J C E M
O N L I N E
A d v a n c e s
i n
G e n e t i c s — E n d o c r i n e
R e s e a r c h
Evidence for Masculinization of Adipokine Gene
Expression in Visceral and Subcutaneous Adipose
Tissue of Obese Women With Polycystic Ovary
Syndrome (PCOS)
M. Ángeles Martínez-García, Rafael Montes-Nieto, Elena Fernández-Durán,
María Insenser, Manuel Luque-Ramírez, and Héctor F. Escobar-Morreale
Diabetes, Obesity and Human Reproduction Research Group, Universidad de Alcalá and Hospital Universitario
Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria, and Centro de Investigación Biomédica en
Red Diabetes y Enfermedades Metabólicas Asociadas, E-28034 Madrid, Spain
Context: Sex hormones, particularly androgens, may influence not only adipose tissue distribution
but also its functions.
Objective: We explored the possibility of sexual dimorphism in adipose tissue and skeletal muscle
function.
Design: This was a case-control study.
Setting: The setting was an academic hospital.
Participants: Participants were severely obese men (n ⫽ 7), control women (n ⫽ 7), and hyperandrogenic women presenting with polycystic ovary syndrome (PCOS) (n ⫽ 7) submitting to bariatric
surgery and an independent series of 40 patients with PCOS and 40 control women matched for
age and body mass index.
Interventions: Samples of subcutaneous (SAT) and visceral adipose tissue (VAT) and skeletal muscle
were obtained during bariatric surgery in severely obese subjects.
Main Outcome Measures: Gene expression of chemerin, lipocalin-2, and omentin-1 in tissue samples was measured. We analyzed the effects of PCOS and obesity on serum concentrations of these
adipokines in the larger series of women with PCOS and in control women.
Results: Expression of chemerin and lipocalin-2 was higher in VAT than in SAT in men and women
with PCOS; the opposite was observed in control women. Omentin-1 expression was higher in VAT
than in SAT in the three groups. No differences were observed in the skeletal muscle expression of
these adipokines. Obesity increased serum chemerin and lipocalin-2 levels and tended to decrease
omentin-1, irrespective of PCOS.
Conclusions: The present results suggest that there is sexual dimorphism in some adipose tissue
functions and that this dimorphism may be related to differences in androgen concentrations
because women with PCOS show a masculinized pattern of expression of some adipokines. (J Clin
Endocrinol Metab 98: E388 –E396, 2013)
ISSN Print 0021-972X ISSN Online 1945-7197
Printed in U.S.A.
Copyright © 2013 by The Endocrine Society
doi: 10.1210/jc.2012-3414 Received September 20, 2012. Accepted November 27, 2012.
First Published Online January 21, 2013
E388
jcem.endojournals.org
Abbreviations: ARE, Androgen response element; BMI, body mass index; Ct, threshold
cycle; GLM, general linear model; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PCOS, polycystic ovary syndrome; SAT, subcutaneous adipose tissue; VAT, visceral
adipose tissue.
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
besity-dependent diabetes mellitus (diabesity) is one
of the most common chronic disorders and a leading cause of morbidity and mortality in westernized countries (1). Adipose tissue, which at present is considered as
a complex and highly active metabolic and endocrine organ (2), plays a central role in the pathophysiology of
diabesity.
Adipose tissue is composed by a heterogeneous mix of
adipocytes, stromal preadipocytes, immune cells, and endothelium (2). Adipokines are molecules, structurally similar to cytokines, that are secreted by adipocytes, macrophages, and other fat cells and mediate most of the
functions of adipose tissue by acting locally and in many
nonadipose organs and tissues (2).
Adipose tissue consists of subcutaneous (SAT) and visceral adipose tissue (VAT) depots with distinctive characteristics (3). The differential expansion of these depots is
of importance because the association with metabolic disorders is stronger for VAT than for SAT (4), possibly because molecules secreted by VAT may reach the liver, an
essential organ for intermediate metabolism, directly
through the portal circulation (2). Moreover, differences
in the gene expression profile and in biochemical and metabolic properties may also contribute to the preferential
association of VAT with metabolic dysfunction (5).
The sexual dimorphism of body composition, especially with regard to the distribution of adipose tissue, is
most evident after puberty and modulates the risk of diabesity (6). Adult men have a predominantly abdominal
and visceral fat distribution compared with the more peripheral and subcutaneous fat distribution of premenopausal
women, yet these differences tend to disappear after menopause (6). The sexual dimorphism on the distribution of body
fat appears to be the result of the effects of androgens and
estrogens and their balance on adipose tissue (6). Interestingly, premenopausal women presenting with androgen excess, ie, women with polycystic ovary syndrome (PCOS),
frequently show a central distribution of body fat and associated abdominal adiposity, obesity, and diabetes (7). Furthermore, androgen excess influences gene expression and
protein abundance in VAT (8, 9).
With the hypothesis that sexual dimorphism is not only
limited to body fat distribution but also extends to the
functions of adipose tissue we studied in adult men, in
premenopausal women, and in women presenting with
androgen excess the SAT, VAT, and skeletal muscle expression of chemerin, lipocalin-2, and omentin-1. In addition, we studied the impact of androgen excess and of
obesity on their circulating concentrations. We selected
this panel of adipokines because of their crucial roles in the
interplay between adipose tissue, intermediate metabolism, and cardiovascular risk: chemerin participates in cell
O
jcem.endojournals.org
E389
recruitment for innate and acquired immune processes
and in adipogenesis and adipocyte differentiation and influences genes regulating glucose and lipid metabolism
(10); lipocalin-2 is involved in the immune response
against bacteria and inhibits insulin-dependent glucose uptake (11); and omentin-1 not only is an insulin-sensitizing
molecule but also shows anti-inflammatory, antiproliferative, antiangiogenic, and vasodilatory properties (12).
Subjects and Methods
Subjects
For gene expression experiments, we obtained SAT, VAT,
and skeletal muscle biopsy samples from severely obese patients
during bariatric surgery, including 7 men, 7 patients with PCOS,
and 7 women presenting without symptoms of androgen excess,
selected to have similar age and body mass index (BMI). The
surgeon aimed to obtain the biopsy samples at similar places
(subcutaneous fat at the point of incision or insertion of laparoscopic trocar, from rectus abdominis muscle and from omentum) from all the patients.
A total of 80 premenopausal women served to analyze the
impact of androgen excess and of obesity on serum adipokine
levels; 40 patients had PCOS and 40 women did not show any
evidence of androgen excess and served as control women. Approximately half of the patients and half of the control women
were obese (BMI ⱖ30 kg/m2). This series did not include the
severely obese women from whom adipose tissue biopsy samples
were obtained.
The methods used to phenotype the subjects have been reported earlier (13). PCOS was defined by the presence of ovulatory dysfunction, clinical hyperandrogenism and/or hyperandrogenemia, and exclusion of specific etiologies (14). Evidence of
oligo-ovulation was provided by serum luteal phase progesterone levels ⬍4 ng/ml in patients with regular menses, by oligomenorrhea, or by amenorrhea. Clinical hyperandrogenism was
defined by the presence of hirsutism (modified Ferriman-Gallwey score ⱖ8), persistence of acne in women older than 20 years,
or the presence of androgenic alopecia (14). Specific etiologies
were excluded as follows. Hyperprolactinemia and thyroid dysfunction were excluded by the finding of serum prolactin and
thyrotropin levels within the normal range. Basal or cosyntropinstimulated 17-hydroxyprogesterone levels served to rule out
nonclassic 21-hydroxylase deficiency. Clinical assessment
served to rule out androgen-secreting tumors, Cushing syndrome, and anabolic drug use. We considered women presenting
without menstrual and ovulatory dysfunction and who had no
evidence of androgen excess as control women.
Although ovarian morphology was not analyzed, by having
hyperandrogenism and oligo-ovulation all patients fulfilled all
the current definitions of PCOS (14 –16). In addition, PCOS was
ruled out reliably in the control women because all these women
did not have hyperandrogenism and showed regular ovulatory
menstrual cycles.
Written informed consent was obtained from all the participants, and the study was approved by the ethics committee.
E390
Martínez-García et al
Androgens, Adipose Tissue, and Adipokines
Gene expression
SAT, VAT, and muscle samples were obtained from the morbidly obese subjects included in this study during elective bariatric surgery. All patients had fasted for at least 12 hours overnight before the surgical procedure. Adipose tissue and muscle
samples were collected, washed in PBS, and immediately stored
at ⫺80°C until submitted for RNA extraction.
Total RNA was isolated from 100 mg of VAT and SAT samples with an RNeasy Lipid Tissue Mini Kit (Qiagen, Gaithersburg, Maryland) or from 25 mg of muscle tissue with an RNeasy
Fibrous Tissue Mini Kit (Qiagen) according to the manufacturer’s
guidelines. The concentration and quality of the extracted RNA
were assessed by measuring the 260/280 and 260/230 absorbance ratios in a NanoDrop spectrophotometer. RNA isolation
from 3 muscle biopsy samples was unsuccessful; therefore, subsequent gene expression analysis of muscle samples was performed in 18 of 21 patients including 7 men, 6 patients with
PCOS, and 5 control women.
First-strand cDNA was synthesized using an equal amount of
total RNA with a High Capacity RNA to cDNA Kit (Applied
Biosystems, Foster City, California) in 20-␮L reactions following the manufacturer’s instructions.
A TaqMan PreAmp Master Mix Kit (Applied Biosystems)
was used to increase the quantity of specific cDNA targets for
subsequent gene expression studies. Before the preamplification
reaction was run, assays of interest including an endogenous
control were pooled together in a 0.2⫻ pooled assay mix. The
preamplification reaction was performed in 50-␮L reactions
containing 25 ␮L of TaqMan PreAmp Master Mix, 12.5 ␮L of
the pooled assay mix, 5 ␮L of reverse-transcribed cDNA sample,
and 7.5 ␮L of nuclease-free water. The preamplification program consisted of a denaturing phase of 10 minutes at 95°C
followed by 14 cycles of 15 seconds at 95°C and 4 minutes at
60°C. The preamplified cDNA products were diluted 1:20 before PCR.
Gene expression reactions were assessed in a StepOnePlus
Real-Time PCR System (Applied Biosystems, Darmstadt, Germany) with TaqMan technology. The reaction was performed in
a final volume of 20 ␮L containing 10 ␮L of TaqMan Gene
Expression Master Mix, 5 ␮L of diluted preamplified product, 1
␮L of TaqMan Gene Expression assay, and 4 ␮L of nuclease-free
water. The cycle program consisted of an initial denaturing phase
of 10 minutes at 95°C plus 40 cycles of 15 seconds at 95°C and
a 1-minute annealing-extension phase at 60°C.
Prevalidated TaqMan Gene Expression assays (Applied Biosystems)wereusedfortherelativequantificationofchemerin(RARRES2;
Hs00161209_g1), lipocalin-2 (LCN2; Hs01008571_m1), and omentin-1 (ITLN1; Hs00914745_m1). Cyclophilin A (PPIA; 4333763F)
was used as an endogenous control to normalize for target gene
expression in each sample (17). A threshold cycle (Ct) was obtained for each amplification curve, and a ⌬Ct value was first
calculated by subtracting the Ct value for cyclophilin A cDNA
from the Ct value of the specific transcript. Data are expressed as
arbitrary units using the following transformation: expression ⫽
log2⫺⌬Ct. All samples were performed in triplicate, and negative
controls were included in all the reactions.
Search for androgen response elements (AREs) in
gene promoter regions
The hypothesis that androgens may be responsible for any
difference among men, patients with PCOS, and control women
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
in the adipose tissue gene expression patterns of the adipokines
studied here requires that the genes encoding these molecules
contain putative AREs. We searched for AREs in the promoter
regions (from ⫺1000 to ⫹ 200 base pairs of the start of transcription site) of the chemerin (RARRES2), lipocalin-2 (LCN2),
and omentin-1 (ITLN1) genes, obtained from the Transcriptional Regulatory Element Database (http://rulai.cshl.edu/cgibin/TRED/tred.cgi?process⫽home). The search was conducted
using MatInspector software, with the V$ARE.01, V$ARE.02,
and V$ARE.03 matrixes contained in the MatBase version 8.4
library (Genomatix Software GmbH, Munich, Germany).
Adipokine assays
Serum was assayed for chemerin, lipocalin-2, and omentin-1
using commercial ELISA kits following the manufacturer’s instructions (BioVendor Laboratorní medicina a.s., Brno, Czech
Republic). The lower limits of detection and intra-assay and interassay coefficients of variation were 0.1 ng/mL, 6.1%, and
7.6% for chemerin; 0.02 ng/ml, 7.7%, and 9.7% for lipocalin-2;
and 0.5 ng/ml, 3.7%, and 4.6% for omentin-1.
Statistical analysis
Nominal variables were analyzed by Pearson ␹2 test. Continuous variables are reported as means ⫾ SD (text and tables) or
means ⫾ SEM (figures). The Kolmogorov-Smirnov statistic was
applied to continuous variables. We applied logarithmic or
square root transformations as needed to ensure normal distribution of the variables. Differences in phenotypic variables
among men, women, and patients with PCOS were analyzed by
1-way ANOVA followed by the least significant difference test
for post hoc comparisons. Then, the differences in gene expression and circulating adipokine concentrations were analyzed by
general linear models (GLMs).
Adipose tissue gene expression data were submitted to repeatedmeasures GLM, introducing adipose tissue depot (SAT vs VAT) as
the within-subjects effect and group of subjects as the betweensubjects effect. The GLM for muscle gene expression data included
only group (men, women, and patients with PCOS) as the betweensubjects effect. We also used 1-way ANOVA and least significant
difference tests to compare the differences in the VAT and SAT gene
expression of adipokines, calculated for each adipokine as expression in VAT minus expression in SAT, among men, patients with
PCOS, and control women. Sample size analysis based on published
differences between men and women in the adipose tissue expression of chemerin (18) indicated that 6 men and 6 women were
needed to achieve 0.80 power for ␣ ⫽ .05.
The comparisons of serum adipokine levels in patients with
PCOS and control women used a 2-way GLM, in which the
effects of having PCOS and of being obese and the interaction
between both effects were analyzed. Sample size analysis was
based on the differences between patients with PCOS and control
women in plasma omentin-1 concentrations reported earlier
(19). A total of 20 patients and 20 control women were needed
to achieve 0.80 power for ␣ ⫽ .05.
We used SPSS Statistics 17.0 (SPSS Ibérica, Madrid, Spain) for
analyses with the exception of sample size analyses, which were
calculated using Ene 3.0 software (Departamento de Biometría,
GlaxoSmithKline, S.A., Tres Cantos, Spain). We set ␣ ⫽ .05 as
the level of statistical significance for all analyses.
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
jcem.endojournals.org
Results
Expression of adipokines in adipose tissue and muscle
The clinical, metabolic, and hormonal characteristics
of 21 morbidly obese subjects included in the study are
summarized in Table 1. As expected from the design, we
found no differences in age and BMI between men, women
with PCOS, and control women. The rates of metabolic
complications (hypertension, dyslipidemia, or diabetes)
were similar among these groups (4 of 7 men, 3 of 7
women with PCOS, and 2 of 7 control women, ␹2 ⫽ 1.17,
P ⫽ .558) and, with the exception of lower high-density
lipoprotein (HDL) cholesterol in men compared with
women, no statistically significant differences were observed among the groups in fasting glucose and insulin
concentrations, homeostasis model assessment of insulin
resistance, and lipid profiles. The waist to hip ratio and
total and free testosterone concentrations increased,
whereas SHBG and HDL cholesterol concentrations decreased, in men compared with those in control women.
Patients with PCOS showed intermediate values between
men and control women, including increased free testosterone concentrations and decreased SHBG levels when
compared with control women.
The expression of chemerin and lipocalin-2 in SAT and
VAT was the opposite in men and women: their expression
was higher in VAT than in SAT in men, whereas control
women showed higher expression levels in SAT than in
E391
VAT (Figures 1 and 2). Interestingly, the results in the
samples obtained from women with PCOS resembled
those of men: expression of chemerin and lipocalin-2 was
higher in VAT than in SAT (Figures 1 and 2). When men,
control women, and women with PCOS were considered as
a whole, no differences in chemerin expression were found
among adipose tissue depots, whereas the expression of lipocalin-2 was higher in VAT than in SAT (Figure 1).
Expression of omentin-1 was higher in VAT than in SAT,
and no differences in the pattern of expression were found
among men, control women, and patients with PCOS (Figures 1 and 2). We found no differences among these groups
in the expression of chemerin, lipocalin-2, and omentin-1 in
skeletal muscle (Figure 1). However, analysis of the promoter
regions showed putative AREs in the promoters of the
chemerin, lipocalin-2, and omentin-1 genes (Table 2).
Effects of PCOS and obesity on serum
concentrations of chemerin, lipocalin-2, and
omentin-1
The effects of obesity, PCOS, and their interaction on
clinical and biochemical variables are summarized in Table 3. Patients with PCOS were characterized by higher
hirsutism scores and androgen levels, insulin resistance
indexes and triglyceride levels, and reduced SHBG and
HDL cholesterol concentrations, irrespective of obesity
(Table 3). Obesity increased the waist to hip ratio, free
Table 1. Clinical, Metabolic, and Hormonal Variables in Severely Obese Men, Patients With PCOS, and Control
Women From Whom Adipose Tissue and Skeletal Muscle Biopsy Samples Were Obtained During Bariatric Surgery for
Gene Expression Studiesa
Age, y
BMI, kg/m2
Waist to hip ratiob,c
Hirsutism score
Ovulatory dysfunction
Total testosterone, ng/dLb,c
Free testosterone, ng/dLb,c,d
SHBG, ␮g/dLb,d
DHEAS, ng/mL
Androstenedione, ng/mL
Total cholesterol, mg/dL
LDL cholesterol, mg/dL
HDL cholesterol, mg/dLb
Triglycerides, mg/dL
Fasting glucose, mg/dL
Fasting insulin, ␮IU/mL
HOMA-IR
Men (n ⴝ 7)
33.6 ⫾ 6.5
49.7 ⫾ 6.2
1.03 ⫾ 0.08
314 ⫾ 89
9.1 ⫾ 2.2
126 ⫾ 45
1953 ⫾ 1327
2.6 ⫾ 0.9
193 ⫾ 35
120 ⫾ 35
31 ⫾ 4
204 ⫾ 89
94 ⫾ 16
34 ⫾ 13
7.9 ⫾ 3.8
Women With PCOS (n ⴝ 7)
30.6 ⫾ 5.1
53.3 ⫾ 5.3
0.83 ⫾ 0.08
7.9 ⫾ 5.1
7 (100)
66 ⫾ 29
1.5 ⫾ 0.5
189 ⫾ 117
1695 ⫾ 737
3.4 ⫾ 2.0
174 ⫾ 46
120 ⫾ 39
35 ⫾ 8
150 ⫾ 106
133 ⫾ 97
26 ⫾ 16
6.8 ⫾ 3.5
Control Women (n ⴝ 7)
32.9 ⫾ 6.1
51.3 ⫾ 6.5
0.85 ⫾ 0.06
1.9 ⫾ 1.6
0 (0)
58 ⫾ 32
0.9 ⫾ 0.4
344 ⫾ 148
1363 ⫾ 553
2.6 ⫾ 0.3
182 ⫾ 27
112 ⫾ 19
46 ⫾ 12
115 ⫾ 53
101 ⫾ 27
21 ⫾ 19
5.3 ⫾ 4.8
P Value
.623
.546
⬍.001
.020
⬍.001
⬍.001
⬍.001
.007
.560
.374
.694
.930
.029
.184
.438
.322
.490
Abbreviations: DHEAS, dehydroepiandrosterone sulfate; HOMA-IR, homeostasis model assessment of insulin resistance.
a
Data are means ⫾ SD or counts (percentage). Data were submitted to 1-way ANOVA followed by the least significant difference post hoc test.
Discontinuous variables were compared by ␹2 test.
b
P ⬍ .05 or less for the difference between men and control women.
c
P ⬍ .05 or less for the difference between men and women with PCOS.
d
P ⬍ .05 or less for the difference between patients with PCOS and control women.
E392
Martínez-García et al
Androgens, Adipose Tissue, and Adipokines
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
Figure 1. Differences between obese men, obese patients with PCOS,
and obese control women in the gene expression of chemerin, lipocalin-2,
and omentin-1 in SAT and VAT and in skeletal muscle. Data are means ⫾
SEM. *P ⬍ .05 or less for the interaction between adipose tissue depot
and group of subjects, meaning that the differences in gene expression
were not the same in men, patients with PCOS, and control women.
†P ⬍ .05 or less for differences between SAT and VAT considering men,
patients with PCOS, and control women as a whole. There were no
differences in the global adipose tissue gene expression of chemerin
(P ⫽ .540), lipocalin-2 (P ⫽ .423), and omentin-1 (P ⫽ .913) between
men, patients with PCOS, and control women.
testosterone concentrations, fasting glucose, indexes of insulin resistance, triglyceride and total and low-density lipoprotein (LDL) cholesterol concentrations, and decreased SHBG and HDL cholesterol levels, irrespective of
PCOS (Table 3). The only interaction observed was caused
by fasting glucose, which was lower in nonobese women
with PCOS and higher in obese patients with PCOS, when
compared with their nonhyperandrogenic counterparts
(Table 3).
Serum concentrations of chemerin and lipocalin-2
showed a statistically significant increase in obese women
compared with that in nonobese women, whereas an opposite tendency (P ⫽ .066) toward a decrease in omentin-1
concentrations in obese women did not reach statistical
significance (Figure 3). The presence of PCOS had no ef-
Figure 2. Differences in the expression of adipokines between VAT
and SAT, as indicated by the expression in VAT (log2⫺⌬Ct) minus the
expression in SAT (log2⫺⌬Ct), among severely obese men, obese
patients with PCOS, and obese control women. Data are means ⫾ SEM. P
values indicate the differences between individual groups according to
1-way ANOVA followed by least significant difference post hoc test.
fect on serum levels of these adipokines, and no interaction
between obesity and PCOS was found (Figure 3).
Discussion
Our present results indicate that there is sexual dimorphism in the adipose tissue expression of certain adipokines and that the expression patterns of these adipokines
in women with androgen excess, such as the patients with
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
jcem.endojournals.org
E393
Table 2. Putative AREs in the Promoter Regions of Adipokine Genesa
Gene (Protein)
RARRES2 (chemerin)
ITLN1 (omentin-1)
LCN2 (lipocalin-2)
Sequenceb
caaggactttctGTGCccc
ctatgtcatgaaGTTCtca
caggacctgcatGTGCtcc
Matrix Similarity
0.940
0.892
0.896
Strand
⫺
⫺
⫺
Family/Matrix
V$GREF/ARE.02
V$GREF/ARE.02
V$GREF/ARE.03
a
The search for AREs was conducted with MatInspector software using the V$ARE.01, V$ARE.02, and V$ARE.03 matrixes contained in the
MatBase version 8.4 library. Matrix similarity values ⬎0.80 are considered optimal, and perfect matches result in values of 1.00.
b
Uppercase letters represent the core sequence.
PCOS studied herein, are similar to those of men. Men and
patients with PCOS had higher expression of chemerin
and lipocalin-2 in VAT than in SAT, whereas control
women showed the opposite, albeit the overall adipose
tissue expression of these adipokines was similar in the
three groups of subjects. The possibility that androgens
played a role in these findings is supported by the presence
of putative AREs in the promoter regions of the genes
encoding chemerin and lipocalin-2, yet only a regulatory
effect of androgens on lipocalin-2 expression is supported
by scientific data derived from rat studies (20, 21).
However, other factors might have been also involved,
obviously including the differences in estrogen concentrations among men, control women, and women with PCOS
and possibly changes in other adipokines and inflammatory mediators that have been found to be abnormal in
women with PCOS (7, 22, 23). Of note, the promoter
region of the gene encoding lipocalin-2 also contains an
estrogen response element (24). Moreover, recent animal
studies demonstrated that both chemerin and lipocalin-2
influence the expression of aromatase, a key enzyme for
the peripheral metabolism of sex hormones that catalyzes
the conversion of androgens into estrogens at the tissue
level, in ovarian granulosa cells (25) and in adipose tissue
(26). In particular, the expression patterns of lipocalin-2 in
SAT (inguinal white fat) and VAT (perigonadal white fat)
from adult mice resembled those of the men and control
women studied here, and lipocalin-2 deficiency reduced
aromatase expression, thereby decreasing estradiol production in adipose tissue (26). If confirmed in humans,
such a decrease in aromatase activity might impair the
conversion of androgens into estrogens in adipose tissue,
further contributing to the imbalance between androgens
and estrogens characteristic of PCOS.
Our present results are in conceptual agreement with
the existence of sexual dimorphism in adipose tissue dis-
Table 3. Clinical and Biochemical Variables of Patients With PCOS and Nonhyperandrogenic Control Women as a
Function of Obesity, Defined by a BMI ⱖ30 kg/m2a
Nonobese Women
(n ⴝ 40)
Age, y
BMI, kg/m2
Hirsutism score
Waist to hip ratio
Total testosterone, ng/dL
Free testosterone, ng/dL
SHBG, ␮g/dL
DHEAS, ng/mL
Androstenedione, ng/mL
Total cholesterol, mg/dL
LDL cholesterol, mg/dL
HDL cholesterol, mg/dL
Triglycerides, mg/dL
Fasting glucose, mg/dL
Fasting insulin, ␮U/mL
HOMA-IR
Insulin sensitivity index
Obese Women
(n ⴝ 40)
PCOS vs
Control Womenb
Nonobese vs
Obese Womenc
Interactiond
PCOS
(n ⴝ 19)
Control
(n ⴝ 20)
PCOS
(n ⴝ 21)
Control
(n ⴝ 20)
F
P
Value
F
P
Value
F
P
Value
26 ⫾ 6
25 ⫾ 4
10.2 ⫾ 4.1
0.76 ⫾ 0.08
56 ⫾ 17
1.0 ⫾ 0.4
333 ⫾ 171
2027 ⫾ 774
3.2 ⫾ 1.1
160 ⫾ 32
94 ⫾ 22
49 ⫾ 15
87 ⫾ 38
87 ⫾ 10
9.4 ⫾ 4.8
2.1 ⫾ 2.9
7.2 ⫾ 4.3
27 ⫾ 6
25 ⫾ 4
2.0 ⫾ 2.2
0.74 ⫾ 0.06
39 ⫾ 14
0.6 ⫾ 0.2
423 ⫾ 180
1621 ⫾ 811
2.3 ⫾ 0.9
172 ⫾ 27
104 ⫾ 22
56 ⫾ 14
61 ⫾ 17
91 ⫾ 6
6.2 ⫾ 3.6
1.3 ⫾ 0.8
8.0 ⫾ 4.5
28 ⫾ 6
35 ⫾ 4
9.2 ⫾ 7.2
0.83 ⫾ 0.07
63 ⫾ 23
1.4 ⫾ 0.7
243 ⫾ 162
2247 ⫾ 1179
3.4 ⫾ 1.1
189 ⫾ 30
120 ⫾ 24
42 ⫾ 10
145 ⫾ 107
99 ⫾ 9
14.4 ⫾ 9.3
3.6 ⫾ 2.3
4.2 ⫾ 4.8
27 ⫾ 5
35 ⫾ 4
1.5 ⫾ 2.2
0.80 ⫾ 0.07
41 ⫾ 13
0.8 ⫾ 0.3
297 ⫾ 108
1694 ⫾ 774
2.6 ⫾ 1.1
177 ⫾ 37
115 ⫾ 28
47 ⫾ 10
80 ⫾ 22
95 ⫾ 6
9.8 ⫾ 4.8
2.3 ⫾ 1.2
5.3 ⫾ 3.2
0.096
0.011
63.08
2.291
25.44
34.10
4.199
5.814
10.45
⬍0.001
0.193
4.862
14.51
0.004
4.146
5.284
4.550
.757
.917
⬍.001
.134
⬍.001
⬍.001
.044
.018
.002
.997
.662
.030
⬍.001
.947
.045
.024
.036
0.763
138.4
0.571
15.04
1.014
8.269
10.24
0.529
2.252
5.692
11.27
9.130
13.48
20.03
4.937
8.240
15.37
.385
⬍.001
.452
⬍.001
.317
.005
.002
.469
.138
.020
.001
.003
⬍.001
⬍.001
.029
.005
⬍.001
0.313
0.015
0.044
0.256
0.383
0.010
0.257
0.060
0.168
2.529
1.839
0.217
0.456
5.216
0.132
0.436
0.995
.578
.904
.834
.614
.538
.922
.614
.807
.683
.116
.179
.642
.501
.025
.718
.511
.322
Abbreviations: DHEAS, dehydroepiandrosterone sulfate; HOMA-IR, homeostasis model assessment of insulin resistance.
a
Data are shown as means ⫾ SD. The differences in continuous variables among groups were analyzed by univariate 2-way general linear models.
b
A significant result indicates a difference between all patients with PCOS and all control women considered as a whole, independent of whether
they were obese or nonobese.
c
A significant result indicates a difference between all the nonobese women and all the obese women considered as a whole, independent of
whether they were patients with PCOS or control women.
d
A significant result indicates a difference between patients with PCOS and control women that was present only in nonobese or only in obese
subgroups and/or a difference between nonobese and obese women that was only present in the patients or only in the control women.
E394
Martínez-García et al
Androgens, Adipose Tissue, and Adipokines
Figure 3. Effects of obesity and PCOS on serum concentrations of
chemerin, lipocalin-2, and omentin-1. Data are means ⫾ SEM.
*P ⬍ .01 or less for the difference between obese and nonobese
women irrespective of having PCOS. No effects of PCOS or interactions
between obesity and PCOS were observed.
tribution and function suggested by previous studies. The
abundance of several proteins related to the pathophysiology of obesity in SAT and VAT is different in men and
in women (27). Men and patients with PCOS have a predominantly abdominal and visceral disposition of body
fat (7), and VAT is associated with low-grade chronic inflammation. Furthermore, sex hormones appear to influence the immune system, resulting in a sexual dimorphism
in the immune response in humans (28), and low-grade
inflammation has been described in association with
PCOS (23). Chemerin and lipocalin-2 are involved in ad-
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
ipogenesis and adipocyte differentiation (10, 29). Their
synthesis is induced by the overexpression of proinflammatory cytokines in VAT, and both chemerin and lipocalin-2 participate in the recruitment and local activation of
inflammatory cells in adipose tissue (10, 30). Therefore,
the predominant expression of these adipokines in the
VAT of men and women with PCOS compared with the
predominant expression in SAT found in control women
may simply derive from the higher intensity of low-grade
chronic inflammation in VAT. In addition, their relative
overexpression in VAT may actually be the result of underexpression in the SAT of men and women with PCOS.
However, recent evidence suggests that chemerin and
lipocalin-2 might exert local autocrine and paracrine antiinflammatory actions in adipose tissue (31, 32). We may
speculate that the predominant expression in VAT found
in men and women with PCOS could be a protective response against inflammation (10, 29). Anyway, the precise
physiological role of these adipokines is still unclear because both chemerin and lipocalin-2 may exert dual effects
on inflammation, depending on the context in which their
effects are explored (10, 29).
Adipose tissue expression of omentin-1 did not show
any evidence of sexual dimorphism, even when putative
AREs are also present in its promoter. This result further
supports the participation of factors other than androgens
in the differences observed in the patterns of adipokine
adipose tissue expression as mentioned above. Omentin-1
expression was higher in VAT than in SAT in all groups,
and the overall expression was not different among men,
patients with PCOS, and control women. In adipose tissue, omentin-1 is predominantly expressed in visceral stromal vascular cells (33), in conceptual agreement with our
present results using whole adipose tissue biopsy samples.
Furthermore, the expression of chemerin, lipocalin-2, and
omentin-1 in skeletal muscle did not show any difference
among the three groups, suggesting that the sexual dimorphism in the expression of chemerin and lipocalin-2 is
specific to adipose tissue.
However, previous studies addressing the impact of sex
and PCOS on the expression of some of these adipokines
showed different results (18, 19, 34). In unselected men
and women submitting to elective surgery, chemerin expression was higher in SAT than in VAT, although this
difference did not persist after separate analysis of men
and women (18). Overweight and mildly obese patients
with PCOS showed increased chemerin expression and
reduced omentin-1 expression in adipose tissue (19, 34).
In these patients, no differences between SAT and VAT
were found for chemerin expression, whereas expression
of omentin-1 was higher in VAT and was barely detectable
in SAT (19, 34). Differences in the grade of obesity (our
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
subjects had much more severe obesity compared with the
less severe excess body fat in previous cohorts), in the exact
location from which adipose tissue biopsy samples were
obtained, and the lack of a male control group in the PCOS
studies and perhaps a less strict statistical control of multiple comparisons in these earlier studies may account for
the differences with our present results. Nevertheless, our
present study was conducted using whole adipose tissue
samples, an important limitation, considering that the adipokines investigated here are involved in inflammatory
processes that may not be the same in the adipocytes and
stromal vascular fractions of adipose tissue.
In our study, the sexual dimorphism and putative effects of androgens in chemerin and lipocalin-2 adipose
tissue expression did not translate into their circulating
concentrations, which were similar in patients with PCOS
and control women. This finding suggests that the serum
concentrations of these adipokines depend on their overall
expression in adipose tissue, irrespective of the fat depot in
which they are predominantly expressed, and that the putative effect of androgens and sexual dimorphism in their
expression in VAT and SAT cannot be evaluated by measuring their circulating levels. An alternative explanation
is that we were not able to find differences in serum adipokine levels of patients with PCOS and control women
simply because this series included lean individuals and
women with less severe excess body fat than those used for
gene expression studies.
Serum chemerin and lipocalin-2 levels increased in
obese women compared with those in nonobese women,
and serum omentin-1 concentrations tended to decrease,
irrespective of the presence or absence of PCOS. Because
omentin-1 enhances insulin-stimulated uptake in adipocytes, the decrease in circulating omentin-1 levels has
been proposed to reflect the dysmetabolic insulin-resistant
state associated with obesity (12).
The effects of obesity on serum adipokine concentrations observed in our study are in agreement with the effects of obesity observed in previous reports in humans
(10, 35, 36), but previous studies addressing the effects of
PCOS on serum levels of chemerin, lipocalin-2, and omentin-1 have shown discrepant results. In the PCOS study
mentioned earlier (34), serum chemerin levels were higher
in overweight and mildly obese patients with PCOS than
in control women. Lipocalin-2 levels have been found to
be reduced (37), normal (38), or even increased (39) in
patients with PCOS, whereas serum omentin-1 levels have
been reported to be decreased (19, 40, 41) or normal (42)
in such patients. Together with our present results, these
discrepancies cast further doubts on the accuracy of serum
levels of these adipokines in translating possible changes in
their expression at the tissue level.
jcem.endojournals.org
E395
In summary, our present study suggests that the sexual
dimorphism of adipose tissue is not restricted to the distribution of body fat but also influences the expression of
several adipokines in the different adipose tissue depots.
Furthermore, the fact that the expression pattern in SAT
and VAT of some of these adipokines is similar in men and
in women presenting with androgen excess suggests that
androgen levels might influence adipose tissue function
and not only its distribution.
Acknowledgments
Address all correspondence and requests for reprints to: Héctor
F. Escobar-Morreale, Hospital Universitario Ramón y Cajal and
Universidad de Alcalá, Carretera de Colmenar km 9⬘1, E-28034
Madrid, Spain. E-mail: [email protected].
This study was supported by Grants FIS PI080944 and
PI1100357 from Instituto de Salud Carlos III, Spanish Ministry
of Economy and Competitiveness, Grant EC10-096 from Dirección General de Farmacia, Spanish Ministry of Health, Social
Services and Equality, and Grant DIASOBS from Centro de Investigación Biomédica en Red Diabetes and Enfermedades
Metabólicas Asociadas.
All authors researched data, contributed to discussion, wrote
and edited the manuscript, and act as guarantors for this
manuscript.
Disclosure Summary: The authors have no conflict of interest
to disclose.
References
1. Astrup A, Finer N. Redefining type 2 diabetes: ‘diabesity’ or ‘obesity
dependent diabetes mellitus’? Obes Rev. 2000;1:57–59.
2. Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin
Endocrinol Metab. 2004;89:2548 –2556.
3. Ibrahim MM. Subcutaneous and visceral adipose tissue: structural
and functional differences. Obes Rev. 2010;11:11–18.
4. Goodpaster BH, Krishnaswami S, Harris TB, et al. Obesity, regional
body fat distribution, and the metabolic syndrome in older men and
women. Arch Intern Med. 2005;165:777–783.
5. Linder K, Arner P, Flores-Morales A, Tollet-Egnell P, Norstedt G.
Differentially expressed genes in visceral or subcutaneous adipose
tissue of obese men and women. J Lipid Res. 2004;45:148 –154.
6. Wells JC. Sexual dimorphism of body composition. Best Pract Res
Clin Endocrinol Metab. 2007;21:415– 430.
7. Escobar-Morreale HF, San Millan JL. Abdominal adiposity and the
polycystic ovary syndrome. Trends Endocrinol Metab. 2007;18:
266 –272.
8. Corton M, Botella-Carretero JI, Benguria A, et al. Differential gene
expression profile in omental adipose tissue in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2007;92:328 –337.
9. Corton M, Botella-Carretero JI, Lopez JA, et al. Proteomic analysis
of human omental adipose tissue in the polycystic ovary syndrome
using two-dimensional difference gel electrophoresis and mass spectrometry. Hum Reprod. 2008;23:651– 661.
10. Ernst MC, Sinal CJ. Chemerin: at the crossroads of inflammation
and obesity. Trends Endocrinol Metab. 2010;21:660 – 667.
11. Moreno-Navarrete JM, Manco M, Ibanez J, et al. Metabolic endo-
E396
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Martínez-García et al
Androgens, Adipose Tissue, and Adipokines
toxemia and saturated fat contribute to circulating NGAL concentrations in subjects with insulin resistance. Int J Obes. 2010;34:
240 –249.
Tan BK, Adya R, Randeva HS. Omentin: a novel link between inflammation, diabesity, and cardiovascular disease. Trends Cardiovasc Med. 2010;20:143–148.
Escobar-Morreale HF, Sanchon R, San Millan JL. A prospective
study of the prevalence of nonclassical congenital adrenal hyperplasia among women presenting with hyperandrogenic symptoms
and signs. J Clin Endocrinol Metab. 2008;93:527–533.
Zawadzki JK, Dunaif A. Diagnostic criteria for polycystic ovary
syndrome: Towards a rational approach. In: Dunaif A, Givens JR,
Haseltine FP, Merriam GR, eds. Polycystic ovary syndrome. Boston,
MA: Blackwell Scientific Publications; 1992:377–384.
Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop
group. Revised 2003 consensus on diagnostic criteria and long-term
health risks related to polycystic ovary syndrome (PCOS). Hum
Reprod. 2004;19:41– 47.
Azziz R, Carmina E, Dewailly D, et al. Positions statement: criteria
for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society guideline. J Clin
Endocrinol Metab. 2006;91:4237– 4245.
Neville MJ, Collins JM, Gloyn AL, McCarthy MI, Karpe F. Comprehensive human adipose tissue mRNA and microRNA endogenous control selection for quantitative real-time-PCR normalization. Obesity (Silver Spring). 2011;19:888 – 892.
Alfadda AA, Sallam RM, Chishti MA, et al. Differential patterns of
serum concentration and adipose tissue expression of chemerin in
obesity: adipose depot specificity and gender dimorphism. Mol
Cells. 2012;33:591–596.
Tan BK, Adya R, Farhatullah S, et al. Omentin-1, a novel adipokine,
is decreased in overweight insulin-resistant women with polycystic
ovary syndrome: ex vivo and in vivo regulation of omentin-1 by
insulin and glucose. Diabetes. 2008;57:801– 808.
Nantermet PV, Xu J, Yu Y, et al. Identification of genetic pathways
activated by the androgen receptor during the induction of proliferation in the ventral prostate gland. J Biol Chem. 2004;279:1310 –
1322.
Asirvatham AJ, Schmidt M, Gao B, Chaudhary J. Androgens regulate the immune/inflammatory response and cell survival pathways
in rat ventral prostate epithelial cells. Endocrinology. 2006;147:
257–271.
Escobar-Morreale HF, Villuendas G, Botella-Carretero JI, et al. Adiponectin and resistin in PCOS: a clinical, biochemical and molecular
genetic study. Hum Reprod. 2006;21:2257–2265.
Escobar-Morreale HF, Luque-Ramirez M, Gonzalez F. Circulating
inflammatory markers in polycystic ovary syndrome: a systematic
review and metaanalysis. Fertil Steril. 2011;95:1048 –1058.e1– e2.
Seth P, Porter D, Lahti-Domenici J, Geng Y, Richardson A, Polyak
K. Cellular and molecular targets of estrogen in normal human
breast tissue. Cancer Res. 2002;62:4540 – 4544.
Wang Q, Kim JY, Xue K, Liu JY, Leader A, Tsang BK. Chemerin,
a novel regulator of follicular steroidogenesis and its potential involvement in polycystic ovarian syndrome. Endocrinology. 2012;
153:5600 –5611.
Guo H, Zhang Y, Brockman DA, Hahn W, Bernlohr DA, Chen X.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
J Clin Endocrinol Metab, February 2013, 98(2):E388 –E396
Lipocalin 2 deficiency alters estradiol production and estrogen receptor signaling in female mice. Endocrinology. 2012;153:1183–
1193.
Insenser M, Montes-Nieto R, Vilarrasa N, et al. A nontargeted proteomic approach to the study of visceral and subcutaneous adipose
tissue in human obesity. Mol Cell Endocrinol. 2012;363:10 –19.
Bouman A, Heineman MJ, Faas MM. Sex hormones and the immune response in humans. Hum Reprod Update. 2005;11:411–
423.
Esteve E, Ricart W, Fernandez-Real JM. Adipocytokines and insulin
resistance: the possible role of lipocalin-2, retinol binding protein-4,
and adiponectin. Diabetes Care. 2009;32(suppl 2):S362⫺S367.
Li C, Chan YR. Lipocalin 2 regulation and its complex role in inflammation and cancer. Cytokine. 2011;56:435– 441.
Rabe K, Lehrke M, Parhofer KG, Broedl UC. Adipokines and insulin
resistance. Mol Med. 2008;14:741–751.
Zhang J, Wu Y, Zhang Y, Leroith D, Bernlohr DA, Chen X. The role
of lipocalin 2 in the regulation of inflammation in adipocytes and
macrophages. Mol Endocrinol. 2008;22:1416 –1426.
Yang RZ, Lee MJ, Hu H, et al. Identification of omentin as a novel
depot-specific adipokine in human adipose tissue: possible role in
modulating insulin action. Am J Physiol Endocrinol Metab. 2006;
290:E1253⫺E1261.
Tan BK, Chen J, Farhatullah S, et al. Insulin and metformin regulate
circulating and adipose tissue chemerin. Diabetes. 2009;58:1971–
1977.
Wang Y, Lam KS, Kraegen EW, et al. Lipocalin-2 is an inflammatory
marker closely associated with obesity, insulin resistance, and hyperglycemia in humans. Clin Chem. 2007;53:34 – 41.
de Souza Batista CM, Yang RZ, Lee MJ, et al. Omentin plasma levels
and gene expression are decreased in obesity. Diabetes. 2007;56:
1655–1661.
Diamanti-Kandarakis E, Livadas S, Kandarakis SA, Margeli A, Papassotiriou I. Serum concentrations of atherogenic proteins neutrophil gelatinase-associated lipocalin and its complex with matrix
metalloproteinase-9 are significantly lower in women with polycystic ovary syndrome: hint of a protective mechanism? Eur J Endocrinol. 2008;158:525–531.
Koiou E, Tziomalos K, Katsikis I, et al. Weight loss significantly
reduces serum lipocalin-2 levels in overweight and obese women
with polycystic ovary syndrome. Gynecol Endocrinol. 2012;28:
20 –24.
Cakal E, Ozkaya M, Engin-Ustun Y, Ustun Y. Serum lipocalin-2 as
an insulin resistance marker in patients with polycystic ovary syndrome. J Endocrinol Invest. 2011;34:97–100.
Mahde A, Shaker M, Al-Mashhadani Z. Study of omentin1 and
other adipokines and hormones in PCOS patients. Oman Med J.
2009;24:108 –118.
Choi JH, Rhee EJ, Kim KH, Woo HY, Lee WY, Sung KC. Plasma
omentin-1 levels are reduced in non-obese women with normal glucose tolerance and polycystic ovary syndrome. Eur J Endocrinol.
2011;165:789 –796.
Akbarzadeh S, Ghasemi S, Kalantarhormozi M, et al. Relationship
among plasma adipokines, insulin and androgens level as well as
biochemical glycemic and lipidemic markers with incidence of
PCOS in women with normal BMI. Gynecol Endocrinol. 2012;28:
521–524.