Upload
l-hellstroem
View
212
Download
0
Embed Size (px)
Citation preview
Mechanisms behind gender differences in circulating leptin
levels
L. HELLSTROÈ M1
, H. WAHRENBERG2
, K. HRUSKA2
, S. REYNISDOTTIR2
& P. ARNER2
From the1
Department of Medicine, Danderyd Hospital, Stockholm, and the2
Karolinska Institute, Department of Medicine and Clinical Research Centre at
Huddinge Hospital, Stockholm, Sweden
Abstract. HellstroÈm L, Wahrenberg H, Hruska K,Wahrenberg H, Reynisdottir S, Arner P (Danderyd
Hospital and Karolinska Institute, Stockholm, Swe-
den). Mechanisms behind gender differences incirculating leptin levels. J Intern Med 2000; 247:
457±462.
Objectives. To investigate gender differences in
circulating leptin levels and adipose tissue productionof leptin.
Design setting and subjects. Thirty-two men and 63
women with a large interindividual variation in bodymass index (BMI), but otherwise healthy, were
investigated after an overnight fast. Body fat
(bioimpedance), abdominal subcutaneous adiposetissue secretion of leptin in vitro and serum leptin were
determined.
Results. Although there was no gender difference inmean BMI or fat cell size, mean percentage body fat
was 49 in women and 36 in men (P , 0.001). At
each level of BMI, serum leptin levels were about two
times higher in women than in men (P , 0.001).Adipose tissue secretion rate of leptin in men was two-
thirds of that in women (P , 0.05). The gender
differences in body fat content, serum leptin and leptinsecretion were observed in obese (BMI . 27 kg m22)
as well as non-obese subjects. Serum leptin levels
(P , 0.001) and leptin secretion rate (P , 0.01)correlated positively with body fat content in either
sex. However, the gender differences in serum leptin
(P , 0.001) and leptin secretion rate (P , 0.01)remained statistically significantly different even
when the values were adjusted for body fat.
Conclusion. The gender difference in circulatingleptin concentrations can be due to at least two
different mechanisms. A higher proportion of adipose
tissue and increased production rate of leptin per unitmass of adipose tissue might explain why women have
higher circulating leptin levels than men.
Keywords: adipose tissue, body fat, fat cell, obesity,
plasma, sex.
Introduction
Leptin, secreted from adipose tissue, appears to play
a major role in regulating the body fat mass in
animal models [1±4]. Genetic studies stronglysuggest such a role of leptin in humans as well,
since deletion mutations in leptin or in leptin
receptors associate with morbid obesity [5, 6].A particular feature with leptin in humans is the
marked gender difference in the circulating hor-mone level (see Saad et al. [7] and Licinio et al. [8] for
detailed discussion). Women have much higher
leptin concentrations than men at all levels of body
mass index (BMI). The mechanism behind this sexvariation is not known, although an influence of sex
hormones is considered [9, 10]. It is possible that
gender variations in circulating leptin levels explainwhy most women have a higher body fat content
than do men.
We recently demonstrated that the leptin secre-tion rate from adipose tissue is a major denominator
of the circulating level in women [11]. The rate of
secretion was markedly increased in obese subjects,which corresponded to increased circulating leptin
levels in the obese state. In the present study we
have compared adipose tissue secretion rate of leptin
Journal of Internal Medicine 2000; 247: 457±462
# 2000 Blackwell Science Ltd 457
in vitro in men and women who were either obese or
non-obese in order to elucidate the mechanisms
behind gender variations in circulating leptin.
Materials and methods
Subjects
The study is an ongoing project with the aim ofstudying leptin secretion from human adipose tissue
[11]. All subjects (n = 96) with no overt disease
except uncomplicated obesity were included. Thisstudy group consisted of healthy volunteers with
varying BMI and obese, otherwise healthy subjects
who later underwent weight reduction treatment(conservative or surgical). None of the subjects was
on a diet or had undertaken a weight reduction
programme within 6 months of the investigation.None was completely sedentary and none was
involved in athletic performance. Thirty-two men
aged 23±59 years and 64 women aged 20±64 yearswere included. BMI ranged from 23 to 53 kg m22 in
men and from 21 to 53 kg m22 in women. Obesity
was defined as BMI . 27 kg m22.The subjects came to the laboratory at about
08.00 h after an overnight fast. Height and weight
and body fat were determined and BMI andpercentage body fat were calculated. The bioimpe-
dance method used to determine body fat content
has been described earlier [12]. In the supineposition, waist-to-hip ratio (WHR) was determined
and a blood sample was obtained for measurement
of serum leptin; thereafter, a specimen of subcuta-neous adipose tissue was removed from the middle
part of the abdominal region. During the first period
of the project, adipose tissue was removed by openbiopsy under local anaesthesia as described pre-
viously [13]. Recently, a needle biopsy procedure
has been described and evaluated [14]. Since thelatter method is more convenient for patients, it was
considered unethical to continue with open biopsies.
Therefore all biopsies in the later part of the studywere performed with the needle aspiration technique
(ISO-Caro).The study was approved by the hospital's
committee on ethics and it was explained in detail
to each participant and his or her consent obtained.
Adipose tissue experiments
Adipose tissue was cut into fragments (about 10±20 mg each). One part was used for isolation of fat
cells and determination of fat cell volume exactly as
described previously [15]. Another part of theadipose tissue fragments was incubated in vitro
exactly as described previously [11]. In brief,
approximately 300 mg of tissue was incubated in3 mL of an albumin-containing buffer for 2 h at
378C. The ingredients in the buffer were from the
same batches throughout the study. An aliquot ofthe medium was freeze-dried and stored at ±70 8C
for subsequent leptin analysis. The incubated
adipose tissue was collected, lipids were extractedand the measured leptin secretion was related to the
whole incubation volume and lipid weight of the
incubated adipose sample as described earlier [11].
Determination of leptin protein
Serum (stored at ±70 8C) was thawed and freeze-
dried incubation medium was redissolved in distilled
water. Leptin in serum or incubation medium wasdetermined with a radioactive immune assay kit
(Linco Research, St Charles, USA)
Statistics
All values are mean 6 SE (standard error of mean).
The data were analysed statistically using linear
regression, analysis of covariance and Student'sunpaired two-tailed t-test. The values for adipose
tissue leptin secretion rate were not normally
distributed, and were transformed to the logarithmicform before analysis.
Results
The relationship between serum leptin and BMI is
depicted in Fig. 1. A significant (P , 0.001) rela-tionship between the two parameters was found in
both women (r = 0.82) and men (r = 0.87). There
was, however, a gender effect on the relationship(F = 11.8, P , 0.001 by analysis of covariance). At
each level of BMI, higher serum leptin values wererecorded in women. A significant relationship
between adipose tissue secretion rate of leptin and
serum leptin levels was also found in both men
4 5 8 L . H E L L S T R OÈ M E T A L .
# 2000 Blackwell Science Ltd Journal of Internal Medicine 247: 457±462
(r = 0.57, P , 0.001) and women (r = 0.52,P , 0.0001).
The clinical data are shown in Tables 1±3.
Women had significantly higher mean values ofpercentage body fat and significantly lower WHR
than men. However, the means for fat cell volume,
BMI and age were comparable in the two groups.Women had 2±3 times higher leptin concentrations
than men. Similar results were recorded in lean and
obese subjects as regards the gender effects.Figure 2 depicts the influence of gender leptin
secretion. The adipose tissue secretion rate inwomen was increased by one-third as compared
with that of men (P , 0.05).
Clinical and adipose secretion data were alsoanalysed after the cohort was separated into lean
and obese subjects (Tables 1±3, Fig. 2). The gender
differences observed in the whole cohort were alsopresent in non-obese or obese subjects.
The influence of body fat content (kg) has also
been taken into account. Total body fat correlated
positively with serum leptin (r = 0.83, P , 0.001
for women, and r = 0.89, P , 0.001 for men) andleptin secretion rate (r = 0.50, P , 0.01 for wo-
men, and r = 0.80, P , 0.001 for men). We also
analysed the gender influence of total body fat onleptin using analysis of covariance. This analysis
showed that gender interacted with the relationships
between total body fat and serum leptin (F = 12.8,P , 0.001) and between total body fat and leptin
secretion (F = 6.9, P = 0.01). However, the gender
differences in serum leptin (P , 0.001) and leptinsecretion rate (P , 0.05) remained when the
influence of body fat content was taken into
account.
Discussion
To the best our knowledge, we have, in the present
study, for the first time demonstrated some mechan-isms which can explain why circulating leptin levels
are higher in women than in men. This was done by
comparing adipose tissue leptin secretion rates invitro in a large population of apparently healthy
men and women (except for obesity).
We could confirm that plasma leptin is related toBMI and adipose secretion rate of leptin. In
agreement with numerous pervious findings [7, 8],
we observed higher serum leptin levels in women. Ateach level of BMI, plasma leptin concentrations were
2±3 times higher in women than in men.At least two factors ± the production rate in
adipose tissue and the fat mass ± might explain the
gender differences in circulating level of leptin. Theincreased rate of leptin secretion in women was
observed in obese as well as lean subjects, and it was
L E P T I N I N M E N A N D W O M E N 4 5 9
120
020
BMI (kg m–2)
Lept
in (
ng m
L–1)
5030 40
100
80
60
40
20
60
Female
Male
Fig. 1 Bivariate plot of serum leptin versus body mass index (BMI)
in women (open symbols) and men (filled symbols).
Table 1 Clinical data on the whole study group
Measure
Females
(n = 63)
Males
(n = 32) P
Age (years) 40 � 1 39 � 2 0.84
BMI (kg m22) 36.7 � 1.2 35.1 � 1.6 0.45WHR 0.893 � 0.011 0.993 � 0.011 <0.0001
Percentage body fat 52 � 2 32 � 2 <0.001
Fat cell volume (pL) 764 � 32 776 � 39 0.81
Serum leptin (mg L21) 38 � 3 17 � 2 <0.001
Values are mean � SE. They were compared using Student's two-tailed unpaired t-test.
BMI, body mass index; WHR, waist-to-hip ratio; n = number of subjects.
# 2000 Blackwell Science Ltd Journal of Internal Medicine 247: 457±462
not due to a gender difference in adipose tissuecellularity, because mean fat cell size was similar in
men and women.
The relative amount of adipose tissue was muchhigher in women, regardless of BMI, than it was in
men. The data on body fat content indicate that one
important reason behind gender variation in plasmaleptin concentration is an adipose tissue mass effect.
This mass effect, in combination with an increased
rate of leptin secretion per unit mass in women,seems to be the major mechanism responsible for the
fact that women, at all levels of body fat, have
higher circulating leptin levels than do men.It should be noted that serum leptin levels vary
more directly with total body fat content than withBMI [16]. In the present study, total body fat was a
strong regressor for serum leptin and leptin secretion
rate. Furthermore, gender interacted significantlywith both relationships. However, the results of
covariance analysis showed that the gender differ-
ences in serum leptin and leptin secretion remainedafter the influence of total body fat content was
accounted for. Thus, gender variations in circulatingleptin and leptin secretion are not only due to the
fact that women in general have more body fat than
do men.The presently recorded secretion rates of leptin are
much higher than reported values for leptin secre-
tion from abdominal subcutaneous tissue in vivo[17]. Methodological reasons most probably explain
the reported differences in leptin secretion rates in
vitro versus in vivo. The in vivo investigatorsexpressed leptin production per total weight of
adipose tissue and we used lipid weight as denomi-
nator. In our hands leptin values differ approxi-mately two times using total weight as compared to
lipid weight as denominator. The in vivo investiga-tors used arterial plasma, which has lower leptin
concentrations than the levels in serum of blood
from the antecubital vein. Finally, some leptin mightbe retained by adipose tissue in vivo, thereby
underestimating the true production rate in vivo
[18]. Because of the observed strong correlationbetween secretion rate of leptin in vitro and the
4 6 0 L . H E L L S T R OÈ M E T A L .
Table 2 Clinical data on the obese subjects
Measure
Females
(n = 44)
Males
(n = 23) P
Age (years) 41 � 2 39 � 2 0.66
BMI (kg m22) 40.4 � 0.8 39.3 � 1.5 0.60WHR 0.927 � 0.010 1.016 � 0.011 <0.0001
Percentage body fat 64 � 2 40 � 2 <0.001
Fat cell volume (pL) 866 � 28 866 � 38 0.99
Serum leptin (mg L21) 52 � 3 22 � 11 <0.001
Values are mean � SE. They were compared using Student's two-tailed unpaired t-test.
Obesity was defined as BMI > 27 kg m22.BMI, body mass index; WHR, waist-to-hip ratio; n, number of subjects.
Table 3 Clinical data on the lean subjects
Measure
Females
(n = 19)
Males
(n = 9) P
Age (years) 38 � 2 40 � 3 0.71
BMI (kg m22) 23.5 � 0.4 24.3 � 0.4 0.27
WHR 0.813 � 0.016 0.929 � 0.017 0.0002Percentage body fat 33 � 1 20 � 1 <0.0001
Fat cell volume (pl) 508 � 39 556 � 37 0.42
Serum leptin (mg L21) 13 � 2 6 � 2 <0.01
Values are mean � SE. They were compared using Student's two-tailed unpaired t-test.
Obesity was defined as BMI > 27 kg m22.
BMI, body mass index; WHR, waist-to-hip ratio; n, number of subjects.
# 2000 Blackwell Science Ltd Journal of Internal Medicine 247: 457±462
serum concentration of leptin, we believe that oursecretion values are representative of true release of
the hormone from fat tissue.
Recently, we have shown that there are regionalvariations in leptin secretion; the rates are higher in
subcutaneous than in visceral adipose tissue [19]. Itis unclear at present if gender variations in leptin
secretion rate also exist in visceral adipose tissue.
The observation that women, in spite of a largerfat mass, have an increased production rate of leptin
compared with men suggests that women have a
less efficient leptin signalling system than do men. Ifwe accept that leptin is a major regulator of body fat
stores [18], it may be inferred that the centres in the
brain that control body fat stores have a higher set-point for leptin in women than in men. This could
explain a higher body fat content in women than in
men. Alternatively, other factors could be respon-sible for gender differences in body fat content, so
that the difference in circulating leptin concentra-
tions between the sexes is merely an adaptivephenomenon.
In summary, this study provides evidence that
gender differences in circulating leptin levels arecaused by at least two different mechanisms: a larger
adipose tissue mass and a higher production rate of
leptin per unit mass of adipose tissue in women than
in men.
Acknowledgements
The study was supported by grants from the Knutand Alice Wallenberg Foundation, the Swedish
Medical Society, the Swedish Foundation for Strate-
gic Research, the Karolinska Institute, the SwedishDiabetes Association, the Swedish Medical Research
Council, the Swedish Heart and Lung Foundation,
Novo Nordisk Foundation, Bergwall Foundation andSoÈderberg Foundation.
References
1 Prolo P, Wong ML, Licinio J, Leptin. Int J Biochem Cell Biol
1998; 30: 1285±90.2 Auwerx J, Staels B, Leptin. Lancet 1998; 351: 737±42.
3 Flier J. What's in a name? In search of leptin's physiologic
role. J Clin Endocrinol Metab 1998; 83: 1407±13.
4 Tritos NA, Mantzoros CS. Leptin: its role in obesity andbeyond. Diabetologia 1997; 40: 1371±79.
5 Montague CT, Farooqi IS, Whitehead JP et al. Congenital
leptin deficiency is associated with severe early-onset obesityin humans. Nature 1997; 387: 903±908.
6 Clement K, Vaisse C, Lahlou N et al. A mutation in the
human leptin receptor gene causes obesity and pituitary
dysfunction. Nature 1998; 392: 398±401.7 Saad MF, Damani S, Gingerich RL et al. Sexual dimorphism
in plasma leptin concentration. J Clin Endocrinol Metab 1997;
82: 579±84.
8 Licinio J, Negrao AB, Mantzoros C et al. Sex differences incirculating human leptin pulse amplitude: clinical implica-
tions. J Clin Endocrinol Metab 1998; 83: 4140±47.
9 Casabiell X, Piniero V, Peino R et al. Gender differences inboth spontaneous and stimulated leptin secretion by human
omental adipose tissue in vitro: dexamethasone and estradiol
stimulate leptin release in women, but not in men. J Clin
Endocrinol Metab 1998; 83: 2149±55.10 Pineiro V, Casabiell X, Peino R et al. Dihydrotestosterone,
stanozolol, androstenedione and dehydroepiandrosterone
sulphate inhibit leptin secretion in female but not in male
samples of omental adipose tissue in vitro: lack of effect oftestosterone. J Endocrinol 1999; 160: 425±32.
11 LoÈnnqvist F, Nordfors L, Jansson M, ThoÈrne A, Schalling M,
Arner P. Leptin secretion from adipose tissue in women.
Relationship to plasma levels and gene expression. J ClinInvest 1997; 99: 2398±404.
12 Large V, HellstroÈm L, Reynisdottir S et al. Human beta2-
adrenoceptor gene polymorphisms are highly frequent inobesity and associate with altered adipocyte beta2-adreno-
ceptor function. J Clin Invest 1997; 100: 3005±13.
13 Arner P, Arner O, OÈ stman J. The effect of local anaesthetic
agents on lipolysis by human adipose tissue. Life Sci 1973;13: 161±69.
14 Kolaczynski JW, Morales LM, Moore JH Jr et al. A new
technique for biopsy for human abdominal fat under local
L E P T I N I N M E N A N D W O M E N 4 6 1
425
50
Lept
in s
ecre
tion
(ng
2 h–1
g–1
lipi
d)
ObeseAll Non-obese
350
275
200
125
P < 0.05P < 0.05
P < 0.05
Fig. 2 Adipose tissue secretion rate of leptin in women (open
bars) and men (filled bars). Values are mean 6 SE. Results for the
whole cohort, obese and non-obese subjects are shown separately.Differences between men and women were analysed by Student's
unpaired t-test.
# 2000 Blackwell Science Ltd Journal of Internal Medicine 247: 457±462
anaesthesia with Lidocaine. Int J Obesity 1994; 18: 161±66.
15 Hirsch J, Galllian E. Methods for determination of adipose
cell size and cell number in man and animals. J Lipid Res
1968; 9: 100±19.16 Rosenbaum N, Nicolson M, Hirsch J et al. Effects of gender,
body composition, and menopause on plasma concentra-
tions of leptin. J Clin Endocrinol Metab 1996; 81: 3424±27.
17 Klein S, Coppack SW, Mohamed-Ali V, Landt M. Adiposetissue leptin production and plasma kinetics in humans.
Diabetes 1996; 45: 984±87.
18 Van Harmelen V, Reynisdottir S, Eriksson P et al. Leptin
secretion from subcutaneous and visceral adipose tissue in
women. Diabetes 1998; 47: 913±17.
19 Friedman JM. Leptin, leptin receptors and the control of body
weight. Nutr Rev 1998; 56: S38±46.
Received 13 September 1999; accepted 25 November 1999.
Correspondence: Peter Arner, professor, MD, Department ofMedicine, CME (M61), Huddinge Hospital, S-141 86 Huddinge,
Sweden (fax: + 46 8 58582407; e-mail: [email protected]
ki.se).
4 6 2 L . H E L L S T R OÈ M E T A L .
# 2000 Blackwell Science Ltd Journal of Internal Medicine 247: 457±462