| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Articles |
Department of Pediatrics, Division of Molecular Genetics, New York Presbyterian Hospital, New York, New York 10032
Address all correspondence and requests for reprints to: Michael Rosenbaum, M.D., Russ Berrie Research Building, 1150 St. Nicholas Avenue, 6th Floor, New York, New York 10032. E-mail: mr475{at}columbia.edu
| Introduction |
|---|
|
|
|---|
Regardless of how body composition is measured, at the same body weight, women generally have a larger adipose tissue mass (4, 5) and have higher circulating concentrations of leptin per unit of fat mass than men (6, 7, 8). These sexual dimorphisms are predominantly accounted for by gonadal steroids (see below) and indicate that systems regulating energy homeostasis (energy intake and expenditure and the partitioning of stored calories between fat and lean body mass) are also strongly affected by the gonadal hormones.
| Body composition |
|---|
|
|
|---|
A two-compartment model is adequate to describe total energy stores and can be generated by densitometry, isotope dilution, measurement of skinfold thicknesses, body electrical conductivity, or bioimpedance (12, 13, 14). It should be emphasized that none of the above methods provides a direct measure of body composition. Dual energy x-ray absorptiometry (15) provides a more direct measure of body composition as well as information regarding body fat distribution and bone mineral density (1, 16, 17), which, as discussed below, are important measures for assessment of the risk of adiposity-related morbidity and osteoporosis. Precise means of directly measuring body composition are not readily available to the clinical practitioner. Current recommendations are to consider an individual to be at risk for adiposity-related morbidity if his/her BMI is greater than 25 kg/m2 (overweight) and to be obese if his/her BMI is greater than 30 kg/m2 (9). These definitions are meant to alert the practitioner regarding levels of body fatness that may convey increased risk of adiposity-related morbidities rather than to provide rigid diagnostic guidelines.
Genetic predispositions, as reflected by a family history of diabetes, hyperlipidemia, or cardiovascular disease, as well as the relative centrality of body fat distribution (see below), as reflected by a high waist to hip ratio (>0.90 in females, >1.00 in males), also constitute significant risk factors for these moribidities at any given body weight (1). The risk of adiposity-related morbidities (e.g. diabetes, hypertension, or dyslipidemia) is better correlated with visceral adipose tissue mass than with relative body fatness (18, 19, 20, 21). The mechanism for the association between body fat distribution and morbidity is believed to be related to the direct venous drainage of the intraabdominal fat depot into the portal circulation. The high concentrations of free fatty acids thus presented to the liver favor increased synthesis of low density and very low density lipoproteins and promote insulin resistance by interfering with first pass catabolism of insulin by the liver (22, 23). The overall lower rate of cardiovascular morbidity in premenopausal females, despite their greater body fatness (see below), is probably related in part to the proportionally greater amount of fat in sc vs. visceral depots in women vs. men. Visceral adipose tissue constitutes approximately 5% of the total fat mass in premenopausal women vs. 10% in men (24).
Adipose tissue possesses estrogen, androgen, and progesterone receptors (25, 26), and expression of these receptors varies by depot (visceral vs. sc) and gender (27, 28). Androgen receptors are more dense in visceral than sc adipose tissue in both sexes, whereas the estrogen-binding capacity of visceral adipose tissue depots is lower than that of sc adipose tissue in males, but not females (27, 28). Ovine studies have shown higher concentrations of the progesterone receptor in sc (gluteal) than visceral (perirenal or omental) adipose tissue depots (29).
Gonadal steroids largely account for the greater degree of body fatness in women. Genetic males with testicular feminization (insensitivity to androgens) have a female body habitus (30). Women given exogenous androgens or suffering from virilizing tumors or disorders such as congenital adrenal hyperplasia will develop a male body habitus, including more central adipose tissue distribution (27, 31, 32, 33). Cessation of gonadal estrogen production at menopause is associated with an increase in the waist to hip ratio and size of the visceral adipose tissue depot (24, 27, 34, 35), i.e. development of a more android body habitus. Administration of estrogen to postmenopausal women is associated with a lowering of the waist to hip ratio (36). Progesterone administration to rodents increases fat mass (37), but the addition of progesterone to estrogen as a hormone supplement for postmenopausal women does not alter the effects of estrogen administration on body fat distribution (36). In addition, estrogen administration increases the concentration of progesterone receptors in adipose tissue (29). Thus, estrogens and progesterone may act synergistically to favor the storage of excess calories as fat, whereas estrogens promote the storage of fat in more peripheral adipose tissue depots. The virilization of the body composition of women with androgen-producing tumors, even in the presence of normal female circulating concentrations of estrogen, indicates that androgens favor an increase in lean body mass and a loss of fat mass that can, under certain circumstances, mask the effects of estrogen that promote fat storage.
A propensity to store excess calories as fat would represent a clear survival advantage for both sexes via the ability to survive the prolonged periods of low caloric intake that probably plagued our forebears. Storage of extra calories as fat would confer further advantages to females in the form of increased fertility, increased resources for breastfeeding offspring, and earlier menarche (1, 38, 39, 40). Genes that increase this propensity to store calories as fat in the presence of estrogen would distinctly increase the likelihood of conception and survival of offspring. Estrogen and androgen receptors are expressed in neurons of ventromedial and lateral hypothalamic nuclei that affect systems of energy homeostasis (including the arcuate and paraventricular nuclei, which are major sites of synthesis and transport of the orexigenic neuropeptide Y and which affect autonomic nervous system tone and other systems that regulate energy homeostasis) (1, 41, 42, 43, 44, 45, 46).
| Sexual dimorphism in circulating concentrations of leptin |
|---|
|
|
|---|
The pulse amplitude of leptin release from adipose tissue into blood is
2- to 3-fold higher in females than in males, but there is no apparent
sexual dimorphism in pulse frequency (49). Rates of leptin messenger
ribonucleic acid (mRNA) expression in sc adipose tissue are
significantly (
2-fold) higher in females than males (50) and in sc
than visceral adipose tissue in both sexes (51, 52, 53). This sexual
dimorphism in the relationship between circulating leptin concentration
and body fat content could be due to primary effects of gonadal
steroids, genes on the X or Y chromosomes, or the greater proportion of
sc vs. visceral adipose tissue and higher percentage of body
fat in females than in males (4, 5, 53).
Adult females have a significantly greater mass of sc adipose tissue
relative to visceral adipose tissue than males (35, 54, and Rosenbaum
M, A Pietrobelli, J Vasselli, S Heymsfield, R Leibel; submitted
manuscript). Nagy et al. (52) reported that prepubertal
girls had a significantly higher percentage of body fat and a
significantly lower fat-free mass than prepubertal boys, but noted no
significant gender-related differences in total fat mass or in the
distribution of fat in abdominal sc vs. visceral adipose
tissue (measured by computed tomography scan). Circulating
concentrations of leptin corrected for fat mass were also significantly
(
50%) greater in girls. Gender was no longer a significant
determinant of circulating leptin concentrations when corrected for
gender-related differences in body composition and adipose tissue
distribution in visceral vs. sc depots. These investigators
suggested that the sexual dimorphism in the circulating
concentration of leptin in children was due to sex-related differences
in sc vs. visceral adipose tissue volumes, even though no
statistically significant sexual dimorphism in adipose tissue
distribution was found. However, the relatively small amount of
visceral (510% of total adipose tissue mass), relative to sc adipose
tissue in most humans of either sex (35, 54) makes it unlikely that
differential rates of leptin production in adipose tissue from these
depots (50, 53) would be sufficient to account for the striking sexual
dimorphism in circulating concentrations of leptin. Furthermore, even
within specific adipose tissue depots, leptin mRNA expression rates are
approximately 2-fold higher in women than in men (56). Therefore, a
strictly anatomical basis for the sexual dimorphism in circulating
concentrations of leptin is unlikely.
Leptin concentrations in cord blood are significantly lower in male compared to female neonates (57), and circulating concentrations of leptin, normalized to fat mass, are significantly greater in females than in males at Tanner stages I, III, IV, and V in one study (58). The observation of a sexual dimorphism in circulating concentrations of leptin at birth and in prepubertal children in this study (52, 58) suggests that this dimorphism is sex chromosome related. However, numerous other studies have not detected significant gender effects on circulating concentrations of leptin normalized to fat mass (59) or body mass index (60) before late puberty (Tanner stage IV or V), suggesting that the reports that leptin concentration per kg fat mass is significantly greater in prepubertal females than males may represent a type I statistical error due to the nonzero intercept of the regression line relating leptin to fat mass (6, 7) discussed above. In the two articles reporting significantly greater concentrations of leptin corrected for fat mass in prepubertal girls (52, 58), the average prepubertal girl had 3366% greater absolute fat mass than the average prepubertal boy [6.5 kg in girls and 4.7 kg in boys (58); 8.3 kg in girls and 5.0 kg in boys (52)]. In the study not reporting a significant sexual dimorphism in circulating concentrations normalized to fat mass before puberty (60), the absolute fat mass was similar between genders (5.1 kg in girls and 5.5 kg in boys), suggesting that the reported value (52, 58) may be an artifact of a bimodal distribution of body fat in the subject population.
Considering the lack of a consistent demonstration of sexual dimorphism in leptin concentrations relative to fat mass before exposure to endogenous gonadal steroids at puberty, the dimorphism observed in neonates (57) is probably due to the peripubertal levels of circulating androgens in the male fetus (61) and/or the elevated estrogens in the female fetus (62). (Although the authors (57) noted no significant gender differences in testosterone and estradiol concentrations in cord blood, sexual dimorphisms in gonadal steroids are, in fact, present from the time of Leydig cell proliferation until the perinatal gonadotropin surge.) A primary sex chromosome-linked basis (distinct from effects on gonad differentiation) for the sexual dimorphism in the relationship of leptin to fat mass is, therefore, as yet unproven.
In vivo and in vitro studies support a primary endocrine basis for this sexual dimorphism. The sexual dimorphism in the relationship of fat mass to leptin in later puberty is eliminated when adjusted for circulating concentrations of gonadal steroids (63). Plasma leptin concentrations rise during early male puberty, but fall later in puberty, suggesting that androgens may inhibit leptin production (64, 65). Leptin concentrations are strongly negatively correlated with circulating concentrations of testosterone in men. Circulating plasma leptin concentrations, normalized to body mass index, are significantly increased in hypogonadal compared to eugonadal men, and leptin concentrations are lowered in hypogonadal men after the administration of testosterone (66). Incubation of adipose tissue with testosterone decreases leptin mRNA expression, and the circulating testosterone concentration accounts for a significant fraction of the variability in circulating concentrations of leptin in obese boys at all stages of puberty (r = -0.35; P < 0.0001) (67) and in adult men (r = -0.32; P < 0.01) (68).
In contrast to males, leptin concentrations continue to rise throughout puberty in females (65). In vivo administration of estrogen increases circulating concentrations of leptin in humans and rodents (8), and some studies have found that circulating concentrations of leptin, normalized to fat mass, are slightly decreased in postmenopausal (hypoestrogenemic) compared to premenopausal women (6, 8). Ovariectomy in adult rats causes a significant decline in circulating leptin concentrations that is reversed by estradiol supplementation (8). Estradiol increases in vitro leptin production in omental adipose tissue from women, but not men (69), and circulating concentrations of estradiol and testosterone have been reported to be significantly correlated with plasma leptin concentrations in adult women (70, 71), but not in female children or adolescents (72).
Administration of estradiol and cyproterone acetate (an androgen receptor antagonist) to transsexual males increased the circulating plasma leptin concentration (160180%), whereas administration of testosterone to transsexual females increased body weight and decreased the circulating leptin concentration (-5060%) (73). In these studies, hormonal sex, but not genetic sex, was a significant covariate of the plasma concentration of leptin normalized to fat mass. This conclusion is not supported by the observation that this sexual dimorphism is not substantially altered in the hyperandrogenemic polcystic ovarian syndrome (74, 75, 76, 77).
| Leptin, body composition, and energy homeostasis |
|---|
|
|
|---|
The relationship of circulating leptin concentration to energy expenditure in weight-stable humans is unclear. In some studies, significant positive correlations of plasma leptin concentrations with resting energy expenditure or 24-h energy expenditure are found (81, 82, 83), whereas in others no correlations (7, 84) or even significant negative correlations between leptin and energy expenditure are noted (85). Some of the discrepancies among these studies appear to result from the extreme colinearity of leptin and fat mass (r2 > 0.90 in males and females) (6). Using stepwise multiple linear regression analyses, we have examined the relationship between leptin and energy expenditure in lean and obese subjects studied while weight stable at their usual body weights, 10% above their usual body weights, and 10% below their usual body weights (7). We found, as have others (84), that there were no significant correlations between circulating leptin concentration and 24-h energy expenditure or resting energy expenditure at any weight plateau in lean or obese subjects. The maintenance of an elevated body weight did not alter the relationship between leptin and fat mass in any subject, whereas the maintenance of a reduced body weight was associated with a significant reduction in concentrations of leptin normalized to fat mass in women, but not in men (7).
These findings are in contrast to the changes in energy expenditure that accompany maintenance of an elevated or reduced body weight (86). The process of weight loss, but not weight gain, causes a significant decline in circulating leptin concentrations normalized to fat mass (7, 87). These observations are consistent with a threshold model for the effects of leptin on energy homeostasis. When the circulating (hence, central nervous system) leptin concentration falls below an individualized threshold concentration, the hypometabolism and hunger that characterize the weight-reduced individual or the rodent with mutant leptin or leptin receptor genes are invoked (1). In such a threshold model, the organism defends a minimum body fat. The storage of moderately excessive body fat stores does not necessarily invoke any counterregulation. In this sense, the system does not behave as if there were an absolute set-point for body fat above or below which there is a compensatory response. As maintenance of an elevated body weight results in hypophagia and hypermetabolism, an alternative system must regulate upper limits of body fatness. Operationally, such a threshold resolves genetic, developmental, and environmental influences and would not necessarily remain constant throughout a lifetime. The thresholds for the wide variety of leptin-mediated physiological effects (3) are probably different. For example, effects on the gonadal or adrenal axis can be achieved without effects on food intake (79).
If such a threshold model is correct, there should be demonstrable effects of gonadal steroids on systems regulating energy homeostasis, even though men and women demonstrate similar metabolic rates normalized to fat-free mass and make similar adaptations to under- and overfeeding (86). If leptin sensitivity or threshold were not influenced by gonadal steroids, then the hypogonadal males or transsexual females who received testosterone (66, 73) should display similar hyperphagia and hypometabolism as a result of the decline in circulating concentrations of leptin accompanying such treatment. They should, in this instance, attempt to rectify their lower circulating concentrations of leptin by increasing adipose tissue mass. This is clearly not the case, as hyperandrogenized women tend to have a lower percentage of body fat than eugonadal women despite lower circulating concentrations of leptin (73). Therefore, alterations of the gonadal steroid milieu, whether exogenous (e.g. administration of androgens or gonadectomy) or endogenous (e.g. puberty) would be predicted to affect leptin sensitivity, the thresholds for leptin-mediated effects on behavior and metabolism below which an organism will demonstrate stigmata of leptin deficiency. Gonadal steroid effects on leptin sensitivity thus represent a hormonally driven sexual dimorphism in the regulation of systems regulating energy homeostasis and response to weight change. As discussed earlier in regard to the sexual dimorphism of body composition, the presence of androgen and estrogen receptors in neurons comprising hypothalamic nuclei that are involved in systems of energy homeostasis (1, 44, 45, 46) make gonadal steroids, either directly or via effects on gonadotropin release, strong candidate signals for changes in the sensitivity of the central nervous system to leptin-mediated signals.
| Body composition, leptin, and fertility |
|---|
|
|
|---|
Although pubescence and fertility are not synonymous, leptin administration has been shown to hasten the onset of puberty (defined on the basis of vaginal opening size, age at first estrus, ovary weight, ovulatory index, and uterine weight and cross-sectional area) in pair-fed rodents (93). A prepubertal rise in the serum leptin concentration has been proposed as a trigger for the onset of puberty in male humans (64) via signaling of the hypothalamic-pituitary-gonadal axis regarding the nutritional state of the organism. In addition, administration of leptin to rodents increases sexual behavior in fed, but not in food-deprived, female hamsters (94). These observations are consistent with the hypothesis that leptin functions as an afferent signal regarding somatic energy stores that influences reproductive activity and behavior to synchronize endocrine and behavioral components of reproductive function with the sufficiency of energy stores. The influences of adipose tissue mass and estrogen on circulating concentrations of leptin would signal that females were both nutritionally and endocrinologically prepared for reproduction. The apparent decrease in leptin sensitivity in response to increased estrogen might function to encourage increased energy intake during pregnancy, lactation, or periods of maximum fertility in preparation for pregnancy.
| Perspectives |
|---|
|
|
|---|
Genes favoring mobility, strength to compete with other males for mating privileges, and aggressiveness during periods of maximum fertility (adolescence in males and estrus in females) would presumably be selected for in males. Both genders would benefit from genes tending to enhance the ability to store excess calories as fat. Because such a tendency would have enabled our distant progenitors to survive periods of prolonged caloric deficiency, it is likely that the human genome would be heavily enriched with such genes. Survival advantages related to genetic predisposition to storage of calories as fat would be greater in women than in men because women are subjected to the additional energy demands of sustaining gestation and of breastfeeding offspring but are less subject than males to selection pressures favoring increased strength, mobility, and aggressiveness.
Leptin integrates systems of energy homeostasis with those controlling the hypothalamic-pituitary-gonadal axis. There are complex reciprocal interactions of these processes so that leptin affects the integrity of the gonadal axis, and the gonadal steroids affect both leptin production and sensitivity. Pregnancy and lactation are examples of periods in the life cycle when it is desirable to increase energy intake and adipose tissue mass. Perhaps placenta-driven increases in circulating estrogen raise the central nervous system threshold to ambient leptin in the gravid female.
| Footnotes |
|---|
Received February 22, 1999.
Accepted March 16, 1999.
| References |
|---|
|
|
|---|
- and ß-mRNA within
the female rat brain. Brain Res Mol Brain Res. 54:175180.[Medline]
This article has been cited by other articles:
![]() |
S. P Garnett, W. Hogler, B. Blades, L. A Baur, J. Peat, J. Lee, and C. T Cowell Relation between hormones and body composition, including bone, in prepubertal children Am. J. Clinical Nutrition, October 1, 2004; 80(4): 966 - 972. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. R. Kraemer, H. Chu, and V. D. Castracane Leptin and Exercise Experimental Biology and Medicine, October 1, 2002; 227(9): 701 - 708. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-Y. Song, J. Kim, M. Horlick, J. Wang, R. N. Pierson Jr., M. Heo, and D. Gallagher Prepubertal Asians have less limb skeletal muscle J Appl Physiol, June 1, 2002; 92(6): 2285 - 2291. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. He, M. Horlick, J. Thornton, J. Wang, R. N. Pierson Jr., S. Heshka, and D. Gallagher Sex and Race Differences in Fat Distribution among Asian, African-American, and Caucasian Prepubertal Children J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2164 - 2170. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. P Goldstone, A. E Brynes, E L. Thomas, J. D Bell, G. Frost, A. Holland, M. A Ghatei, and S. R Bloom Resting metabolic rate, plasma leptin concentrations, leptin receptor expression, and adipose tissue measured by whole-body magnetic resonance imaging in women with Prader-Willi syndrome Am. J. Clinical Nutrition, March 1, 2002; 75(3): 468 - 475. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. J. O'Sullivan, A. Martin, and M. A. Brown Efficient Fat Storage in Premenopausal Women and in Early Pregnancy: A Role for Estrogen J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4951 - 4956. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. B. Kaplowitz, E. J. Slora, R. C. Wasserman, S. E. Pedlow, and M. E. Herman-Giddens Earlier Onset of Puberty in Girls: Relation to Increased Body Mass Index and Race Pediatrics, August 1, 2001; 108(2): 347 - 353. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Ghizzoni, G. Mastorakos, M. Ziveri, M. Furlini, A. Solazzi, A. Vottero, and S. Bernasconi Interactions of Leptin and Thyrotropin 24-Hour Secretory Profiles in Short Normal Children J. Clin. Endocrinol. Metab., May 1, 2001; 86(5): 2065 - 2072. [Abstract] [Full Text] |
||||
![]() |
M. C. Henson and V. D. Castracane Leptin in Pregnancy Biol Reprod, November 1, 2000; 63(5): 1219 - 1228. [Abstract] [Full Text] |
||||
![]() |
L. B. Williams, R. L. Fawcett, A. S. Waechter, P. Zhang, B. E. Kogon, R. Jones, M. Inman, J. Huse, and R. V. Considine Leptin Production in Adipocytes from Morbidly Obese Subjects: Stimulation by Dexamethasone, Inhibition with Troglitazone, and Influence of Gender J. Clin. Endocrinol. Metab., August 1, 2000; 85(8): 2678 - 2684. [Abstract] [Full Text] |
||||
![]() |
M. B. HORLICK, M. ROSENBAUM, M. NICOLSON, L. S. LEVINE, B. FEDUN, J. WANG, R. N. PIERSON Jr., and R. L. LEIBEL Effect of Puberty on the Relationship between Circulating Leptin and Body Composition J. Clin. Endocrinol. Metab., July 1, 2000; 85(7): 2509 - 2518. [Abstract] [Full Text] |
||||
![]() |
A. M. Isidori, F. Strollo, M. Morè, M. Caprio, A. Aversa, C. Moretti, G. Frajese, G. Riondino, and A. Fabbri Leptin and Aging: Correlation with Endocrine Changes in Male and Female Healthy Adult Populations of Different Body Weights J. Clin. Endocrinol. Metab., May 1, 2000; 85(5): 1954 - 1962. [Abstract] [Full Text] |
||||
![]() |
A. M. Isidori, M. Caprio, F. Strollo, C. Moretti, G. Frajese, A. Isidori, and A. Fabbri Leptin and Androgens in Male Obesity: Evidence for Leptin Contribution to Reduced Androgen Levels J. Clin. Endocrinol. Metab., October 1, 1999; 84(10): 3673 - 3680. [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 |