects in different populations of obese subjects.

Acknowledgements

This work was supported in part by Deutsche Forschungsgemeinschaft Grant Jo 284/3-1. J.J. is the recipient of a Helmholtz scholarship of the Max-Delbrueck-Center of Molecular Medicine.

Abbreviations:

 
• BMI,

  Body mass index;

 
• high-bl,

  high protein-bound leptin plasma levels;

 
• low-bl,

  low protein-bound leptin plasma levels;

 
• MSNA,

  muscle sympathetic nerve activity;

 
• NTP,

  nitroprusside;

 
• PHE,

  phenylephrine.

References