| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Lilly Research Laboratories (W.F.B., M.L.H., M.B., A.M.A.); University Children’s Hospital (W.F.B., P.E., S.H., W.K., W.R.), Giessen, Germany; Rigshospitalet (A.J., N.T.H., J.M., N.E.S.), Copenhagen, Denmark
Address correspondence and requests for reprints to: Dr. Werner F. Blum, Lilly Germany GmbH, Saalburgstrasse 153, D-61350 Bad Homburg, Germany.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Leptin levels increased in girls with age (r = 0.47, P < 0.0001), but decreased in boys (r = -0.34, P < 0.0001). An analysis according to pubertal stage showed a steady increase in girls between 2.51 µg/L (median) at Tanner stage 1 to 6.24 µg/L at Tanner stage 5. In boys, leptin levels were highest at Tanner stage 2 (2.19 µg/L) and declined thereafter to 0.71 µg/L at Tanner stage 5.
A strong exponential relationship was observed for leptin levels with body mass index (BMI) and percentage body fat as determined by bioelectric impedance measurements in a subgroup of subjects. This relationship was similar between boys and girls at Tanner stages 1 and 2. In boys, there was a significant decline of leptin at a given BMI with further progression of puberty that was much less pronounced in girls. Although the relative increase of leptin with BMI and percent body fat was the same in both genders, the absolute values at a given BMI or percent body fat were significantly lower in boys in late puberty and in adolescents. In boys, but not in girls, there was an inverse correlation with testosterone concentrations (r = -0.43, P < 0.0001), which explained 10.5% of the variation of leptin levels in a multiple regression model. Since BMI proved to be the major influencing variable, reference ranges were constructed using a best-fit regression line of the form leptin = a*e(b*BMI) and stratifying ranges according to gender and pubertal stage.
In conclusion, these data suggest that 1) plasma leptin levels increase in girls and decrease in boys after Tanner stage 2 as the pubertal development proceeds; 2) they show a significant gender difference especially in late puberty and adolescence, even after adjustment for BMI or percent body fat; 3) the lower levels in males may be explained at least in part by a suppressive effect of androgens; 4) reference ranges with BMI as the independent variable should be stratified according to gender and pubertal stage.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All these actions may, at least in part, be explained by the suppressive effect of leptin on neuropeptide Y (NPY) expression and secretion by neurons in the arcuate nucleus (13, 19, 20). Further, there is evidence that leptin also functionally antagonizes NPY action (19, 21). NPY is a strong stimulator of appetite (21, 22) and is known to be involved in the regulation of various pituitary hormones, e.g. suppression of GH through stimulation of somatostatin (23, 24), suppression of gonadotropins (24), or stimulation of the pituitary-adrenal axis (22).
The most important variable that determines circulating leptin levels is body fat mass (25, 26, 27). Obviously, under conditions of regular eating cycles, leptin reflects the proportion of adipose tissue (28). This constitutive synthesis of leptin is modulated by a number of hormones. Strong stimulators in both rodents and humans are insulin (10, 12, 29, 30, 31, 32) and glucocorticoids (12, 33, 34, 35, 36). GH exerts short-term a stimulating and long-term a suppressing effect (37). Suppression has also been shown for cAMP and ß3-adrenoceptor agonists (34). From these findings it becomes clear that leptin is an integral component of various metabolic (21, 22) and endocrine feedback loops.
Because childhood and especially puberty are periods of pronounced changes in the endocrine milieu, we investigated whether various hormonal regulators of growth and sexual development are associated with changes of circulating plasma leptin levels. Because our preliminary findings suggested that leptin levels in normal young children are at the limit of sensitivity of previously employed immunoassays, we developed a new radioimmunoassay that allowed accurate measurement of low plasma leptin concentrations. To provide a basis for the question of what is normal and what is pathologic, leptin was studied in a large cohort of healthy children and adolescents, aiming at the construction of reference ranges with respect to the major influencing variables. Based on reference values, obese subjects may be classified more precisely in those with low leptin levels (relative or absolute leptin deficiency) or with high leptin levels (potential leptin resistance). Thus, leptin measurements may aid to unveil different pathophysiologic causes of obesity.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
713 healthy children and adolescents (312 males, age 6.0–19.9 yr, mean age 13.5 yr; 401 females, age 5.8–19.7 yr, mean age 13.1 yr) were investigated. All children or their parents were asked for written informed consent and agreed to participate in the study. None of the individuals had obvious signs of acute or chronic illness. Standing height was measured with a portable Harpender stadiometer (Holtain Ltd., Crymych, United Kingdom) to the nearest 0.1 cm. Weight was determined using a digital weight scale with a precision of 0.1 kg (SECAdelta, model 707). BMI was calculated as weight (kg) divided by the square of height (m). The pubertal stage (development of breast, genitalia, and pubic hair) was documented in 311 males and 333 females according to the classification by Tanner. For the analysis of subgroups, the classification was based on pubic hair development. Blood samples were drawn from the antecubital vein between 0800 and 1300 h. After centrifugation, serum was stored at -20 C, until measurement of the various parameters.
In 179 subjects (71 males, age 6.4–16.5 yr, mean age 11.3 yr; 108
females, age 6.1–16.8 yr, mean 11.3 yr), the body composition was
determined by bioelectric impedance measurement (BIA) using a Holtain
Body Composition Analyser (Holtain Ltd.). Total body water and fat free
mass were calculated using the equations reported by Davies et
al. (38). For comparison, an alternative equation by Wabitsch
et al. was used (39). The percentage of body fat was
calculated by the following equation:
 |
Methods
Leptin radioimmunoassay Assay components.
New Zealand white rabbits were immunized with recombinant human leptin (Eli Lilly and Company, Indianapolis, IN). The serum with the highest affinity was used for further studies. Tracer was prepared with 2.5 µg leptin as described in detail using the chloramine T method (40). The radiolabeled protein was further purified by exclusion chromatography on a 1.5 x 90 cm column of Sephadex G-50 (Pharmacia, Freiburg, Germany) and was stored at -20 C. Standards were prepared by geometrical dilutions of recombinant human leptin in assay buffer (0.05 mol/L sodium phosphate, pH 7.4, 0.1 mol/L NaCl, 0.1% (vol/vol) gelatine from teleost fish (Sigma, Munich, Germany), 0.1% (vol/vol) triton x 100 (Serva, Heidelberg, Germany), 0.05% (wt/vol) NaN3) between 12.5 and 0.049 µg/L.
Assay procedure. Standards or diluted samples (100 µL; 1:3 to 1:10 in assay buffer) were pipetted into polystyrene tubes, and 100 µL each of first antibody [1:10,000 in assay buffer, containing 150 µg/mL rabbit gammaglobulin (Sigma)] and radioiodinated leptin (150,000–200,000 cpm/mL in assay buffer) were added. After incubation overnight at room temperature, separation of free and bound tracer was achieved by adding 500 µL of cold 4% (wt:vol) polyethylene glycol 6000 (Serva) containing a goat-antirabbit IgG serum (DSL, Webster, MA; 1:150). After 30 min at 4 C, the bound tracer was precipitated by centrifugation (15 min x 2000 g) and the radioactivity was measured in a gamma counter.
Assay performance. Maximal binding in the absence of unlabeled leptin (B0) was 37–45%. Nonspecific binding with normal rabbit serum was 0.61 ± 0.08% (n = 10). Half-maximal binding (50% B/B0) occurred at 0.092 ± 0.007 ng per tube (n = 10). Sensitivity with undiluted samples was 0.03 µg/L, corresponding to 0.003 ng per tube. The intra- and interassay coefficients of variation were 0.8 and 8.5% respectively (n = 10). Excellent parallelism with the standard curve was obtained with serial dilutions of human sera. Spiking experiments with 0.1 ng per tube yielded a recovery of 97 ± 2.1%. Prolonged incubation of the assay mixture (1, 2, 4, and 8 days) at room temperature yielded identical results. Serum leptin levels by RIA were stable upon repeated freezing and thawing, up to 10 cycles, and also during incubation at 4 C and at room temperature up to 8 days. At 37 C leptin levels started to decline steadily after 24 h and had almost disappeared after 8 days.
Other hormones
Insulin-like growth factor (IGF)-I and IGF-II were measured by RIA after acid-ethanol extraction as described previously (41, 42). IGF binding protein (IGFBP)-1 (43), IGFBP-2 (44), and IGFBP-3 (45) were measured by specific RIAs. Estradiol (RIA; Immunodiagnostic System Ltd., Boldon, United Kingdom), testosterone (RIA; Diagnostic Products Corp., Los Angeles, CA), DHEA-S (RIA; Immunotech International, Marseille, France), and SHBG (Delfia; Wallac Oy, Turku, Finland) were measured with commercial assay kits.
Statistics
The distribution of leptin levels in various age or BMI groups was log normal. Therefore, the values were transformed to their logarithms before calculating means and standard deviations, and Student’s t test was used to test for significance of differences between groups. Nonlinear regression analysis was used to calculate ranges that include 90% of the values related to BMI. Correlations between log leptin and the various other parameters are given by the Pearson correlation coefficient r. To test which variables contribute to the variation of leptin levels, multiple step-wise backward regression analysis was performed.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). This
difference in age-dependence became especially clear after logarithmic
transformation of leptin values (Fig. 1b). While in girls levels
increased with age (r = 0.47, P < 0.0001), serum
leptin decreased in boys (r = -0.34, P <
0.0001). As a consequence of this finding, the following analyses were
performed separately for both genders.
|
). In
contrast, leptin levels increased in girls continuously with
progression of puberty, while they decreased significantly in boys from
Tanner stage 3 on despite a further increase in BMI. This opposite
pattern resulted eventually in more than 8-fold higher levels in
females at Tanner stage 5, although BMI values were comparable.
|
). In Tanner stages 3, 4, and 5,
however, average leptin levels decreased at a given BMI with
progression of the pubertal development. This phenomenon was much less
pronounced in girls (Fig. 2b). The comparison of boys and girls,
classified according to pubertal stage, revealed that the relative
increase of serum leptin with BMI is identical in both genders as
indicated by parallel lines in the semilogarithmic plots (Fig. 3). However, while
absolute values showed a less pronounced difference in pre- and early
puberty (Figure 3a), a large difference was observed in Tanner
stage 5. This stratified analysis according to pubertal stages clearly
improved the associations between leptin levels and BMI especially in
males (Table 2).
|
|
|
), indicating that
the relative increase of leptin with body fat is the same in both
sexes. However, for a given percentage of body fat, females had higher
absolute leptin levels than males. At a mean value of 16.26% BF, the
average leptin level was 2.74 µg/L in girls vs. 1.59
µg/L in boys. Using the errors of the nonlinear regression model in
both sexes, this difference proved significant (P =
0.002). Corresponding results were obtained when, alternatively, the
model of Wabitsch et al. (39) was employed for calculating
percent BF. Although the correlation of BF values with those obtained
with the model of Davies et al. was high (r = 0.97),
absolute values were lower by about 30%.
|
).
Significant correlations were found with logarithmically transformed
leptin levels, which were, surprisingly, mostly opposite when comparing
genders. This may be related to the opposite age-dependence of leptin
levels. When a multiple step-wise backward regression analysis was
performed, the total contribution of all parameters, including age and
BMI, accounted for 52.3% of the leptin variation in males. Besides age
and BMI, only testosterone (10.5%) and IGFBP-2 (3.2%) contributed
significantly to this variation. In females, 69.5% of the variation
was explained by all parameters, with BMI as the major influencing
variable and only a minor contribution by age and by IGFBP-2
(5.4%).
|
. These reference ranges can be used to
adjust leptin levels for BMI in a given comparator group by calculating
standard deviation scores (SDS, Z-scores). Accounting for the
logarithmic distribution, the deviation from the mean reference value
is given by the following equation:
 |
The d-values of the various subgroups are also shown in Table 2.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
When leptin levels were related to BMI, an exponential relation was observed in accordance with reports in adults (26, 27) indicating that fat mass is the main regulator of leptin levels. However, there was also a clear dependence upon age and pubertal stage that was also observed in a smaller number of subjects by Hassink et al. (47). In particular, there was a decrease in males with age as compared to females. At first sight, this finding may be explained by the divergent pattern of development of fat and lean body mass during puberty. In males, the increase of BMI in late puberty may be mainly caused by the accumulation of muscle mass owing to the action of androgens. This view is supported by the fact that the difference between male and female leptin levels was much less pronounced when leptin was referred to percent BF instead of BMI. However, although the relative increase of leptin vs. BMI or percent BF as measured by BIA was identical in both genders, i.e. for a given increase in BMI or percent BF, leptin levels increased by the same factor, absolute values were significantly different, suggesting that an additional factor influences plasma leptin.
The question, whether sex steroids are causally involved in this gender difference, which has also been observed by others (47, 48, 49), cannot conclusively and directly be answered by the results of this study. As revealed by multiple regression analysis, however, about 10% of the variation of leptin levels could be contributed to testosterone in males, but not in females. Since absolute values of testosterone were on average about 20-fold higher in male adolescents, it cannot be excluded that a threshold effect plays a role. In a separate study with hypogonadal men receiving androgen replacement (50), and in in vitro studies with human adipocytes (51), we were able to clearly demonstrate that androgens suppress leptin levels and leptin expression. Therefore, the divergent pattern of absolute leptin levels during pubertal development of males and females may be explained by the rising levels of androgens in males that have a direct effect and also an indirect one by shifting the relative contribution of the increase of BMI towards muscle mass.
As suggested by multiple regression analysis, IGF-I as an important regulator of growth does not significantly contribute to the variation of leptin, which is consistent with in vitro experiments (Dr. M. Wabitsch, personal communication). Also, leptin did not significantly contribute to the variation of GH-dependent variables such as IGF-I or IGFBP-3. This was unexpected because, hypothetically, leptin should be expected to stimulate GH secretion through suppression of NPY (23, 24) and since it restores starvation-induced suppression of GH secretion in rats (Dr. M. Aubert, personal communication). Therefore, the question, whether leptin as a signal reflecting available energy stores is of any importance for growth, which would teleologically make sense, remains to be answered by more specifically designed studies.
To appropriately interpret measured leptin concentrations (e.g. to detect pathology), reference values are required. For this purpose, leptin should be referred to the predominant (independent) influencing variable. With respect to leptin levels, the major regulator proved to be fat mass. Absolute measured values of fat mass, however, depend largely on the methodology used, despite the fact that good correlations between methods may be obtained. Using the impedance values of the current study, absolute percent of BF values differed largely depending on the regression models used, although their relation to leptin levels was comparable. On the other hand, BMI reflects body fat mass somehow. It is easy to determine, it is retrospectively mostly available, and it is the most precise measure for short-term longitudinal changes of body fat in a single patient, e.g. during restricted food intake. Therefore, on this basis it appeared reasonable to provide reference ranges of leptin levels referring to BMI rather than relative fat mass or age.
Of course, these ranges (taken as normal ranges), are, in a strict
sense, limited to the BMI ranges of the studied cohorts. Adjusting the
leptin concentrations of patients that fall within this range according
to gender, pubertal stage and BMI is straightforward. However, the
extrapolation of these ranges to extremely low or high BMI values is
not trivial. A BMI value of, say, 40 kg/m or less can be considered
neither normal nor healthy. Therefore, the ranges provided by the
functions derived from this study should be considered instead as
expectation ranges. Nevertheless, these expectation ranges may be a
useful tool for detecting discrepancies between a given BMI value and
aberrant leptin levels. Removal of the influences of gender, pubertal
stage, and BMI by calculating leptin SDS values as suggested in Table 2
can thus be helpful in detecting both pathologic leptin production (or
clearance) and the influence of additional regulating factors.
|
Footnotes |
|---|
Received March 20, 1997.
Revised May 29, 1997.
Accepted March 20, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. A. DIVALL and S. RADOVICK Pubertal Development and Menarche Ann. N.Y. Acad. Sci., June 1, 2008; 1135(1): 19 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. L. Hui, C. M. Schooling, S. S. L. Leung, K. H. Mak, L. M. Ho, T. H. Lam, and G. M. Leung Birth Weight, Infant Growth, and Childhood Body Mass Index: Hong Kong's Children of 1997 Birth Cohort Arch Pediatr Adolesc Med, March 1, 2008; 162(3): 212 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Sloboda, R. Hart, D. A. Doherty, C. E. Pennell, and M. Hickey Age at Menarche: Influences of Prenatal and Postnatal Growth J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 46 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.-E. W. Kavey, V. Allada, S. R. Daniels, L. L. Hayman, B. W. McCrindle, J. W. Newburger, R. S. Parekh, and J. Steinberger Cardiovascular Risk Reduction in High-Risk Pediatric Patients: A Scientific Statement From the American Heart Association Expert Panel on Population and Prevention Science; the Councils on Cardiovascular Disease in the Young, Epidemiology and Prevention, Nutrition, Physical Activity and Metabolism, High Blood Pressure Research, Cardiovascular Nursing, and the Kidney in Heart Disease; and the Interdisciplinary Working Group on Quality of Care and Outcomes Research: Endorsed by the American Academy of Pediatrics Circulation, December 12, 2006; 114(24): 2710 - 2738. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Nobili, M. Manco, P. Ciampalini, V. Diciommo, R. Devito, F. Piemonte, D. Comparcola, R. Guidi, and M. Marcellini Leptin, free leptin index, insulin resistance and liver fibrosis in children with non-alcoholic fatty liver disease. Eur. J. Endocrinol., November 1, 2006; 155(5): 735 - 743. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Tam, F. de Zegher, S. P. Garnett, L. A. Baur, and C. T. Cowell Opposing Influences of Prenatal and Postnatal Growth on the Timing of Menarche J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4369 - 4373. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Spinazzi, P. G. Andreis, G. P. Rossi, and G. G. Nussdorfer Orexins in the regulation of the hypothalamic-pituitary-adrenal axis. Pharmacol. Rev., March 1, 2006; 58(1): 46 - 57. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. K. Volkl, D. Simm, C. Beier, and H. G. Dorr Obesity Among Children and Adolescents With Classic Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency Pediatrics, January 1, 2006; 117(1): e98 - e105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Eikelis and M. Esler The neurobiology of human obesity Exp Physiol, September 1, 2005; 90(5): 673 - 682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gerber, A. Boettner, B. Seidel, A. Lammert, J. Bar, E. Schuster, J. Thiery, W. Kiess, and J. Kratzsch Serum Resistin Levels of Obese and Lean Children and Adolescents: Biochemical Analysis and Clinical Relevance J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4503 - 4509. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Wudy, S. Hagemann, A. Dempfle, G. Ringler, W. F. Blum, L. D. Berthold, G. Alzen, L. Gortner, and J. Hebebrand Children With Idiopathic Short Stature Are Poor Eaters and Have Decreased Body Mass Index Pediatrics, July 1, 2005; 116(1): e52 - e57. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J H Davies, B A J Evans, and J W Gregory Bone mass acquisition in healthy children Arch. Dis. Child., April 1, 2005; 90(4): 373 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. F. Liew, S. D. Wise, K. P. Yeo, and K. O. Lee Insulin-Like Growth Factor Binding Protein-1 Is Independently Affected by Ethnicity, Insulin Sensitivity, and Leptin in Healthy, Glucose-Tolerant Young Men J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1483 - 1488. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Meissner, C. Hanisch, I. Ostreicher, I. Knerr, K.-H. Hofbauer, W. F. Blum, I. Allabauer, W. Rascher, and J. Dotsch Differential Regulation of Leptin Synthesis in Rats during Short-Term Hypoxia and Short-Term Carbon Monoxide Inhalation Endocrinology, January 1, 2005; 146(1): 215 - 220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Plant and M. L. Barker-Gibb Neurobiological mechanisms of puberty in higher primates Hum. Reprod. Update, January 1, 2004; 10(1): 67 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Boonstra, Y. van Pareren, P. Mulder, and A. Hokken-Koelega Puberty in Growth Hormone-Treated Children Born Small for Gestational Age (SGA) J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5753 - 5758. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. GHIZZONI and G. MASTORAKOS Interactions of Leptin, GH, and Cortisol in Normal Children Ann. N.Y. Acad. Sci., November 1, 2003; 997(1): 56 - 63. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. APTER The Role of Leptin in Female Adolescence Ann. N.Y. Acad. Sci., November 1, 2003; 997(1): 64 - 76. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Wilson, J. Fisher, K. Chikazawa, R. Yoda, A. Legendre, D. Mook, and K. G. Gould Leptin Administration Increases Nocturnal Concentrations of Luteinizing Hormone and Growth Hormone in Juvenile Female Rhesus Monkeys J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 4874 - 4883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Mann, A. O. K. Johnson, T. Gimpel, and V. D. Castracane Changes in Circulating Leptin, Leptin Receptor, and Gonadal Hormones from Infancy until Advanced Age in Humans J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3339 - 3345. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Jeon, V. J. Harber, and R. D. Steadward Leptin response to short-term fasting in sympathectomized men: role of the SNS Am J Physiol Endocrinol Metab, March 1, 2003; 284(3): E634 - E640. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Saar, F. Geller, F. Ruschendorf, A. Reis, S. Friedel, N. Schauble, P. Nurnberg, W. Siegfried, H.-P. Goldschmidt, H. Schafer, et al. Genome Scan for Childhood and Adolescent Obesity in German Families |
)
and girls (•) vs. age. a) Linear scale of the
ordinate; b) logarithmic scale of the ordinate. The regression lines in
panel a) are given by leptin = 3.1624*e(-0.0792*age)
in males and leptin = 1.1777*e(0.0963*age) in
females.
