Department of Medicine, University of Innsbruck, Anichstrasse 35, A-6020 Innsbruck, Austria. E-mail: email@example.com
Objective: In obesity, plasma leptin is high and soluble leptin receptor (sOb-R) levels are low, resulting in a low fraction of bound leptin. The aim of this study was to investigate the influence of insulin resistance (IR) and the metabolic syndrome (MS) on sOb-R concentration and the bound-free ratio of leptin.
Research Methods and Procedures: sOb-R, leptin levels, and homeostasis model assessment (HOMA) index for IR were determined in 76 middle-aged obese or overweight men.
Results: Concentration of sOb-R and soluble receptor-bound fraction of leptin were lowest in the highest tertile of HOMA-IR. sOb-R and the bound-free ratio of leptin correlated with HOMA-IR, leptin concentration, and waist-to-hip ratio independently of age, BMI, and fat mass. Leptin and waist-to-hip ratio were the sole independent determinants of sOb-R concentration, and BMI, HOMA-IR, and visceral adipose tissue were independent determinants of the bound fractin of leptin. sOb-R concentration and the bound fraction of leptin decreased with increasing numbers of components of the MS, resulting in lower sOb-R concentration and a lower fraction of bound leptin in men with the MS.
Discussion: IR and abdominal obesity are associated with low sOb-R concentration and low bound-free ratio of leptin independent of fat mass. Low sOb-R concentration and low bound-free ratio of leptin segregate with components of the MS. We suggest that low sOb-R levels and a low fraction of specifically bound leptin are markers of leptin resistance, which is independently associated with IR and abdominal obesity and may constitute an additional component of the MS.
Obesity is considered to be a major contributor to overall and cardiovascular morbidity and mortality (1). Epidemiological studies demonstrate that the incidence and prevalence of obesity are increasing. The metabolic syndrome, comprising insulin resistance (IR),1 visceral obesity, hypertension, dyslipidemia, and microalbuminuria, is considered a major risk factor for atherosclerosis in obesity (2, 3). The adipocyte-derived cytokine leptin acts as a satiety factor and is significantly raised in obese subjects (4, 5). Recent studies indicate that leptin is an independent risk factor for coronary artery disease (6). Although leptin is highly correlated with the amount of adipose tissue, leptin is correlated with IR independently of fat mass, suggesting that hyperleptinemia is an independent component of the metabolic syndrome (7, 8). The exact interactions between insulin and leptin are not fully elucidated. However, in analogy to IR, a putative leptin resistance in obesity has been postulated (9). Studies derived from mouse models demonstrate a resistance to leptin in the hypothalamus (10, 11), but only few data suggest a putative peripheral leptin resistance (12, 13). Besides membrane-bound isoforms of the leptin receptor with varying cytoplasmic length, a soluble form of the soluble leptin receptor (sOb-R) could be demonstrated. sOb-R represents the main leptin-binding compound in plasma resulting in a fraction of bound and a fraction of free leptin in plasma (14). In contrast to rodents, where an alternative splicing results in sOb-R (15), no m-RNA coding for sOb-R could be detected in humans. Recent work demonstrated that sOb-R is generated by cleavage of the membrane-bound form of the Ob-R (16). The exact function of the sOb-R is not clear. In obesity, levels of the sOb-R are decreased compared with lean controls, resulting in an increased fraction of free leptin (17). Reduction of body weight through diet or surgical procedures significantly increases the concentration of circulating sOb-R and, thus, increases the fraction of bound leptin (18). Thus, sOb-R might act as a modulating factor of leptin action and plays an important role in leptin resistance. The exact physiological mechanisms regulating sOb-R plasma concentration are not known.
The objective of this study was to elucidate the influence of IR, as the main component of the metabolic syndrome, and other components of the metabolic syndrome on sOb-R levels and the bound-free ratio of leptin. To this end, we investigated the plasma concentration of sOb-R, plasma leptin concentration, and the ratio of specifically bound leptin in 76 obese or overweight middle-aged men in relation to varying degrees of IR and other components of the metabolic syndrome.
Research Methods and Procedures
Seventy-six middle-aged men (40 to 60 years, mean 50.0 ± 4.1 years) with a BMI > 28.1 kg/m2 selected from the Salzburg Atherosclerosis Prevention program in subjects at High Individual Risk (SAPHIR) study were investigated. Subjects with apparent coronary artery disease, type 2 diabetes, or concomitant severe illness were excluded. Diabetes was diagnosed when fasting blood glucose exceeded 126 mg/dL or when hypoglycemic drugs had been prescribed. The study protocol was approved by the local ethics committee, and all participants gave written informed consent.
Venous blood was drawn after an overnight fast, and plasma was obtained by centrifugation at 3000 rpm at 4 °C immediately after blood collection. Samples were either used for measurements immediately or stored frozen at −80 °C. Fasting plasma insulin concentrations were measured using a microparticle enzyme immunoassay on an IMx analyzer (IMx System No. 2A10-20, Abbott Diagnostics, Abbott Parc, IL). Glucose was determined using a commercially available enzymatic kit (Roche Diagnostic, Mannheim, Germany). Plasma leptin concentrations were measured by an enzyme-linked immunosorbent assay according to the manufacturer's instructions (R&D Systems, Wiesbaden, Germany).
sOb-R concentration was measured using an sOb-R enzyme-linked immunosorbent assay (Chemicon International, Temecula, CA). The sensitivity of the assay is 0.4 U leptin receptor/mL, and the range of detection is 2 to 100 U leptin receptor/mL. According to the manufacturer of the assay, 1 U/mL leptin receptor approximates 2 ng native human sOb-R/mL (18). The percentage of bound leptin was calculated as published previously (18).
Insulin sensitivity was estimated by the homeostasis model assessment (HOMA) index for IR (19). Subjects were considered as insulin resistant when HOMA-IR was in the highest quartile of HOMA-IR of all men included in the SAPHIR study (n = 560) at the time of analysis (20).
Definition of the Metabolic Syndrome
According to the proposal of the World Health Organization (WHO), the metabolic syndrome was defined as IR plus at least two of the following components: 1) obesity and/or waist-to-hip ratio >0.9, 2) dyslipidemia [high-density lipoprotein-cholesterol (HDL-C) < 35 mg/dL and/or triglycerides >150 mg/dL], 3) hypertension (systolic blood pressure > 140 mm Hg and/or diastolic blood pressure > 90 mm Hg), and 4) microalbuminuria (urinary albumin excretion rate ≥ 20 μg/min). According to the National Cholesterol Education Program (NCEP), the metabolic syndrome was diagnosed when at least three of the following criteria were present: 1) waist > 102 cm, 2) fasting glucose ≥ 110 mg/dL, 3) HDL-C < 40 mg/dL, 4) triglycerides ≥ 150 mg/dL, and 5) blood pressure ≥ 130/85 mm Hg.
Assessment of Body Composition
Body composition was determined by body impedance analysis (B.I.A. 2000-M, Data-Input, Frankfurt, Germany). Areas of abdominal subcutaneous and visceral fat were assessed by computed tomography scan (Picker CT MXTWIN, Picker International, Cleveland, OH) using a single cross section at position L4/5.
Blood pressure was measured using ambulatory 24-hour monitoring (−2430, Boso, Jungingen, Germany). Mean values for systolic and diastolic pressure were calculated from all available measurements (8 am to 8 pm).
Descriptive data are expressed as means ± SD or ± SEM as indicated. Variables not normally distributed were naturally log-transformed. The significance of differences in means was calculated using the unpaired Student's t test, and correlation coefficients were calculated using the Pearson's method. Multiple linear regression was used to detect independent variables. Multiple step-wise linear regression was performed, entering the independent variable with the highest correlation coefficient at each step. A p value < 0.05 was considered statistically significant. All statistical analyses were performed using the SPSS/PC statistical program (version 10.0 for Windows; SPSS, Inc., Chicago, IL).
Clinical characteristics are presented in Table 1. No statistical significant difference in age, BMI, fat or lean mass, waist, waist-to-hip ratio, area of subcutaneous or visceral adipose tissue, and blood pressure could be detected among the three tertiles of insulin sensitivity. According to the definition of the three groups, the HOMA-IR and fasting insulin differed significantly between the groups. In accordance with previous reports, serum leptin levels were highest in subjects with the lowest insulin sensitivity, although no difference in BMI or body composition could be detected.
A strong trend for decreasing sOb-R concentration and bound leptin across the tertiles of HOMA-IR could be detected (p = 0.01 and p < 0.001, respectively). Serum levels of sOb-R were highest in the tertile with the highest insulin sensitivity (20.8 U/mL), intermediate in the second tertile (19.7 U/mL), and lowest in men with IR (17.4 U/mL) (Figure 1 A). Seventy-nine percent (i.e., 1.68 ng/mL free leptin) of leptin was bound to sOb-R in the first tertile, 61% (i.e., 4.77 ng/mL free leptin) in the second tertile, and 46% (i.e., 7.25 ng/mL free leptin) in the third tertile with insulin-resistant subjects (Figure 1B).
Serum concentration of sOb-R was negatively correlated with serum leptin concentration, BMI, fat mass, waist-to-hip ratio, subcutaneous fat area, fasting insulin, and HOMA-IR (Table 2). The ratio of specifically bound leptin was negatively correlated to insulin, HOMA-IR, plasma leptin concentration, BMI, and fat mass and positively correlated to sOb-R levels (Table 2).
After correction for age, BMI, and fat mass, the correlation of sOb-R with HOMA-IR (r = −0.323, p = 0.005), leptin (r = −0.379, p < 0.001), and waist-to-hip ratio (r = −0.460, p < 0.001) persisted. The fraction of specifically bound leptin was negatively correlated with HOMA-IR and waist-to-hip ratio independently of age, BMI, and fat mass (r = −0.469, p < 0.001; r = −0.383, p = 0.002, respectively).
To account for the multiple correlations among the variables, a multiple linear regression was performed to determine independent factors influencing sOb-R levels, showing that only leptin and waist-to-hip ratio are independently correlated to sOb-R concentration (Table 3). Step-wise linear regression demonstrated that leptin [standardized coefficient (SC) = −0.535, p < 0.001] and waist-to-hip ratio (SC = −0.338, p = 0.001) are the only independent determinants, explaining 47.3% of the variance in the sOb-R concentration.
Table 3. Multiple linear regression
Bound leptin (%)
SC, standardized coefficient; VAT, area of visceral adipose tissue; SAT, area of subcutaneous adipose tissue; ND, not determined.
Multiple linear regression using the fraction of bound leptin as a dependent variable demonstrated that HOMA-IR, fat mass, waist-to-hip ratio, and the area of visceral adipose tissue are independently correlated to the bound fraction (Table 3). However, using step-wise linear regression, HOMA-IR (SC = −0.441, p < 0.001), BMI (SC = −0.354, p = 0.001), and the area of visceral adipose tissue (SC = −0.270, p = 0.008) are the sole independent determinants of the fraction of bound leptin, explaining 47.7% of the variance.
The prevalence of components of the metabolic syndrome and, consequently, the prevalence of the metabolic syndrome significantly increased as IR increased (χ2p < 0.001). In the lowest tertile of HOMA-IR, no subject had the stigmata of the metabolic syndrome, and in the median tertile and in the highest tertile, 35% and 72% of the men, respectively, fulfilled the definition of the metabolic syndrome as defined by the WHO criteria (Table 4).
Components of the metabolic syndrome according to the WHO definition: insulin resistance [defined as the highest quartile of HOMA-IR (>2.37) from all men taking part in the SAPHIR study], obesity (BMI > 30 kg/m2 and/or waist-to-hip ratio > 0.9), dyslipidemia (HDL-C < 35 mg/dL and/or TG > 150 mg/dL), hypertension (systolic blood pressure > 140 mm Hg and/or diastolic blood pressure > 90 mm Hg assessed using ambulatory blood pressure monitoring from 8 AM to 8 PM).
Definition of the metabolic syndrome according to the WHO proposal: insulin resistance plus at least two of obesity, dyslipidemia, hypertension, and microalbuminuria.
p < 0.001 for differences in distribution between the tertiles of insulin sensitivity (χ2 test).
As the number of components of the metabolic syndrome increased, sOb-R concentration decreased from 23 U/mL in the subjects without any component of the metabolic syndrome to 18 U/mL in subjects with 4 components (Figure 2 A). As the number of the components of the metabolic syndrome increased, the ratio of bound leptin decreased significantly from 92.4% (i.e., 0.46 ng/mL) to 46.6% (i.e., 8.4 ng/mL) (p = 0.027 for trend, ANOVA) (Figure 2B). Moreover, subjects fulfilling the criteria for the metabolic syndrome according to the proposal of the WHO had significantly lower sOb-R levels and a significantly lower sOb-R bound fraction of leptin (20.08 U/mL vs. 17.98 U/mL, p = 0.03 and 70.43%, i.e., 2.93 ng/mL, free leptin vs. 47.88%, i.e., 7.54 ng/mL, free leptin, p = 0.001, respectively). However, using the definition of the metabolic syndrome of the NCEP, no difference in sOb-R concentration and percent bound leptin could be detected (Figure 3 A and B).
Previous studies have shown that sOb-R concentration is decreased in obesity (21, 22) and that sOb-R increases with weight loss induced by diet or surgical intervention (18). The mechanisms regulating sOb-R concentration are not clear. In vitro studies and studies in mice have suggested that leptin suppresses the expression of its own receptor (10, 23, 24). Recently, Chan et al. demonstrated that leptin administration suppresses sOb-R levels in humans (25). Our results confirm and extend these reports, showing an independent negative correlation of sOb-R with plasma leptin concentration.
However, this is the first study, to our knowledge, to investigate the influence of IR and the metabolic syndrome on sOb-R and the bound-free ratio of leptin. We demonstrate that sOb-R concentration is lowest in subjects who are insulin resistant, although BMI and total fat mass were not higher than in insulin-sensitive men. Besides leptin, waist-to-hip ratio emerged as an independent determinant of sOb-R concentration from a step-wise multiple linear regression model, explaining together half of the variance in sOb-R concentration. This indicates a suppressive effect of abdominal obesity on sOb-R levels.
sOb-R represents the main binding capacity for leptin in the circulation. We can show that in the state of IR, <50% of leptin circulates as bound to the receptor, whereas in obese insulin-sensitive subjects, 80% of leptin is specifically bound to sOb-R, although no difference in measures of fat mass could be detected, resulting in a significantly higher level of free leptin in IR. Furthermore, analyses of our data could demonstrate a strong and independent effect of IR and visceral obesity on the ratio of bound and free leptin, explaining almost 50% of the variation in the bound fraction, again strengthening the important role of visceral obesity on the leptin-sOb-R stoichiometry.
Recently, the metabolic syndrome has gained increasing interest as a clustering of known cardiovascular risk factors, with IR as the central factor (26). We can show that sOb-R levels decrease as the number of components of the metabolic syndrome increase. Furthermore, subjects fulfilling the WHO criteria of the metabolic syndrome have significantly decreased sOb-R levels, although no difference in total fat mass could be detected. As observed for sOb-R, the fraction of bound leptin decreases as components of the metabolic syndrome cluster, resulting in a significantly decreased fraction of bound leptin in subjects with the metabolic syndrome (47%), compared with subjects without the metabolic syndrome (70%). Applying the NCEP definition of the metabolic syndrome, no differences in sOb-R and bound leptin can be detected in this study population. Several reasons could explain these potentially conflicting results: First, when using the NCEP and WHO definitions in our study population, 28 men (36.8%) were not concordantly classified (χ2 = 3.781, p = 0.052, Table 5). Secondly, the WHO defines IR as the mandatory component of the metabolic syndrome, whereas the NCEP definition uses fasting glucose, as an indicator for IR, as one of the five criteria for the definition of the metabolic syndrome. In contrast, we used the HOMA-IR, which is considered to be an accurate estimate of IR as recently shown by Bonora et al., who demonstrated a high correlation between the euglycemic hyperinsulinemic clamp and the HOMA-IR (27).
Table 5. Frequency of the metabolic syndrome according to the definition of the WHO and NCEP, respectively
IR is highly correlated to hyperleptinemia and—as a result of this study—with decreased sOb-R levels and higher free leptin levels. We conclude that decreased sOb-R levels, increased free leptin levels, and a decreased bound-free ratio of leptin are independent characteristics of the metabolic syndrome.
Although leptin signals satiety to the hypothalamus, leptin concentration is paradoxically raised in obesity. In analogy to hyperinsulinemia, hyperleptinemia was interpreted as leptin resistance. Several studies could demonstrate that IR and serum leptin concentration are independently correlated, even after correction for BMI and absolute or relative fat mass, suggesting a pathophysiological interplay between IR and leptin (7, 8). Our observations showing the highest leptin levels in the subjects with the lowest insulin sensitivity and a positive correlation between leptin and HOMA index independently of BMI or fat mass are in accordance with this view. Based on the data from this study, we suggest the further extension of this hypothesis and propose that a decreased sOb-R concentration and a decreased fraction of bound leptin resulting in an increased level of free leptin are additional features of the metabolic syndrome.
The leptin receptor belongs to the class I cytokine receptors. Another member of the class I cytokine receptor family existing in a membrane-bound and soluble form is the human growth hormone (hGH) receptor. By acting as a reservoir for circulating hGH, the soluble receptor compensates for diurnal changes in concentration of hGH and regulates bioavailability of hGH (28, 29). It is reasonable to argue that sOb-R acts in a similar manner. It is unclear whether the bound or the free form of leptin is the biologically active one. Because in the cerebrospinal fluid leptin exists exclusively in its free form, it has been suggested that free leptin is the active form. Based on the observations of this study, we suggest that high concentrations of free leptin are indicative for leptin resistance, with the limitation of the small sample size and gender specific manner of this study.
In conclusion, IR and clustering of components of the metabolic syndrome decrease the concentration of the sOb-R and increase leptin levels in obese or overweight middle aged men, resulting in a decreased fraction of bound leptin, further emphasizing the close relationship between the insulin and the leptin axis. We suggest that low levels of sOb-R and a high concentration of free leptin are independent components of the metabolic syndrome.
This work was supported by a grant of the Medizinische Forschungsgesellschaft Salzburg to B. Paulweber.
Nonstandard abbreviations: IR, insulin resistance; sOb-R, soluble leptin receptor; SAPHIR, Salzburg Atherosclerosis Prevention program in subjects at High Individual Risk; HOMA, homeostasis model assessment; WHO, World Health Organization; NCEP, National Cholesterol Education Program; hGH, human growth hormone; SC, standardized coefficient.