Introduction
The onset of menopause is a milestone in women health as it is accompanied by an increase in the occurrence of several health conditions like osteoporosis, sarcopenia or cardiovascular diseases [1] which are also a common consequence of sedentary lifestyles [2]. This suggests these health conditions may not only be a direct consequence of estrogen lack in skeletal, muscular or cardiovascular systems but also the result of increased sedentary behaviors following menopause.
Results from several studies suggest that estrogen is a chief regulator of motor activity and that reduced circulating estrogen levels are associated with sedentary behaviors [3, 4]. Ovariectomized (OVX) rats for instance, display significantly less running activity when compared to intact controls [5] and estrogen replacement is shown to effectively reverse OVX induced physical inactivity in mice [3, 4] in a somewhat dose dependent fashion [6]. Moreover, locomotor activity variations in female rat [7] and mice [8] have been shown to accompany estrogen variations along the estrous cycle, with simultaneous occurrence of peak activity and estrogen levels. Although findings in higher primates [9, 10] and humans [11] are not so consistent as in other animals, there is also evidence suggesting that motor activity in women varies along the menstrual cycle as well [12, 13]. However, the mechanisms by which estrogen influences motor activity are still largely unknown, partially because estrogen has the potential of targeting different tissues involved in motor activity control.
Skeletal muscle in several species [14, 15] including humans [16, 17] express estrogen receptors (ER) which makes it a target tissue for estrogen action. Decreases in muscle mass and strength are a well described outcome of menopause [18] which are able to be prevented by estrogen replacement therapy [19]. Moreover, there is evidence that low estrogen levels following OVX are associated with a significant decrease in mice soleus muscle maximal strength [20] and that this decrease is successfully restored after treatment with estrogen [21]. Significant changes in the contractile properties of mice soleus muscle were also reported following OVX due to a significant shift in myofiber composition from fast to slow myosin heavy chain [22]. Furthermore, estrogen enhances skeletal muscle antioxidant status since lipid peroxidation damage is decreased [23] and glutathione peroxidase mRNA expression is increased [24] in OVX mice supplemented with estrogen. This protective mechanism migth explain the anti-apoptotic effects of estrogen described in mice skeletal muscle [25, 26]. Therefore the decline in motor activity identified following estrogen loss [3, 4], could result from impaired muscular function due to a direct low estrogen effect on muscle tissue.
However, some of the features identified in skeletal muscle of OVX animals are absent when the animals are stimulated by exercise [22] which suggests that estrogen lack per se is not the only cause for the previously described skeletal muscle alterations. There is also evidence that estrogen effects at the central nervous system (CNS) level are a potential cause for the estrogen deficiency induced physical inactivity [27-29]. Estrogen receptors are present on brain cells [27] and the systemic administration of radio-labeled estrogen shows that the preoptic area and the ventromedial hypothalamus have high affinity for estrogen [30]. Interestingly, damage to these brain structures hinders the increase in running activity elicited by estrogen replacement therapy on OVX rats [28]. In addition, targeted inhibition of the ventromedial hypothalamus ERα was shown to cause a decrease in voluntary running activity [29] while local administration of estrogen in the medial preoptic and anterior hypothalamic areas were able to enhance running activity in OVX rats [31]. However, there is also evidence that estrogen administration to postmenopausal woman increases GH release by the hypothalamus [32] which further stimulates IGF-I expression. In addition, blockage of ER is shown to decrease GH release [33, 34]. Therefore, it is possible that estrogen actions over the CNS may have a direct repercussion on skeletal muscle function by interfering with the release of mitogenic hormones [35].
Therefore, considering these evidences, physical inactivity induced by estrogen loss could result from two main mechanisms: i) due to the direct effect of estrogen lack on skeletal muscle structural and contractile properties, which in turn would hamper the ability to perform motor actions, ii) or on the other hand, the inactivity could be the result of a decrease in volitional will towards motor activity due to effects of estrogen lack on the CNS.
The objective of our research was to investigate witch of these two possible mechanisms better explain physical inactivity induced by estrogen loss. We hypothesized that skeletal muscle changes observed by others [20] reflect largely muscle atrophy due a decrease in voluntary motor activity and that skeletal muscle changes are not be the primary cause for the decreases in locomotor activity. We believe that our results will have meaningful clinical significance as they will allow clarification of the causes for estrogen deficiency induced physical inactivity and therefore to more appropriately define strategies for its prevention.
For achieving our purpose we have used OVX and sham operated female Wistar rats either housed in standard sedentary conditions or with accesses to running wheel. We then analyzed motor activity patterns and soleus muscle structure for identification of alterations that would indicate loss of functionality.
Materials and methods
Animal models and experimental design
All procedures involving animal care and sacrifice were approved by the local ethics committee. Following one week of quarantine after arrival, 25 nulliparous female Wistar rats aged 5 months (Charles River laboratories, Barcelona) were randomly ovariectomized (OVX; n=13) or sham operated (SHAM; n=12). Bilateral ovariectomy was performed by standard ventral approach [36] under anesthesia with 4% sevoflurane. Sham surgery consisted in exposure without removal of the ovaries. After one week of recovery OVX and SHAM rats were separated in two sub-groups and housed in different conditions.
Two groups of rats were individually housed in cages with activity wheel and distance counter (floor area 800 cm2; Tecniplast, Italy) allowing them to perform voluntary running activity (OVX+VR, n=7; SHAM+VR, n=6) and two control groups were housed in identical conditions (OVX+C, n=6; SHAM+C, n=6) without the running wheel in the cage. Distance traveled by each rat in the running wheel was recorded daily and body weight and food intake were recorded weekly at the end of the light phase.
All animals were maintained in an inverted 12h light/dark cycle in humidity (50-60%) and temperature (21-22oC) controlled conditions. Standard rat chow and water were provided ad libitum throughout the experimental period.
Animal sacrifice and morphometric analysis
At 14 months of age all rats were anesthetized with 4% sevoflurane and sacrificed by exsanguination. The blood was collected from vena cava and was later used for biochemical analysis. Intra-abdominal fat (intraperitoneal + retroperitoneal fat), whole heart, left ventricle, both hid limbs soleus and gastrocnemius muscles were dissected, washed in could PBS (pH 7.2) and weighted in a precision balance (resolution 0.01 mg; Kern 870). Intra-abdominal fat was used as an estimator of whole body fat as described previously [29]. Right tibia was also collected and its length measured with a digital caliper (resolution 0.01mm, Powerfix) for adjusting weight comparisons to differences in animal size. Whole heart and left ventricle weight were used as estimators of myocardial adaptation to physical activity [37]. After weighted, right soleus muscle was transversely divided in two portions and each was processed immediately for light or transmission electron microscopy. Left soleus muscle was minced and homogenized in a ratio of 1:10 in ice-cold extraction medium containing 50mM Tris/base and 1mM EDTA (Sigma) pH 7.4 with a Potter-Elvehjem homogenizer and pestle. The homogenate was then collected, sonicated for 1min in ice cold water (Sonorex, Bandelin Electronic) and centrifuged at 700g for 10min at 4oC. The supernatant was collected for biochemical assessment of CS and GAPDH activity and type I myosin expression by Slot-Blot. Total protein content was determined according to the method described by Lowry et al [38] using BSA (Sigma) as standard.
Histology
Tissue processing for light and transmission electron microscopy
One of the right soleus muscle portions was fixed overnight in a solution containing 4% paraformaldehyde, 2.5% sucrose (Sigma) and 0.1% gluteraldehyde (TAAB) in PBS (pH 7.2) at 4oC, dehydrated through graded ethanol solutions, cleared in xylene and mounted in paraffin. Transverse 6 µm thick sections were cut, deparaffinized in xylene, rehydrated in graded ethanol solutions and stained according to the objective. Images from each section were recorded with a digital camera coupled to a light microscope (Axio Imager A1, Carl Zeiss; Germany).
The remaining portion of the right soleus muscle was sectioned into smaller fragments (≈2 mm3) and fixed overnight in 2.5% gluteraldehyde in 0.2M sodium cacodylate buffer (pH 7.2) at 4oC. After rinsing with 0.2M sodium cacodylate buffer (pH 7.2) soleus samples were post fixed with osmium tetroxide in 0.2M sodium cacodylate, dehydrated in graded ethanol and embedded in Epon resin (TAAB) until polymerization at 60oC. Ultra-thin (100 nm) sections were cut with an ultramicrotome and diamond knife, contrasted with uranyl acetate and lead citrate (Sigma) and examined in a transmission electron microscope (Zeiss EM10A) at an accelerating voltage of 60 kV.
Soleus muscle fiber cross-sectional area
For determination of soleus muscle cross sectional area, sections were stained with H&E and acquired photomicrographs analyzed with ImageJ software (Image Processing and Analysis in Java). Cross sectional area was determined in a total of 150 fibers per muscle. Results for each muscle were calculated as the average area of the 150 fibers analyzed.
Assessment of fibrous tissue accumulation
For determination of fibrous tissue accumulation, soleus muscle sections were stained with picrosirius red (PSR) according to the method of Sweat et al. [39]. Sections were incubated in a solution containing 0.1% Sirius Red in saturated picric acid for 1h, rinsed in 0.5% acetic acid, dehydrated in absolute ethanol and cleared in xylene. PSR technique stains collagen bright red and muscle tissue yellow. Obtained images were analyzed with Image-Pro Plus 6.0 software (Media Cybernetics, Inc.) by quantification of the percentage area covered by collagen (red) and muscle tissue (yellow).
Assessment of Soleus Muscle Apoptosis
The presence of apoptotic nuclei in soleus muscle was assayed by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL), using a commercially available kit (in situ cell death detection kit AP) according to the manufacturer instructions (Roche). After deparaffinization, sections were immersed in 0.1M citrate buffer (pH 6.0) and microwave irradiated for 1 min (750W). After rinsing in PBS, sections were first blocked with 3% BSA in 0.1M Tris-HCl (pH 7.5) for 30min at 20oC and then incubated in freshly prepared TUNEL reaction solution (nucleotide mixture + terminal deoxynucleotidyl transferase) in a humidified chamber for 60 min at 37oC in the dark. Negative controls were simultaneously prepared by incubation with label solution only (nucleotide mixture). Sections were analyzed with a fluorescent microscope (comprimento de onda da fonte de fluorescência) coupled to a digital camera (Axio Imager A1, Carl Zeiss). Apoptotic cells were identified as brightly fluorescent in opposition to the pale green background staining.
Assessment of Soleus muscle type I myosin expression by immunohistochemistry
After deparaffinization and rehydration, sections were rinsed in 0.1% TBS-T and non specific binding was blocked by incubation with 3% BSA in 0.1% TBS-T for 30 min at 37oC. Sections were then incubated with mouse monoclonal anti type I myosin (M8421, Sigma) primary antibody diluted 1:100 in 0.1% TBS-T overnight in a humidified chamber. After rinsing in 0.1% TBS-T, sections were subsequently incubated with goat anti-mouse alkaline phosphatase conjugated secondary antibody (SC-3698, Abcam) diluted 1:100 in 0.1% TBS-T for 1h at 37oC in a humidified chamber. Detection was performed by incubation with Fast Red TR/Naphthol AS-MX Tablets (SigmaFast, Sigma). Sections were then counterstained with hematoxylin. Negative controls were performed for each bone section by omission of the primary antibody.
Biochemical analysis
Determination of serum estrogen levels
Serum was separated from the blood samples by centrifugation at 4oC. Estradiol (17-β-estradiol) concentration was assayed by solid phase competitive binding ELISA using a commercially available kit (Estradiol ELISA DE2693) and a spectrophotometer (iEMS, Labsystems) according to the manufacturer recommendations (Demeditec Diagnostics) for confirmation of OVX animal model implementation. Briefly, 25µl of serum and 200µl of enzyme conjugate were dispensed in each plate well and incubated for 120 min at 25oC. After complete washing, 100µl of substrate solution was added to each plate well and incubated for a further 15min. Absorbance was read at 450nm after addition of 50µl stop solution. Duplicates were analyzed for each sample and triplicates for each standard. Assay range is between 9.7 to 2000 pg/mL.
Type I myosin immunoblot
Semi-quantification of type I myosin content was performed by slot blot as described previously [40]. Briefly, samples of soleus muscle homogenate containing 10µg of protein were loaded into a nitrocellulose membrane (Hybond; Amersham Biosciences) with a slot blot filtration manifold device (Hybri-Slot; Gibco BRL) and vacuum pump (KNF Neuberger). The membrane was then blocked in 5% non-fat dried milk in 0.1% TBS-T for 1h, incubated with primary antibody (mouse monoclonal anti type I myosin; M8421, Sigma) diluted 1:1000 in 5% non-fat dried milk in 0.1% TBS-T for 2h and then incubated in secondary antibody (………………………) conjugated with horseradish peroxidase diluted 1:1000 in 5% non-fat dried milk in 0.1% TBS-T for 2h. Detection was performed by enhanced chemiluminescence after membrane incubation with ECL (Amersham). The bands imprinted on the photographic film (Kodak Biomax Ligth Film; Sigma) were analyzed with imageJ software by calculation of their integrated density.
Determination of citrate synthase (CS) activity
CS activity was measured as described previously [41] by spectrophotometric measurement at 412nm of the amount of 5,5-dithio-bis(2-nitrobenzoate) that reacted with acetyl-coenzyme A (CoA) after its release from the reaction of acetyl-CoA with oxaloacetate. Brifly 150µl of ultrapure water, 20µl of freshly prepared assay buffer containing 1.0mM DTNB and 1% Triton X-100 (pH 8.1) and 20µl of the soleus homogenate were dispensed to each plate well. The absorbance was read for 5min with 15 seconds interval in a spectrophotometer (iEMS, Labsystems) at 37oC following the addition of 10µl of 10mM oxaloacetate. The activity was calculated using CS from porcine heart (C3260, Sigma) as standard and expressed per mg of protein content.
Determination of Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) activity
GAPDH activity was measured as described previously [42] as an increase in absorption at 340nm following the reduction of NAD. Briefly freshly prepared 173µl of assay buffer containing 15mM sodium pyrophosphate and 30mM sodium arsenate pH 8.5, 6.6µl 7.5 mM NAD, 6.6µl 100mM DTT (Sigma) and 6.6µl of the soleus homogenate were dispensed to each plate well. Absorbance was read at 340 nm for 5 min with 15 seconds interval in a spectrophotometer (iEMS, Labsystems) at 37oC following the addition of 6.6µl of 7.5mM glyceraldehyde 3-phosphate. The activity was calculated using GAPDH from rabbit muscle (C2267, Sigma) as standard and expressed per mg of protein content.
Statistical analysis
The Kolmogorov-Smirnov test was used to investigate within-group normality for a given variable. Levene's test was used to assess the homogeneity of variance. Statistical analysis was performed by one-way analysis of variance (ANOVA) with Bonferroni post hoc test for pair wise comparisons between groups if normality or equality of variance existed (comparisons were not adjusted for animals size as no significant correlation was identified between tibia length and organs weight). When normality or equality of variance assumptions was not met, root square transformation was employed (heart weight; intra-abdominal fat weight) to reestablish the necessary assumptions for ANOVA. Results are expressed as mean ± standard deviation (SD) and differences were considered significant when p<0.05.
Results
Discussion
Acknowledgments
The authors wish to express their gratitude to Celeste Resende for her skilled technical involvement. The present work was supported by Fundacão para a Ciência e Tecnologia (FCT) grant PTDC/DES/103047/2008. First and second authors are supported by FCT fellows SFRH/BD/38110/2007 and SFRH/BD/33123/2007 respectively. All authors have no conflicts of interest to disclose.
