Metabolic effects of estrogen substitution in

Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
Contents lists available at SciVerse ScienceDirect
Journal of Steroid Biochemistry and Molecular Biology
journal homepage: www.elsevier.com/locate/jsbmb
Metabolic effects of estrogen substitution in combination with targeted exercise
training on the therapy of obesity in ovariectomized Wistar rats
Nora Zoth a,∗ , Carmen Weigt a , Sinan Zengin a , Oliver Selder a , Nadine Selke a , Michael Kalicinski a ,
Marion Piechotta b , Patrick Diel a
a
b
Department of Cellular and Molecular Sports Medicine, Institute of Cardiovascular Research and Sports Medicine, German Sport University Cologne, Germany
Clinics for Cattle, Institute of Endocrinology, University of Veterinary Medicine Hannover, Germany
a r t i c l e
i n f o
Article history:
Received 28 July 2011
Received in revised form 10 January 2012
Accepted 12 January 2012
Keywords:
Obesity
Metabolic syndrome
Estrogens
Physical activity
Leptin
Respiratory quotient
a b s t r a c t
Postmenopausal women tend to have a higher risk in developing obesity and thus metabolic syndrome.
Recently we could demonstrate that physical activity and estrogen replacement are effective strategies
to prevent the development of nutritional induced obesity in an animal model. The aim of this study was
to determine the combined effects of estrogen treatment and exercise training on already established
obesity. Therefore ovariectomized (OVX) and sham-operated (SHAM) female Wistar rats were exposed
to a high fat diet for ten months. After this induction period obese SHAM and OVX rats either remained
sedentary or performed treadmill training for six weeks. In addition OVX rats were treated with 17␤Estradiol (E2 ) alone, or in combination with training. Before and after intervention effects on lipid and
glucose metabolism were investigated. Training resulted in SHAM and OVX rats in a significant decrease
of body weight, subcutaneous and visceral body fat, size of adipocytes and the serum levels of leptin,
cholesterol, low-density lipoprotein and triglycerides. In OVX animals E2 treatment resulted in similar
effects. Often the combination of E2 treatment and training was most effective. Analysis of the respiratory
quotient indicates that SHAM animals had a better fat burning capacity than OVX rats. There was a
tendency that training in SHAM animals and E2 treatment in OVX animals could improve this capacity.
Analysis of glucose metabolism revealed that obese SHAM animals had higher glucose tolerance than
OVX animals. Training improved glucose tolerance in SHAM and OVX rats, E2 treatment in OVX rats.
The combination of both was most effective. Our results indicate that even after a short intervention
period of six weeks E2 treatment and exercise training improve parameters related to lipid as well as
glucose metabolism and energy expenditure in a model of already established obesity. In conclusion a
combination of hormone replacement therapy and exercise training could be a very effective strategy
to encourage the therapy of diet-induced obesity and its metabolic consequences in postmenopausal
women.
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Obesity, often regarded as an incipiently factor for metabolic
syndrome, supports insulin resistance, dyslipidemia and hypertension. The combination of these metabolic complaints is clinically
interpreted as an indicator for increased risk of cardiovascular diseases and diabetes mellitus type 2 [1,2].
Especially postmenopausal women tend to have a higher risk
in developing obesity and thus metabolic syndrome because estrogens play an important role in the control of energy homeostasis in
∗ Corresponding author at: German Sport University Cologne, Institute of Cardiovascular Research and Sports Medicine, Am Sportpark Müngersdorf 6, 50933
Cologne, Germany. Tel.: +49 221 49825860; fax: +49 221 49828370.
E-mail address: [email protected] (N. Zoth).
0960-0760/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jsbmb.2012.01.004
females [3,4]. It is well established that estrogen deficiency seems
to have an unfavorable impact on mobilization of fatty acids, body
fat distribution and glucose-absorbing capacity of different tissues
[5–7]. Total body fat mass in postmenopausal women, for example, is around 28% higher compared to premenopausal women
[6,8]. Similarly, ovariectomy in rats, and accordingly deprivation of
estrogens, leads to increases in body weight and fat mass whereas
estrogen treatment antagonizes these effects in a positive manner [9–11]. Ovariectomy has also been reported to enhance overall
insulin resistance in rats, while hormone replacement therapy
(HRT) ameliorates these effects [12]. On the one hand, the use of
HRT for treating estrogen deficiency symptoms has been demonstrated to be beneficial but on the other hand, there are reports
demonstrating that HRT increases the risk of developing or promoting breast cancer [13,14]. Therefore, there is an ongoing discussion
about alternative strategies for the treatment of menopausal and
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
postmenopausal disorders. In this context, it is still ambiguous, if
targeted exercise training might be a suitable alternative for HRT,
especially in the treatment of metabolic syndrome.
Physical training in animals decreases fat deposition and
enhances insulin sensitivity [15,16]. According to Latour et al. [17]
exercise training improves glucose stimulated insulin response and
activates glucose transporter concentration in skeletal muscle in
ovariectomized (OVX) rats as well.
A number of studies have shown that exercise training or
estrogen treatment seems to have the potency to prevent the
development of metabolic syndrome [11,16,18]. However there are
limited data on such effects in the therapy of already existing obesity. Therefore, the aim of this study was to determine the combined
effects of estrogen substitution and exercise training on the therapy of obesity and on obesity-related comorbidities in the animal
model of OVX Wistar rats.
OVX female Wistar rats were fed with a high fat diet (HF)
for ten months. After developing obesity the animals remained
sedentary, performed treadmill training, received 17␤-Estradiol
(E2 ), or underwent both exercise training and estrogen treatment. Sham-operated (SHAM) animals served as controls. Body
weight, subcutaneous and visceral body fat, insulin sensitivity, leptin, insulin, serum lipids (triglyceride, cholesterol, high-density
lipoprotein (HDL) and low-density lipoprotein (LDL)) as well as
energy expenditure were investigated. In addition, PPAR␣ mRNA
expression in the liver and GLUT4 expression in the m. gastrocnemius have been determined.
2. Materials and methods
2.1. Animals
Thirty-one female Wistar rats (Janvier, Le Genest St Isle,
France) were maintained under controlled conditions of illumination (12/12-h day/night cycle) and constant room temperature
(20 ± 1 ◦ C, relative humidity 50–80%). Weighing 70–90 g on arrival,
they were housed in groups of five and had free access to food
(SSniff® GmbH, Soest, Germany) and water. All animal experiments
were in accordance with the guidelines of the Committee on Animal
Care and complied with accepted veterinary medical practice.
2.2. Animal treatment and tissue preparation
The experimental design of the present study is presented in
Fig. 1. Throughout the experiment all animals were fed with a
HF diet consisting of pellet rat chow (EF R/M acc. D12451 (I)
mod.*, SSniff® GmbH, Soest, Germany). The gross energy averaged
22.6 MJ/kg and the metabolizable energy 19.2 MJ/kg. Details of the
diet, especially the fatty acids, are presented in Table 1.
Four weeks after delivery estrous synchronization as well as
ovariectomy was performed, rats on the same cycle were shamoperated (n = 10), the remaining were ovariectomized (n = 21).
During eight months of overnutrition metabolic changes were
documented intermittently. Weight, blood glucose and energy
expenditure were measured in equal periods. After developing
obesity, all rats were randomly assigned to one of the experimental groups for six weeks. A subset of the OVX group (n = 10)
was substituted with E2 (4 ␮g kg−1 day−1 ). The 17␤-Estradiol
(estra-1,3,5(10)-trien-3,16␣,17␤-diol) provided by Sigma–Aldrich
(Deisenhofen, Germany) was administered via ALZET® osmotic
mini pumps. Beyond that all experimental groups were classified
into a sedentary and an exercise training group. Before and after the
treatment period blood samples were collected and intraperitoneal
glucose tolerance tests (ipGTT) were carried out.
65
Table 1
Composition of the high fat diet (HF).
Ingredients
Dry matter (%)
Crude fat (%)
Fatty acids (%)
C12:0
C14:0
C16:0
C16:l
C18:0
C18:2
C18:3
C20:0
C20:4
Cholesterol (mg/kg)
Crude protein (%)
Crude fibre (%)
Crude ash (%)
Nitrogen-free extracts (%)
Starch (%)
Sugar/dextrines (%)
Amount
HF diet
96.2
23.1
0.02
0.29
5.15
0.62
2.83
8.99
3.19
0.37
0.01
175
22.5
4
5.9
40.7
6.7
31.7
At the end of the experiment all animals were decapitated
after light anesthesia with CO2 inhalation and complete blood
was collected. Whole abdominal body fat (periovarian, perirenal,
mesenteric and omental), liver, m. gastrocnemius and uteri were
prepared free of fat and wet weights were determined. Periovarian
fat pads were immediately fixed in 10% neutrally buffered formalin
for morphometric analysis. To measure subcutaneous fat content, a
piece of 4 cm2 from the back of each animal was dissected and wet
weight was determined. The m. gastrocnemius and the livers were
directly frozen in liquid nitrogen for analysis of molecular markers.
2.3. Open-circuit calorimeter “Oxymax”
The energy expenditure, in terms of oxygen consumption and
carbon dioxide production, was measured using open-circuit indirect calorimeter “Oxymax” (Colinst, Columbus Ohio). During the
test the percent O2 and CO2 gas levels of the test chamber environment were quantified periodically. Hence, O2 consumption (VO2 ),
CO2 production (VCO2 ) and respiratory exchange ratio (RER) could
be calculated. RER resulted from the ratio between carbon dioxide
production and oxygen consumption (VCO2 /VO2 ). In the present
study energy expenditure at rest was recorded isochronously (data
not shown). In addition, before and after the intervention energy
expenditure during exercise training was measured.
2.4. Intraperitoneal glucose tolerance test (ipGTT)
Eight months after HF diet, immediately before and after treatment period, all animals were submitted to an ipGTT. After a 12-h
fasting period tail blood was taken before glucose application by
intraperitoneal injection (2 g/kg bwt.) and 15, 30, 60 and 90 min
afterwards. Blood samples (1.5 ␮l) were analyzed proximately via
ACCU CHECK® Compact plus (Roche diagnostics GmbH, Mannheim,
Germany).
2.5. Exercise training protocol
The exercise training, in terms of endurance training, was in line
with the protocol of previous experiments [11]. It took place on a
motor-driven rodent treadmill (Columbus Instruments Ohio, USA)
five days/week over a six-week period. The intensity progressively
increased from 10 min once a day at 22 m/min, 5% upgrade, up to
15 min twice a day at 25 m/min, 5% upgrade, after the first week of
practicing.
66
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
Fig. 1. Experimental design of the study. SHAM = sham-operated; OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no ex = rest; T0 = before treatment
period; T1 = after treatment period.
2.6. Determination of serum lipids
Blood samples were collected at the day of decapitation. All
samples were centrifuged and serum was stored at −20 ◦ C. Serum
levels of cholesterol, high-density lipoprotein cholesterol (HDL)
and low-density lipoprotein cholesterol (LDL) were determined
via photometric systems using DIALAB (Wiener Neudorf, Austria).
Serum levels of triglyceride were analyzed by colorimetry using
ABX Pentra (ABX Diagnostics Montpellier, France).
2.7. Determination of leptin and insulin
Leptin levels were determined by a commercially available
Enzyme-Immunoassay (Alpco Diagnostics, Salem NH) which utilizes two specific polyclonal antibodies for mouse and rat leptin.
The intra-assay variation coefficient was 5.6%. Serum levels of
insulin were assessed with an Enzyme-Immunoassay (Alpco diagnostics, Salem, NH) in which mouse monoclonal antibodies for
insulin were used. The intra-assay variation coefficient was 4.8%.
2.8. Histological analysis
normalization to an endogenous reference gene (Cytochrom c
Ocidase, subunit 1A) following the CT method. Based on
genomic sequences available at the UCSC Genome Bioinformatics
database the following specific primer pairs were designed by
Primer3 (Whitehead Institute for Biomedical Research, Cambridge, MA, USA; http://www-genome.wi.mit.edu/cgi-bin/primer/
primer3 www.cgi/) and confirmed by sequences in the NCBI
database (http://www.ncbi.nlm.nih.gov/). All the following
primers were synthesized by InvitrogenTM (Germany): 1A fwd
5 -CGTCACAGCCCATGCATTCG-3 , 1A rev 5 -CTGTTCATCCTGTTCCAGCTC-3 , PPAR␣ fwd 5 -GGCTGCTATAATTTGCTGTGG-3 , PPAR␣
rev 5 -TTCTTGATGACCTGCACGAG-3 , Glut4 fwd 5 -TATTTGGCTTTGTGGCCTTC-3 , Glut4 rev 5 -CGGCAAATAGAAGGAAGACG-3 .
2.10. Statistical analysis
All available data are expressed as arithmetic means ± S.E.M.
To establish statistical significance, a two-way analysis of variance
was applied followed by pair-wise comparison of selected means
using Mann–Whitney-U-test. Regarding glucose tolerance, differences between the groups were calculated using one-way ANOVA
with a following Tukey’s HSD test (Fig. 5b–d) or Student’s t-test
(Fig. 5a). p ≤ 0.05 indicated statistical significance. All analyses were
statistically evaluated by SPSS 19.0 (SPSS, Chicago).
To determine size of adipocytes sections (7 ␮m) were cut from
paraffin embedded periovarian fat tissue and were stained with
hematoxylin and eosin to visualize adipocytes’ morphometry. The
cells were examined using Zeiss KS 300 3.0 Image software (Carl
Zeiss Jena, Germany).
3. Results
2.9. mRNA preperation and realtime PCR
3.1. Body weight
Total RNA from the frozen m. gastrocnemius and the liver
were extracted using the standard TrIzol method (Invitrogen).
RNA was quantified by spectrophotometry (NanoDropTM 1000,
Thermo Scientific, Wilmington, DE 19810, USA) and cDNA was
synthesized from 1 ␮g RNA using a Reverse Transcription System
(QuantiTect ®, Qiagen). Real-time PCR was performed with Taq
DNA polymerase (Invitrogen, Germany) in the presence of a
fluorescent dye (SYBR Green, BioRad) on an Mx3005PTM qPCR
System (Stratagene). All reactions were run in triplicate in a
50 ␮l total volume. The PCR program was as follows: 95 ◦ C for
3 min for 1 cycle, followed by 35 cycles of 30 s at 95 ◦ C, 30 s at
58 ◦ C, 30 s at 72 ◦ C, and 1 cycle at 72 ◦ C for 1 min. Fluorescence
was quantified during the 58 ◦ C annealing step and product
formation was confirmed by melting curve analysis (58–95 ◦ C).
Relative mRNA amounts of target genes were calculated after
In Table 2 it is clearly visible that the uterine wet weights in the
respective groups demonstrate the efficiency of ovariectomy and
hormone treatment. OVX rats showed significant lowered uterine
wet weights compared to SHAM rats whereas treatment with E2
led to a strong stimulation of uterine wet weights.
As expected, at the end of the experiment sedentary OVX rats
had the highest body weights, which were about 24% (p ≤ 0.05)
higher than those of SHAM rats (Table 2). Six weeks of exercise
training, estrogen substitution as well as a combined treatment
resulted in a significant reduction of body weights (Fig. 2A). In
Fig. 2B–D individual developments of body weights between start
(T0) and end of intervention (T1) are shown. It is visible that in
SHAM and OVX animals exercise training resulted in a decrease of
body weight, whereas in untrained animals body weight further
increased significantly (Fig. 2B and C). Fig. 2D demonstrates that E2
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
67
Table 2
Body and tissue weights at the day of section.
Variable
Body weight (g)
SHAM ex
SHAM no ex
OVX ex
OVX no ex
OVX + E2 ex
OVX + E2 no ex
400.5
386.2
410.8
478.4
439.0
413.2
±
±
±
±
±
±
Uterus wet weight (mg/kg b.wt.)
27.1*
32.9*
42.2*
53.6
16.8
25.1*
2095.8
1866.9
148.3
155.2
858.6
888.6
±
±
±
±
±
±
Subcutan fat (g/kg b.wt.)
1178.4*
702.0*
30.4
41.3
111.4*
111.6*
6.8
8.4
12.0
14.3
8.0
9.1
±
±
±
±
±
±
1.8*
2.4*
2.9
3.6
0.8*
2.3*
Visceral fat (g/kg b.wt.)
84.3
102.4
83.3
124.7
89.8
94.8
±
±
±
±
±
±
19.0*
16.9*
13.5*
12.1
11.2*
14.2*
SHAM = sham-operated; OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no ex = rest, data shown are means ± S.E.M. Mean values were significantly
different for the following comparisons: * vs. OVX no ex (p ≤ 0.05).
treatment also resulted in a decrease of body weight. Combination
of E2 treatment and training seems to have an additive effect.
3.2. Serum lipid levels
(A) body weight process T0-T1
*
*
600
Compared to SHAM rats, estrogen deprivation by ovariectomy
resulted in a significant increase in visceral and subcutaneous body
fat (Table 2 and Fig. 3A). Intervention either by training or E2 treatment resulted in a significant decrease of subcutaneous and visceral
body fat in OVX animals (Table 2 and Fig. 3A). Particularly, at the
end of intervention period (T1) OVX rats revealed subcutaneous
fat contents, which were about 70% higher than those of SHAM
rats (Table 2). This rise of subcutaneous fat could be significantly
antagonized by estrogen substitution (36%, p ≤ 0.05) as well as a
combination with exercise training (44%, p ≤ 0.05). Accordingly,
sedentary OVX rats possessed the largest adipocytes (p ≤ 0.05),
which could also be minimized through training or combined treatment (n.s.) (Fig. 3B).
In good agreement to the effects observed for visceral fat,
in obese OVX animals serum leptin levels were reduced by E2
(B) SHAM animals
*
*
27
500
400
300
200
T0
100
T1
0
(C) OVX animals
body weight process (%)
body weight process (g)
Serum levels of cholesterol, HDL, LDL and triglycerides at the
end of the intervention (T1) are presented in Table 3.
Sedentary OVX rats revealed the highest serum levels of
cholesterol, triglyceride and LDL. Their values were about 33%
(cholesterol), 56% (LDL) and even 71% (triglyceride) higher than
those of the SHAM rats (each p ≤ 0.05). These OVX induced increases
could be antagonized by exercise training or by E2 treatment.
Especially, serum levels of triglycerides could be suppressed by
55% through exercise training (p ≤ 0.05). In addition, the strongest
reduction of cholesterol and LDL was observed in OVX rats with
combined intervention (each p ≤ 0.05).
3.3. Body fat and leptin
SHAM1 ex
22
SHAM2 ex
17
SHAM3 ex
12
SHAM4 ex
7
SHAM5 ex
2
-3
SHAM1 no ex
T0
T1
SHAM2 no ex
-8
SHAM3 no ex
-13
SHAM4 no ex
-18
SHAM5 no ex
(D) OVX+E2 animals
OVX1 ex
22
OVX2 ex
OVX3 ex
17
OVX4 ex
12
OVX5 ex
7
OVX1 no ex
OVX2 no ex
2
-3
-8
-13
-18
T0
T1
OVX3 no ex
OVX4 no ex
OVX5 no ex
OVX6 no ex
body weight process (%)
body weight process (%)
0
27
-2
-4
-6
-8
-10
-12
T0
T1
OVX+E2 ex
OVX+E2 ex
OVX+E2 ex
OVX+E2 ex
OVX+E2 ex
OVX+E2 no ex
OVX+E2 no ex
OVX+E2 no ex
-14
OVX+E2 no ex
-16
OVX+E2 no ex
-18
Fig. 2. Body weight process of each single animal (T0–T1). SHAM = sham-operated; OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no ex = rest;
T0 = before treatment period; T1 = after treatment period, data shown are means ± S.E.M. Mean values were significantly different for the following comparisons: (A) *
marks significant differences as indicated by brackets (p ≤ 0.05).
68
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
Table 3
Serum lipid levels at the day of section.
Variable
Cholesterol (mg/dl)
SHAM ex
SHAM no ex
OVX ex
OVX no ex
OVX + E2 ex
OVX + E2 no ex
80.4
80.5
90.2
107.0
79.6
86.0
±
±
±
±
±
±
12.6*
12.1*
4.8
17.0
19.7*
9.6*
HDL (mg/dl)
60.2
63.0
62.0
75.2
56.8
62.8
±
±
±
±
±
±
7.0
11.9
3.1
13.1
12.8*
5.7
LDL (mg/dl)
12.0
15.2
23.1
23.7
15.4
17.9
±
±
±
±
±
±
4.9*
3.5*
3.8
3.9
6.3*
3.1*
Triglyceride (mg/dl)
91.2
77.25
59.6
131.8
73.4
76.5
±
±
±
±
±
±
22.8
29.1*
13.4*
35.2
35.0*
37.1*
SHAM = sham-operated; OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no ex = rest, data shown are means ± S.E.M. Mean values were significantly
different for the following comparisons: * vs. OVX no ex (p ≤ 0.05).
treatment or training (each p ≤ 0.05, Fig. 3C). The highest reduction
of serum leptin was observed in OVX rats with combined treatment
(p ≤ 0.05).
training or E2 treatment resulted in a reduced RER during exercise
training (n.s.) (Fig. 4B).
3.5. Insulin levels and glucose tolerance
3.4. Energy expenditure
Energy expenditure during exercise training, in terms of glucose
or fat oxidation, can be measured via indirect calorimetry. From RER
(VCO2 /VO2 ) can be gathered to energy consumption as well as proportional energy supply (carbohydrate, fat or protein). RER usually
ranges from 0.7 (pure fat oxidation) up to 1.0 (pure carbohydrate
oxidation) [19]. Compared to SHAM rats, OVX induced estrogen
deprivation resulted in a significant increase of RER during exercise training (Fig. 4A). In SHAM as well as OVX rats intervention by
No significant differences in serum insulin levels among the
treatment groups could be observed (data not shown).
To analyze changes in glucose sensitivity, induced by intervention, an ipGTT (Fig. 5) was performed right before (T0) and after the
treatment period (T1). After eight months of overnutrition SHAM
animals showed still a significant higher glucose tolerance compared to OVX animals (Fig. 5A). In SHAM rats exercise training
resulted in an improvement of glucose tolerance (Fig. 5B). In OVX
animals exercise training (Fig. 5C) as well as E2 treatment (Fig. 5D)
led to a significant improvement of glucose tolerance.
Fig. 3. Visceral body fat content, adipocyte size and serum levels of leptin. SHAM = sham-operated; OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no
ex = rest, data shown are means ± S.E.M. Mean values were significantly different for the following comparisons: (A–C) * vs. OVX no ex (p ≤ 0.05).
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
RER
(A) RER SHAM vs. OVX T0
0,88
0,87
0,86
0,85
0,84
0,83
0,82
0,81
0,8
0,79
0,78
0,77
0,76
*
SHAM
OVX
RER
(B) RER T0 –T1
0,88
0,87
0,86
0,85
0,84
0,83
0,82
0,81
0,8
0,79
0,78
0,77
0,76
SHAM
OVX
OVX+E2
T0
T1
Fig. 4. Energy expenditure during exercise training before and after treatment period (T0–T1). (A) SHAM vs. OVX (T0), (B) subdivided into treatment groups (T0–T1). RER = respiratory exchange rate; SHAM = sham-operated;
OVX = ovariectomized; E2 = treated with 17␤-Estradiol; T0 = before treatment
period; T1 = after treatment period, data shown are means ± S.E.M. Mean values
were significantly different for following comparisons: (A) * vs. SHAM (p ≤ 0.05).
3.6. Gene expression in liver and m. gastrocnemius
To confirm the present findings on a molecular level mRNA
expression of PPAR␣ in the liver as a marker for fatty acid oxidation and GLUT4 expression in the m. gastrocnemius as a marker
for effects in the insulin signaling pathway have been measured
[20,21]. As shown in Fig. 6A, exercise training significantly reduced
GLUT4 expression in m. gastrocnemius, whereas E2 treatment had
no effect. PPAR␣ expression in the liver remained unaffected in all
groups (Fig. 6B).
4. Discussion
The role of estrogen deficiency in the development of obesity
after menopause is well known [22,23]. Results of our previous
studies support that physical activity as well as estrogen substitution has a considerable impact on the prevention of obesity
in ovariectomized rats. There are several reports describing a
stimulation of lipid metabolism by estrogens [11,24]. However,
mechanistically data about the effects of E2 , training and combinations in the therapy of an already established obesity are limited.
Therefore, in the present study we have investigated combined
effects of E2 and training on obese SHAM and OVX female rats. Longterm feeding (eight months) of OVX rats with a high fat diet resulted
in manifestation of obesity and obesity-related risk factors, such as
increased body weights, unfavorable changes of serum lipoprotein
profile, decreased fat oxidation and impaired glucose tolerance. An
69
only six weeks lasting intervention of combined estrogen substitution and exercise training ameliorated these effects partially.
In Table 2 it is clearly visible that at the end of the intervention period sedentary OVX rats showed significant higher body
weights compared to SHAM and E2 treated rats. Due to the fact
that no differences regarding food intake were recognized body
weight gain of OVX rats could not be caused by hyperphagia. Consistent with previous studies intervention by estrogen treatment
but also training resulted in a decrease of body weight. Especially
the induction of weight loss after estrogen treatment is in agreement to a variety of previous investigations [17,25,26]. An effect
of training and E2 treatment on weight loss is clearly visible comparing body weights of individual animals between start (T0) and
end (T1) of the intervention (Fig. 2). There is a tendency that in
the treatment of already established obesity a combination of both
interventions acts in an additive manner. This fits to our recently
published data demonstrating that the combination of estrogen
treatment and exercise training also acts additive in preventing
the development of nutritional induced obesity [11]. However, in
our therapy model the combinatory effect is not as obvious as in
the prevention model. A possible explanation could be that in this
short period of intervention time (six weeks) the single interventions already resulted in a maximum effect. With respect to this
it has to be considered that E2 and training resulted in an average weight loss of about 10% in six weeks by unchanged nutrition.
Consequently the reduction of body weight by E2 treatment allows
assuming that there must be metabolic changes under control of
estradiol. Wade [27] and Gray and Wade [28] claim the increase
of body fat content for the weight gain in OVX rodents. Further
studies on mice deficient in estrogen receptor ␣ also exhibited significant increases in adipose tissue [29]. Our data are in line with
those results. In our study OVX rats had the highest body fat contents as well as the largest adipocytes (Table 2 and Fig. 3A and B).
Subcutaneous and visceral body fat content (Table 2) as well as
adipocyte size (Fig. 3B) have been reduced by E2 treatment and
exercise training. These data confirm the assumption that estrogens play an important regulatory role in adipocyte metabolism.
Furthermore, exercise training suppressed visceral body fat of OVX
rats by one third, which implies that OVX induced fat gain could
completely be prevented by exercise training. These results are in
agreement with literature [9,15,25]. A similar study on HF diet fed
Sprague-Dawley rats provided preventive effects even of voluntary wheel running on the characteristics of adipose tissue [18].
Body fat gain due to estrogen deficiency and in succession obesity
seems to be associated with decreased lipolysis as well as fat oxidation [30]. Presumably exercise training stimulates lipolysis via an
increase in sympathetic activity and consequently exerts influence
on body fat content [31,32]. This proposition can be supported by
the present study because the measurement of RER during exercise training showed increased values in the OVX group, whereas
estrogen substitution suppressed them. Considering that a RER
around 0.7 indicates fat oxidation a decline can be interpreted as
an increase of energy supply through fat oxidation. After six weeks
of training all treatment groups exhibited lower RER values compared to the beginning. Therefore we assume that an adaptation to
the training had already occurred even if it was not significant. A
longer intervention period may result in more pronounced effects.
Moreover, data in the literature indicate that visceral body
fat strongly correlates with serum leptin levels [27,33,34]. Leptin, mainly expressed in adipose tissue, is characterized as an
adipocytokine [35]. It plays a relevant role in regulating appetite or
energy expenditure and is able to resist insulin secretion [34,36].
As expected, elevated visceral body fat contents in OVX rats were
found to be associated with strong increases in serum leptin levels (Fig. 3B). In the present study E2 substitution decreases leptin
concentrations of sedentary OVX rats slightly but significantly. Our
70
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
(A) ipGTT SHAM vs OVX T0
(B) ipGTT SHAM ex vs SHAM no ex (T0-T1)
*
500
*
400
300
*
OVX
SHAM
200
550
500
450
400
350
300
250
200
150
100
50
glucose (mg/dl)
glucose (mg/dl)
600
100
0
0
15
30
60
120
T0 SHAM ex
T0 SHAM no ex
T1 SHAM ex
T1 SHAM no ex
0
time after glucose application
*
*
T0 OVX ex
T0 OVX no ex
T1 OVX ex
T1 OVX no ex
0
15
30
60
30
60
120
(D) ipGTT OVX+E2 ex vs OVX+E2 no ex (T0-T1)
120
time after glucose application
glucose (mg/dl)
glucose (mg/dl)
(C) ipGTT OVX ex vs OVX no ex (T0-T1)
550
500
450
400
350
300
250
200
150
100
50
15
time after glucose application
*
550
500
450
400
350
300
250
200
150
100
50
T0 OVX+E2 ex
T0 OVX+E2 no ex
T1 OVX+E2 ex
T1 OVX+E2 no ex
0
15
30
60
120
time after glucose application
Fig. 5. ipGTT before and after treatment period, B and C subdivided into treatment groups. ipGTT = intraperitoneal glucose tolerance test; SHAM = sham-operated;
OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no ex = rest; T0 = before treatment period; T1 = after treatment period, data shown are means ± S.E.M.
Mean values were significantly different for the following comparisons: (A) * vs. SHAM at the same time point (p ≤ 0.05), (B) no significance, (C) * vs. T1OVX ex at the same
time point (p ≤ 0.05), (D) * vs. T1OVX + E2 ex at the same time point (p ≤ 0.05).
data do not confirm the results of Pelleymounter et al. [37] who
did not find a significant influence of estradiol treatment on leptin levels. Furthermore, serum leptin levels did also not differ in
pre- and postmenopausal women and were not affected by HRT
[38]. Interestingly, serum leptin levels in OVX rats could be lowered by exercise training but most effectively in combination with
E2 substitution (Fig. 3C).
Hyperlipidemia, known to be an obesity-related risk factor
for cardiovascular diseases, has also been investigated in this
study. In line with reports in literature [16,25,39] the removal of
estrogens by ovariectomy also led to significant increases in serum
levels of cholesterol, LDL and triglycerides (Table 3). This increase
could be antagonized by E2 treatment alone and in combination
with exercise training. Moreover, exercise training alone lowered
triglycerides of OVX rodents even by 55%. This observation is in
disagreement to published data from a human intervention study
demonstrating that exercise training did not affect serum lipid levels [40]. However, other studies support our findings and describe
Fig. 6. mRNA expression of (A) GLUT4 (m. gastrocnemius) and (B) PPAR␣ (liver). GLUT4 = glucose transporter 4; PPAR␣ = peroxisome proliferator-activated receptor ␣;
SHAM = sham-operated; OVX = ovariectomized; E2 = treated with 17␤-Estradiol; ex = exercise; no ex = rest; T0 = before treatment period; T1 = after treatment period, data
shown are means ± S.E.M. Mean values were significantly different for the following comparisons: (A) * vs. SHAM no ex (p ≤ 0.05); + vs. OVX no ex (p ≤ 0.05); # vs. OVX E2 no
ex (p ≤ 0.05), (B) no significance.
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
beneficial effects of physical activity on blood lipids [41]. In general,
findings regarding the effects of E2 treatment and exercise training
seem to be enormous heterogeneous [16,25,42]. This might be due
to the experimental design and aspects such as diet, the subject’s
metabolic conditions and the duration of the studies. However,
based on observations of our study we believe that the combination of E2 treatment and exercise training might be a promising
strategy in the prevention of arteriosclerosis in postmenopausal
women.
According to Woods et al. [43] HF diet itself leads to obesity
as well as obesity-related risk factors such as restricted insulin
sensitivity. In our study serum insulin levels were not affected by
any intervention and in a similar range in all experimental groups.
Therefore, differences in glucose tolerance will be interpreted as
changes in insulin sensitivity of the respective groups. The present
data of GTT have shown that the absence of estrogens resulted in a
decreased insulin sensitivity (Fig. 5A SHAM vs. OVX), which could
be increased by E2 treatment (Fig. 5D). These observations are in
line with findings that excess visceral fat accumulation due to estrogen deficiency is associated with insulin resistance [18,44,45]. In
our study exercise training had also a significant impact on insulin
sensitivity. In SHAM animals (Fig. 5B) as well as in OVX animals
(Fig. 5C) kept on a high fat diet for eight months only six weeks of
exercise training improved insulin sensitivity. These results prove
a study also demonstrating enhanced insulin sensitivity after exercise training or E2 substitution [18] whereas other studies disprove
our results [17,46]. Although E2 as well as exercise training seems
to result in increased insulin sensitivity our data clearly show that
a combination of both does not result in additive effects (Fig. 5D).
Again, like discussed with respect to weight loss, we believe that in
the short period of time the single interventions already resulted
in maximum effects.
Of course it is of great interest to identify the underlying molecular mechanisms of our observations. Getting first insights into these
mechanisms we have studied PPAR␣ expression in the liver as a
marker for fatty acid (FA) oxidation (Fig. 6B) and GLUT4 expression in the m. gastrocnemius (Fig. 6A) as a marker for effects in the
insulin signaling pathway [20,21]. The major physiological function of PPAR␣ is to promote FA utilization [47]. In the present
study none of our interventions resulted in a modulation of PPAR␣
expression in the liver, which could be interpreted as FA oxidation, at least in this tissue, is not modulated by E2 and exercise
training. In contrast expression of GLUT4 in m. gastrocnemius was
strongly affected by exercise training but not by E2 treatment. Thus,
we conclude that insulin signaling is mainly affected by training.
We speculate that the insulin sensitivity induced by training has
been increased due to a combined effect of changes in muscle mass
and insulin signaling. This assumption needs to be confirmed by
future investigations including a variety of molecular markers in
different tissues such as pAkt, IRS1/2, Pi-3-kinase, PPAR␥ and many
more.
In conclusion, the present study indicates that the combination
of estrogen treatment and exercise training is an effective strategy to reduce body weight, body fat content, blood serum lipids
and results in an improvement of energy expenditure as well as
an increase of insulin sensitivity in OVX rats. Therefore, we believe
that our data provide evidence that HRT combined with targeted
exercise training might be a very effective strategy to support the
therapy of diet-induced obesity. Prospectively, the dose of estrogen used in HRT could be lowered if treatment is combined with
physical activity.
Acknowledgement
This project was in part supported by the DFG Grant DI 716/10-3.
71
References
[1] R.A. De Fronzo, E. Ferrannini, Insulin resistance. A multifaceted syndrome
responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease, Diabetes Care 14 (1991) 173–194.
[2] P.W.F. Wilson, W.B. Kannel, H. Silbershatz, R.B. D’Agostino, Clustering of
metabolic factors and coronary heart disease, Arch. Intern. Med. 159 (1999)
1104–1109.
[3] Y.W. Park, S. Zhu, L. Palaniappan, S. Heshka, M.R. Carnethon, S.B. Heymsfield,
The metabolic syndrome. Prevalence and associated risk factor findings in the
US population from the third national health and nutrition examination survey,
1988–1994, Arch. Intern. Med. 163 (2003) 427–436.
[4] E. Simpson, M. Jones, M. Kylie, R. Hill, L. Maffei, C. Carani, W.C. Boon, Estrogen,
a fundamental player in energy homeostasis, J. Steroid Biochem. Mol. Biol. 95
(1–5) (2005) 3–8.
[5] F. Tremollieres, J. Pouilles, C. Ribot, Relative influence of age and menopause on
total and regional body composition changes in postmenopausal women, Am.
J. Obest. Gynecol. 175 (1996) 1594–1600.
[6] M. Toth, A. Tchernof, C. Sites, E. Poehlman, Menopause-related changes in body
fat distribution, Ann. N.Y. Acad. Sci. 904 (2000) 502–506.
[7] M. Jones, K.J. Mc Innes, W.C. Boon, E. Simpson, Estrogen and adiposity – utilizing
models of aromatase deficiency to explore the relationship, J. Steroid Biochem.
Mol. Biol. 106 (1–5) (2007) 3–7.
[8] C.J. Ley, B. Lees, J.C Stevenson, Sex- and menopause-associated changes in bodyfat distribution, Am. J. Clin. Nutr. 55 (1992) 950–954.
[9] D. Richard, L. Rochon, Y. Deshaies, Effects of exercise training on energy balance
of ovariectomized rats, Am. J. Physiol. Regul. Integr. Comp. Physiol. 253 (1987)
R740–R745.
[10] T. Hertrampf, M.J. Gruca, J. Seibel, U. Laudenbach, K.H. Fritzemeier, P. Diel, The
bone-protective effect of the phytoestrogen genistein is mediated via ER alphadependent mechanisms and strongly enhanced by physical activity, Bone 40
(6) (2007) 1529–1535.
[11] N. Zoth, C. Weigt, U. Laudenbach-Lechowski, P. Diel, Physical activity and estrogen treatment reduce visceral body fat and serum levels of leptin in an additive
manner in a diet induced animal model of obesity, J. Steroid Biochem. Mol. Biol.
122 (2010) 100–105.
[12] C.J. Bailey, H. Ahmed-Sorour, Role of ovarian hormones in the long-term control
of glucose homeostasis. Effects on insulin secretion, Diabetologia 19 (1980)
475–481.
[13] V. Beral, Breast cancer and hormone-replacement therapy in the Million
Women Study, Lancet 362 (2003) 419–427.
[14] W.Y. Chen, Exogenous and endogenous hormones and breast cancer, Best Pract.
Res. Clin. Endocrinol. Metab. 22 (2008) 573–585.
[15] L. Hao, Y. Wang, Y. Duan, S. Bu, Effects of treadmill exercise training on liver
fat accumulation and estrogen receptor alpha expression in intact and ovariectomized rats with or without estrogen replacement treatment, Eur. J. Appl.
Physiol. 109 (2010) 879–886.
[16] V. Saengsirisuwan, S. Pongseeda, M. Prasannarong, K. Vichsiwong, C.
Toskulkao, Modulation of insulin resistance in ovariectomized rats by
endurance exercise training and estrogen replacement, Metabolism 58 (2008)
38–47.
[17] M.G. Latour, M. Shinoda, J.M. Lavoie, Metabolic effects of physical training in ovariectomized and hyperestrogenic rats, J. Appl. Physiol. 90 (2001)
235–241.
[18] K.S. Gollisch, J. Brandauer, N. Jessen, T. Toyoda, A.N. Ayer, M.F. Hirshman, L.J.
Goodvear, Effects of exercise training on subcutaneous and visceral adipose
tissue in normal- and high-fat-diet-fed rats, Am. J. Physiol. Endocrinol. Metab.
297 (2009) E495–E504.
[19] A.E. Jeukendrup, G.A. Wallis, Measurement of substrate oxidation during exercise by means of gas exchange measurements, Int. J. Sports Med. 26 (2005)
S28–S37.
[20] J.D. Tugwood, I. Issemann, R.G. Anderson, K.R. Bundell, W.L. McPheat, S. Green,
The mouse peroxisome proliferator activated receptor recognizes a response
element in the 5 flanking sequence of the rat acyl CoA oxidase gene, EMBO J.
11 (1992) 433–439.
[21] R.T. Watson, M. Kanzaki, J.E. Pessin, Regulated membrane trafficking of the
insulin-responsive glucose transporter4 in adipocytes, Endocr. Rev. 25 (2)
(2004) 177–204.
[22] J.S. Green, P.R. Stanforth, T. Rankien, A.S. Leon, Dc D. Rao, J.S. Shinners,
The effects of exercise training on abdominal visceral fat, body composition, and indicators of the metabolic syndrome in postmenopausal women
with and without estrogen replacement therapy: the HERITAGE family study,
Metabolism 53 (2004) 1192–1196.
[23] C.L. Khoo, M. Perera, Diabetes and the menopause, J. Br. Menopause Soc. 11
(2005) 6–11.
[24] E.S. Campbell, M.A. Febbraio, Effects of ovarian hormones on exercise
metabolism, Curr. Opin. Clin. Nutr. Metab. Care 4 (2001) 515–520.
[25] M. Shinoda, M.G. Latour, J.M. Lavoie, Effects of physical training on body composition and organ weights in ovariectomized and hyperestrogenic rats, Int. J.
Obes. 26 (2002) 335–343.
[26] G. Bryzgalova, L. Lundholm, N. Portwood, J.A. Gustafsson, A. Khan, S. Efendic, K.
Dahlman-Wright, Mechanisms of antidiabetogenic and body weight-lowering
effects of estrogen in high-fat diet-fed mice, Am. J. Physiol. Endocrinol. Metab.
295 (2008) E904–E912.
[27] G.N. Wade, Regulation of body fat content? Am. J. Physiol. Regul. Integr. Comp.
Physiol. 286 (2004) R14–R15.
72
N. Zoth et al. / Journal of Steroid Biochemistry & Molecular Biology 130 (2012) 64–72
[28] J.M. Gray, G.N. Wade, Food intake, body weight, and adiposity in female rats:
actions and interactions of progestins and antiestrogens, Am. J. Physiol. 240
(1981) E474–E481 (Endocrinol. Metab.).
[29] P.A. Heine, J.A. Taylor, G.A. Iwamoto, D.B. Lubahn, P.S. Cooke, Increased adipose
tissue in male and female estrogen receptor-␣ knockout mice, Proc. Nat. Acad.
Sci. U.S.A. 97 (2000) 12729–12734.
[30] J.F. Horowitz, S. Klein, Whole body and abdominal lipolytic sensitivity to
epinephrine is suppressed in upper body obese women, Am. J. Physiol. 278
(2000) E1144–E1152.
[31] J.A. Romijn, S. Klein, E.F. Coyle, L.S. Sidossis, R.R. Wolfe, Strenuous endurance
training increases lipolysis and triglyceride fatty acid cycling at rest, J. Appl.
Physiol. 75 (1993) 108–113.
[32] J. Bülow, Physical activity and adipose tissue metabolism, Scand. J. Med. Sci.
Sports 14 (2004) 72–73.
[33] R.C. Frederich, A. Hamann, S. Anderson, B. Lollmann, B.B. Lowell, J.S. Flier, Leptin
levels reflect body lipid content in mice: evidence for diet-induced resistance
to leptin action, Nat. Med. 1 (1995) 1311–1314.
[34] K.E.L Hashimi, D.D. Pierroz, S.M. Hileman, C. Bjorbaek, J.S. Flier, Two defects
contribute to hypothalamic leptin resistance in mice with diet-induced obesity,
J. Clin. Invest. 105 (2000) 1827–1832.
[35] Y. Zhang, R. Proenca, M. Maffei, M. Barone, L. Leopold, J. Friedman, Positional
cloning of the mouse obese gene and its human homologue, Nature 372 (1994)
425–432.
[36] J.M Friedman, J.L Halaas, Leptin and the regulation of body weight in mammals,
Nature 395 (1998) 763–770.
[37] M.A. Pelleymounter, M.B. Baker, M. McCaleb, Does estradiol mediate leptin’s
effects on adiposity and body weight? Am. J. Physiol. Endocrinol. Metab. 276
(1999) 955–963.
[38] P. Havel, S. Kasim-Karakas, G. Dubuc, W. Mueller, S. Phinney, Gender differences
in plasma leptin concentrations, Nat. Med. 2 (1996) 949–950.
[39] R.D. Leite, J. Prestes, C.F. Bernardes, G.E. Shiguemoto, G.B. Pereira, J.O. Duarte,
M.M. Domingos, V. Baldissera, S.E. De Andrade Perez, Effects of ovariectomy and resistance training on lipid content in skeletal muscle, liver, and
heart; fat depots; and lipid profile, Appl. Physiol. Nutr. Metab. 34 (2009)
1079–1086.
[40] S.R. Lindheim, M. Notelovitz, E.B. Feldman, S. Larsen, F.Y. Khan, R.A. Lobo, The
independent effects of exercise and estrogen on lipids and lipoproteins in postmenopausal women, Obest. Gynecol. 83 (1994) 167–172.
[41] R.S. Mazzeo, S.M. Horvath, Effects of training on weight, food intake, and body
composition in aging rats, Am. J. Clin. Nutr. 44 (1986) 732–738.
[42] M. Mohanka, M. Irwin, S.R. Heckbert, Y. Yasui, B. Sorensen, J. Chubak, S.S.
Tworoger, C.M. Ulrich, A. Mc Tierman, Serum lipoproteins in overweight/obese
postmenopausal women: a one-year exercise trial, Med. Sci. Sports Exerc. 38
(2006) 231–239.
[43] S.C. Woods, R.J. Seeley, P.A. Rushing, D. D’Alessio, P. Tso, A controlled high-fat
diet induces an obese syndrome in rats, J. Nutr. 133 (2003) 1081–1087.
[44] C.K. Sites, G.D. L’Hommedieu, M.J. Toth, M. Brochu, B.C. Cooper, P.A. Fairhurst,
The effect of hormone replacement therapy on body composition, body fat
distribution, and insulin sensitivity in menopausal women: a randomized,
double-blind, placebo-controlled trial, J. Clin. Endocrinol. Metab. 90 (2005)
2701–2707.
[45] A. Tchernof, A. Desmeules, C. Richard, P. Laberge, M. Daris, J. Mailloux, C.
Rhéaume, P. Dupont, Ovarian hormone status and abdominal visceral adipose
tissue metabolism, J. Clin. Endcrinol. Metab. 89 (2004) 3425–3430.
[46] A. Damirchi, R. Mehdizade, M.M. Ansar, B. Soltani, P. Babaei, Effects of aerobic exercise training on visceral fat and serum adiponectin concentration in
ovariectomized rats, Climacteric 13 (2010) 171–178.
[47] M. Vacca, C. Degirolamo, R. Mariani-Costantini, G. Palasciano, A. Moschetta,
Lipid-sensing nuclear receptors in the pathophysiology and treatment of the
metabolic syndrome, WIREs Syst. Biol. Med. (2011), doi:10.1002/wsbm.137.