| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Human Nutrition (C.M.M., I.T., E.B., S.T., A.A.), Centre of Advanced Food Research, The Royal Veterinary and Agricultural University, DK-1958 Frederiksberg, Denmark; Danish Epidemiology Science Centre (T.I.A.S.), Institute of Preventive Medicine, Copenhagen University Hospital, DK-2100 Copenhagen, Denmark; and Steno Diabetes Centre (O.P.), DK-2820 Gentofte, Denmark
Address all correspondence and requests for reprints to: Associate Professor Inge Tetens, Department of Human Nutrition, The Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark. E-mail: ite{at}kvl.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Bone remodeling is a complex process characterized by an ongoing coordinated resorption and formation of new bone, regulated by systemic hormones and local factors. The fact that bone remodeling occurs independently at multiple skeletal locations have generally been viewed as indirect evidence for a local regulation.
Leptin, the soluble 16-kDa product of the ob gene, is secreted by white adipose tissue, and its circulating levels are correlated with the size of the fat mass (10, 11). Several lines of evidence have suggested that the effects of fat mass on bone mineral density (BMD) may be mediated by hormonal factors, with the prime candidate being serum leptin level. Studies in ob/ob (leptin deficient) and db/db (lack of functional leptin receptor) mice have produced divergent results. Early studies found that systemic leptin treatment of ob/ob mice stimulated bone formation (12) and bone growth (13). Ducy et al. (14) found that both mice mutants had high bone mass because of increased bone formation, and histomorphometric and physiologic analysis demonstrated that leptin was a selective inhibitor of bone formation. Ducy et al. (14) did not discover any leptin receptors among osteoblast cells and intracerebroventricular infusion of leptin in ob/ob, db/db, and wild mice led to a rapid and massive decrease in bone mass, leading to the suggestion that leptin may inhibit bone formation through its binding to leptin receptors in the hypothalamus and to the hypothesis that bone remodeling is under central control.
Recently it has been discovered that human bone marrow stromal cells also express high-affinity receptors for leptin and that leptin induces differentiation of stromal cells toward the osteoblastic lineage (15). In addition, Reseland et al. (16) have shown that the transcription, translation, and secretion of leptin take place from primary human osteoblasts. Moreover, it seems that leptin may stimulate fetal bone metabolism by decreasing bone resorption (17). In growing bone metabolism among pubertal girls, Matkovic et al. (18) found a direct relationship between serum leptin levels and total body bone area.
In mainly cross-sectional studies, several researchers have found a correlation between serum leptin and BMD in humans. Some report that circulating leptin levels are not associated with BMD in humans (19, 20, 21, 22, 23), others report that leptin is positively associated with BMD (24, 25, 26, 27), and some report a negative association between leptin levels and BMD (28, 29, 30). Much of the apparent inconsistency in the published data can be seen as a result of the arbitrary way that different authors have chosen to consider either adjusted or unadjusted data.
Especially the male studies published have been limited to an Asian population, not included obese subjects, and used different anthropometric methods to assess body composition and to a national population-based sample. Because of the crucial role of adipose tissue in the secretion and circulating concentration of leptin and the concomitant effect of body fatness on BMD, we found it pertinent to study this phenomenon in Caucasian men with a wide variation in body fat, also including obese subjects.
This study explores the relationship between fasting serum leptin level and BMD in a group of healthy obese and nonobese men, representative of the Danish male population.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study population consisted of two cohorts of Caucasian men, one juvenile obese cohort and one control cohort, identified from draft board examination records of 1953–1977 for the Eastern part of Denmark. The two cohorts have been described in detail previously (31). The initial anthropometric measurements were performed at the draft board examination, which is mandatory for all Danish men aged 18 yr. For the present study individuals (n = 234) from the juvenile obese cohort, representing all men with juvenile onset obesity (BMI = 31 kg/m2) at the draft board were invited to take part. Another group (n = 323) was invited from the control group cohort, representing a random sample of 0.5% of all draftees. All participants were healthy by self-report.
The Danish Data Surveillance Agency and the Ethical Committee for Copenhagen and Frederiksberg approved the study (VEK-nr 01-389/97). The study was in accordance with the guidelines of the second Helsinki Declaration. All participants signed a written consent before participating in the study.
Assessment of anthropometric estimates
Fat mass (kilograms), lean tissue mass (kilograms), and BMD (grams per square centimeter of the entire skeleton) were evaluated by total-body scanning employing dual-energy x-ray absorptiometry (DXA; LUNAR DXA-IQ DEXA, Madison, WI). All scans were performed in slow mode and analyzed using Lunar smart scan version 4.6c with the slowest scan mode (32). Weight was measured with the participants wearing their underwear and no shoes on an electronic scale (Lindell 8000, Kristianstad, Sweden), with the results given to nearest 0.05 kg. Height was measured with a wall-mounted stadiometer (Hultafors AB, Hultafors, Sweden) and the results given to the nearest 0.5 cm. Abdominal sagittal diameter was measured as the distance between the top of the examination table and a spirit level placed horizontally above the abdomen at the level of iliac crest. The measurements of the hip, waist, and sagittal diameter were repeated three times, and the means were calculated.
Assessment of fasting serum leptin concentration
Blood samples were taken in the morning after a 12-h overnight fast, and centrifuged at 20 C and stored at -80 C before assay. Fasting serum leptin was assessed by RIA (human leptin RIA kit, Alta Diagnostica, Marburg, Germany). The laboratory intraassay precision (coefficient of variation value) for the RIA kit used was 0.037 and the interassay precision (coefficient of variation value) was 0.118. Detection limit was 0.5 ng/ml.
Assessments of physical activity and smoking habits
Assessments of physical activity and smoking habits were estimated from a retrospective questionnaire on smoking habits since 1994 and habitual physical activity levels. Nonoccupational physical activity was coded into two categories: 1) almost no physical activity up to light physical activity 2–4 h/wk and 2) light physical activity more than 4 h/wk or heavier physical activity several times a week (exercise with sweating). Occupational physical activity was coded into two categories and defined as: 1) mostly sitting or standing, sometimes walking and 2) walking or heavy work.
Smoking habits were coded into two categories as no smoking and smoking.
Statistical analysis
Statistical analysis was performed using SPSS for Windows 98 version 11.0 (SPSS Inc. 1989–1999, Chicago, IL). Data are expressed as means and SDs unless otherwise stated. To assess the relationship between variables, simple correlations were calculated. In the correlation analysis, serum leptin was log transformed to obtain normal distribution. Multiple linear regression with BMD as the dependent variable was used to explore the relationship of fasting serum leptin level to BMD. Homogeneity of variance and normal distribution among residuals were checked by plots of residuals and Shapiro-Wilk’s test for normal distribution, respectively. Statistical significance was defined as *, P < 0.05, **, P < 0.01, and ***, P < 0.001.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. As expected, the juvenile obese subjects had higher fasting serum leptin levels, fat mass, body mass index (BMI), and abdominal sagittal diameter than the controls. The age of the control and juvenile obese subjects was (mean ± SD) 49.9 ± 6.0 and 47.5 ± 5.1 yr, and the BMI was 26.1 ± 3.7 and 35.9 ± 5.9 kg/m2, respectively.
|
presents correlations for both groups among fasting serum leptin level, BMD, age, body weight, height, fat mass, lean body mass, abdominal sagittal diameter, physical activity level, and smoking habits. All the variables except serum leptin were normally distributed.
|
Linear regression analyses with BMD as a dependent variable and fasting serum leptin (nanograms per milliliter), fat mass (kilograms), lean tissue mass (kilograms), age, smoking habits, and occupational and nonoccupational physical activity as independent variables, presented in Table 3. In the control group, fat mass (kilograms) (P = 0.025), lean body mass (kilograms) (P < 0.001), and occupational physical activity (P = 0.014) were positively associated with BMD. In the juvenile obese group, lean body mass (kilograms) (P < 0.001) was positively associated with BMD, and smoking (P = 0.019) was negatively associated with BMD. No association was found between BMD and fasting serum leptin in either of the groups. Control and juvenile obese subjects were analyzed together after testing for interactions between the groups (control and juvenile obese). Significant interactions were included in the present model. Lean body mass (kilograms) (P < 0.001), fat mass (kilograms) (P = 0.019), and occupational physical activity (P = 0.011) were positively associated with BMD.
|
and 2 illustrate total BMD (gram per square centimeter) according to quintiles of serum leptin for the control and the juvenile group, respectively.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The strength of the present study is that, because of the sampling design, the juvenile obese group represented the most extremely obese of all draftees, whereas the control group was representative of all draftees. The same analyses were thus performed in two groups of middle-aged men with very different weight histories. However, the study also has some limitations. First, DXA scanning of obese individuals, especially those with a high abdominal sagittal diameter, may result in underestimation of BMD. Second, fasting serum leptin levels and fat mass employed as independent variables in this study are closely related, and the results of the multiple linear regression analysis should therefore be interpreted with caution. Finally, the present study illustrates how adjustment of data (in this case adjusting BMD for body weight) are important for the interpretation of the results. Moreover, the inherent covariation between the variables under examination is a severe shortcoming of the cross-sectional design.
Cross-sectional studies of the relationship between leptin and BMD in men provide data that are not very conclusive. Thomas et al. (25) were the first to publish the hypothesis of an association between serum leptin levels and BMD in men, and they concluded that fat mass and leptin are weakly and inconsistently predictive of BMD in men. However, this study did not include obese subjects.
Sato et al. (29) noted an inverse association between serum leptin levels adjusted for fat mass and BMD in a group of 221 Japanese men. Although Asian men are not comparable with Caucasians with regard to physical stature, bone mass, and leptin levels, our results may be congruent. In a large North American national, population-based sample with oversampling of the elderly, non-Hispanic blacks, and Mexican Americans, BMD increased with increasing leptin concentration in men. However, after adjustment for BMI and other bone-related factors, an inverse association emerged, being most evident in men younger than 60 yr (23).
There are several lines of evidence to suggest a role of leptin in human bone remodeling. People with generalized lipodystrophy, characterized by low serum leptin levels, have been shown to exhibit osteosclerosis (33). Obese individuals appear to have higher plasma-cerebrospinal fluid ratio of leptin than lean control subjects. In addition, obese individuals generally have increased BMD, compared with lean subjects. Apart from other factors, obesity may be related to resistance to the effects of leptin, defective leptin transport into the cerebrospinal fluid, or a combination of both factors. Ducy et al. (14) proposed that leptin resistance may be the mechanism of the protective effect of obesity on bone mass. There are studies to support that with increasing levels of body fatness BMD is not increased correspondingly, such as in obese children, in which BMD was shown to be low for their body weight (34). In humans, circulating leptin concentrations vary from 2 ng/ml to 30 ng/ml (35, 36), and individuals with leptin concentrations above 30 ng/ml can be regarded as resistant to leptin. Higher serum leptin levels may be necessary for the peripheral effects of leptin on bone metabolism. Leptin resistance may explain the missing association between BMD and adjusted fasting serum leptin in the juvenile obese group in the present study.
It is generally accepted that obesity is a major factor protecting against osteoporosis in women. Whether the protective effect is due to the increased weight load or enlarged adipose tissue mass exerting specific effects or to other possibilities mentioned in the introduction, remain an open question. It is conceivable that obesity exerts opposite influences on BMD: a positive mechanical effect of weight load and an independent negative effect of fat mass possibly mediated by leptin. This is supported by the present data on males from the control group in which a direct correlation between BMD and body weight (r = 0.474, P < 0.01) exceeds the correlation between BMD and body fat (kilograms) (r = 0.269; P < 0.01). When adjusted for differences in body weight, BMD was inversely correlated with serum leptin level in both the control group (r = -0.186, P < 0.01) and the juvenile obese group (r = -0.135; P < 0.05) (results not shown). In the linear regression analysis, no relationships were found between BMD and fasting serum leptin levels in either the control or juvenile group (Table 3
). Lean body mass was positively correlated with BMD in both groups, whereas BMD was positively correlated only to fat mass in the control group. Obviously, the interpretation of the present data has the inherent problem that body weight through its covariation with body fat is positively correlated with serum leptin level.
Body weight is predominantly made up of two components: lean body mass and fat body mass, and which of these two components that are related to bone mass is a subject of contention. Edelstein and Barret-Connor (37) found that bone mass in elderly men was related to both lean body mass and fat mass. In contrast, Reid et al. (38) reported that neither lean mass nor fat mass was significantly associated with bone mass in men. This disagreement may be related to the statistical problem of collinearity and to the way in which bone mass is expressed. For example, fat mass was found to be the main determinant when bone mass was expressed as volumetric density (gram per cubic centimeter), whereas lean mass was the independent determinant when bone mass was expressed as areal density (gram per square centimeter) (39). This may explain why lean tissue mass is more strongly related to BMD than fat mass in our study. Different results have been obtained for the association between BMD and fat mass depending on the method of measurement employed (DXA or skin fold measurements) (40). In the study of van Langendonck et al. (40), fat mass derived from the DXA scan was significantly and positively related to bone mass, whereas this was not the case with fat quantified by skin fold measurements. These results contradict those of the present study, in which fat mass derived from DXA was significantly associated with BMD in only the control group. Particular attention has been focused on the risk to male health of abdominal obesity because this kind of fat distribution is linked to several comorbidities in men. The hypothesis that abdominal fat accumulation could be a male risk factor for a decrease in BMD is based on observations that abdominal obesity is related to several metabolic and hormonal disturbances, i.e. hypercortisolemia, reduced androgen and GH levels, and relative hyperestrogenism (41). All these hormonal configurations are generally known to promote the development of male osteopenia and/or osteoporosis (42, 43). In our study abdominal obesity, assessed by abdominal sagittal diameter, was associated only with BMD in the control group studied, and this does not support the hypothesis that abdominal fat accumulation is an independent risk factor for low BMD in men. Rather, the present results indicate that there is a positive relationship between adiposity, expressed as fat mass, and bone mineral status in a healthy male population but no association in a juvenile obese males.
Several studies have suggested that smoking may decrease BMD. Many of these studies have been based on observational data and so are susceptible to confounding by lifestyle and health-associated factors such as poor micronutrient status and body weight. However, there is some evidence that smoking is associated with lower BMD, and our data support these earlier findings only for the control group.
Exercise is often advocated for the prevention of osteoporosis, but there are still many unresolved questions regarding its effect on BMD in men. Indirect evidence tends to indicate a positive association between physical activity and BMD in men. In the present study, occupational physical activity level was positively associated with BMD in the control group and when the groups were analyzed together. It is remarkable that nonoccupational physical activity was not significant when the groups were analyzed separately. This may be explained by the sample method, i.e. self-report by questionnaire, in which it is possible that the answers regarding the variable, nonoccupational physical activity, were susceptible to bias.
Our study does not support the existence of a direct link between adipose tissue and bone metabolism via a common signal cascade, which may be leptin. However, a comparison of a single measurement of a hormone such as leptin with BMD, which is accumulated over a lifetime, cannot reveal possible long-term relationships between bone metabolism and fat mass. Further intervention studies in humans are needed to explain a possible regulatory role of leptin in bone metabolism, the possible difference between locally produced leptin and circulating leptin, and degree to which the circulating leptin levels have a crucial effect on bone metabolism.
Conclusion
Data from the present study support the hypothesis that leptin is inversely associated with BMD and play a significant direct role in the bone metabolism in nonobese and obese Danish males. The present results indicate that there is a positive relationship between adiposity, expressed as fat mass, and bone mineral status in a healthy male population but no association in juvenile obese males, suggesting that increasing fat mass does not protect obese men against osteoporosis. However, the present result also stresses the fact that the strong covariation between the examined variables is a shortcoming of the cross-sectional design.
|
Footnotes |
|---|
Abbreviations: BMD, Bone mineral density; BMI, body mass index; DXA, dual-energy x-ray absorptiometry.
Received March 20, 2003.
Accepted August 21, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. B. Richards, A. M. Valdes, K. Burling, U. C. Perks, and T. D. Spector Serum Adiponectin and Bone Mineral Density in Women J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1517 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Jurimae and T. Jurimae Influence of insulin-like growth factor-1 and leptin on bone mineral content in healthy premenopausal women. Experimental Biology and Medicine, November 1, 2006; 231(10): 1673 - 1677. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Bajoria, S. R. Sooranna, and R. Chatterjee Type 1 Collagen Marker of Bone Turnover, Insulin-Like Growth Factor, and Leptin in Dichorionic Twins with Discordant Birth Weight J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4696 - 4701. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
