If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported in part by the U.S. Department of Agriculture and NIDDK, National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by a Canadian Institute for Health Research doctoral fellowship and a Woodrow-Wilson Johnson and Johnson Dissertation Fellowship for Research in Women's Health.
Menopause is associated with increased adiposity and greater risk of metabolic disease. In the ovariectomized (OVX) rodent model of menopause, increased adiposity is prevented by estrogen (E2) replacement, reflecting both anorexigenic and potentially metabolic actions of E2. To elucidate metabolic and molecular mechanisms by which E2 regulates fat storage and fat mobilization independently of reduced energy intake, C57 BL/6 mice were ovariectomized, randomized to estrogen (OVX-E2) or control pellet implants (OVX-C), and pairfed for 40 days. E2 treatment was associated with reduced adipose mass and adipocyte size and down-regulation of lipogenic genes in adipocytes under the control of sterol-regulatory element-binding protein 1c. Adipocytes of OVX-E2 mice contained >3-fold more perilipin protein than adipocytes of pairfed control (OVX) mice, and this difference was associated with enhanced ex vivo lipolytic response to catecholamines and with greater levels of serum-free fatty acids following fasting. As in adipose tissue, E2 decreased the expression of lipogenic genes in liver and skeletal muscle. In the latter, E2 appears to promote the partitioning of free fatty acids toward oxidation and away from triglyceride storage by up-regulating the expression of peroxisome proliferation activator receptor-δ and its downstream targets and also by directly and rapidly activating AMP-activated protein kinase. Thus, novel genomic and non-genomic actions of E2 promote leanness in OVX mice independently of reduced energy intake.
). The classical genomic mechanism of E2 action involves activation of its nuclear receptor (estrogen receptor (ER) α or β), receptor dimerization, and subsequent binding to ER response elements (EREs) located in the promoters of target genes. More recently, E2 has been shown to have rapid, non-genomic biological effects, believed to be mediated through membrane-bound subpopulations of ER-α and ER-β and/or the newly described G protein-coupled receptor 30 (GPRC30) (
). Here, we provide evidence that E2 has significant effects on energy metabolism through both genomic and non-genomic mechanisms and that these effects collectively promote leanness in pairfed ovariectomized (OVX) mice.
Systemic loss of estrogen at menopause is associated with increased adiposity, which is implicated in the elevated risk of age-related metabolic disease in women (
). For example, women randomized to hormone replacement therapy in the Women's Health Initiative were leaner, more insulin sensitive, and less likely to develop Type 2 diabetes than women randomized to placebo (
The molecular and cell-biological mechanisms underlying the metabolic actions of estrogen are poorly understood. Much of our current view of the actions of estrogen on adiposity and metabolism derive from the ovariectomized rodent model of menopause. OVX rodents rapidly become obese and E2 administration prevents the increase in body fat (
). Thus, reduced energy intake contributes to reduced adiposity in E2-treated OVX rodents (OVX-E2). This differential energy intake confounds our elucidation of how and to what extent, reduced adiposity in OVX-E2 rodents reflects estrogen actions on metabolically critical tissues (adipose, liver, and muscle).
Employing a pairfeeding paradigm, the present study defines for the first time how estrogen protects against increased adiposity in OVX mice independently of differences in energy intake. We demonstrate that estrogen inhibits lipogenic gene expression, promotes catecholamine-stimulated lipolysis in adipocytes, and stimulates lipid-oxidative pathways in muscle. These metabolic effects reflect the genomic actions of estrogen on SREBP-1c and PPAR-δ-regulated gene expression, as well as novel, non-genomic estrogen actions on AMPK-activated protein kinase (AMPK).
MATERIALS AND METHODS
Animals—Female C57BL/6 mice were obtained from Taconic Laboratories (Germantown, NY). All animals were housed in an AAALAC-approved animal facility with 12-h light/dark cycles, given free access to water, and fed standard rodent chow (PROLAB, Syracuse, NY). Bilateral ovariectomy was performed on 10-week-old mice as previously described (
). After a 7-day recovery, mice were randomized to either placebo pellets (OVX-C) or E2 pellets (OVX-E2) containing 0.25 mg of 17β-estradiol released over 60 days (Innovation Research, Sarasota, FL) (n = 4 and 5, respectively). This supplementation protocol re-establishes and maintains circulating physiological levels of estrogen in ovariectomized mice (
). OVX-C mice were pair-fed to OVX-E2 mice. To do this, OVX-C mice were individually housed and fed the average food intake of the OVX-E2 group from the previous day. Food intake and body weight were monitored daily. On the day prior to tissue harvest, mice were fed at 16:00 h and food was removed at 20:00 h for an overnight fast.
Plasma Collection and Systemic Measures—After 40 days of treatment, blood was obtained from overnight-fasted mice by terminal exsanguination under isoflurane anesthesia. Plasma-free fatty acids (FFA) were measured using a NEFA-C assay kit (Wako Chemicals, Richmond, VA). Plasma adiponectin was measured by radioimmunoassay and plasma resistin, insulin, and leptin were measured using a mouse adipokine multiplex assay (Linco Diagnostics, St. Charles, MI). Glucose was measured using a Beckman Glucose Analyzer 2 (Beckman, Fullerton, CA).
Adiposity Index and Fat Pad Weight—Periovarian, perirenal, inguinal subcutaneous, mesenteric, and omental fat pads were excised and weighed. Adiposity index, a quantitative measure of total fat mass, was calculated using the previously described equation (
) with minor modifications. The samples were gassed (5% CO2, 95% O2), capped, and incubated at 37 °C with shaking until digestion was complete (30–40 min).
Adipocyte Lipolysis—Adipocytes were aliquoted in triplicate to measure lipolysis under basal (200 nm phenyl isopropyl adenosine) and stimulated (200 nm phenyl isopropyl adenosine + 10 μm epinephrine) conditions. Lipolytic rate was assessed as glycerol released into the media over 3 h using a Free Glycerol Determination kit (Sigma) (
). Lipid content (mg/ml) was determined by organic extraction of lipids from isolated adipocytes. Adipocytes were visually sized (100–150 adipocytes per mouse) using a light microscope and a 100-μm ocular micrometer. Cell number was then calculated using measures of lipid concentration and adipocyte diameter (
). On day 5, cells were serum-depleted in Dulbecco's modified Eagle's medium for 16 h prior to experiments. 17β-Estradiol and ICI 182,780 were purchased from Sigma and Tocris (Ellisville, MO), respectively. ICI 182,780 was added to the appropriate wells 30 min prior to the addition of E2. Incubations were terminated by addition of ice-cold phosphate-buffered saline containing 10 nm NaF and 1 mm Na3VO4.
Western Blotting—SDS-PAGE and Western blotting of adipocyte lysates for perilipin and hormone-sensitive lipase (HSL) was performed as previously described (
) was used to quantitatively recover perilipin from the adipocyte fat cake. Protein concentration was determined by the bicinchoninic acid method (Pierce). Blots were probed with polyclonal antibodies to perilipin A (
), HSL (a generous gift from Dr. F. A. Kraemer), and β-actin (Sigma), which served as a loading control. Chemiluminescence (Super Signal, Pierce) was quantified by laser densitometry within the linear range of detection.
E2-induced phosphorylation of AMPK on threonine 172 was assessed in lysates of mouse skeletal muscle (rectus femoris) and/or C2C12 myocytes following sonication in SDS lysis buffer containing phosphatase and protease inhibitors. Lysates were electrophoresed, transferred to nitrocellulose, and probed with antibodies against AMPK and phosphorylated AMPK (Cell Signaling, Beverly MA).
Real-time PCR—Adipose tissue (perirenal), liver, and muscle (rectus femoris) were dissected, immediately placed in RNAlater (Qiagen, Valencia California), and subsequently frozen in liquid nitrogen and stored at –80 °C. Total RNA was extracted from ∼100 mg of frozen tissue using commercially available kits (Qiagen, Valencia, CA). RNA was quantified by RiboGreen Quantitation Assay (Molecular Probes, Eugene, OR) and cDNA was synthesized from 1 μg of total RNA using a Reverse Transcription System (Promega). Real-time PCR was performed on each sample in triplicate on an ABI PRISM® 7700 Sequence Detection System in a 20-μl total volume using SYBR® Green PCR Master MIX (Applied Biosystems, Foster City, CA). Primers were designed using Primer Express or from previously published sequences (sequences available on request). Data were analyzed using a comparative critical threshold (Ct) method (
), with the amount of target gene normalized to the average of two endogenous control genes (18 S ribosomal RNA and cyclophilin B). Percent difference was calculated by 2–ΔΔCt.
Statistical Analysis—Data are reported as mean ± S.E. and were analyzed by SPSS 10.0.7 (SPSS, Chicago, IL). Treatment effects were evaluated by independent T tests of group means using Bonferroni adjustments for multiple comparisons. Statistical significance was defined as p < 0.05.
Estrogen Decreases Adiposity and Adipocyte Size in Pairfed Ovariectomized Mice—Estrogen-treated mice (OVX-E2) consumed less food on day 1 (∼30%) as compared with ad libitum fed controls (OVX-C) (Fig. 1A). After day 1, OVX-C received only the amount of food consumed by OVX-E2 mice on the previous day. On day 40, body weights and indices of linear growth (femur length, nasoanal length) were equivalent in OVX and OVX-E2 mice. There was no difference in femur length or nasoanal length, indicating linear growth was not affected by treatment (Fig. 1B). Uterine weight was 10-fold higher in OVX-E2 mice, confirming the efficacy of estrogen treatment in this study (Fig. 1B).
Whereas body weights and linear growth were similar, total adiposity was reduced (∼30%) in OVX-E2 mice (Fig. 2A). The associated reductions in adipose mass were observed in both intra-abdominal (perirenal, periovarian, and mesenteric/omental) and subcutaneous depots, although the effect was not statistically significant in subcutaneous fat pads (trend, p = 0.1) (Fig. 2B). Adipose mass reflects both adipocyte number and adipocyte size, the latter a direct manifestation of fat (triglyceride) storage. Decreased adipose tissue mass in OVX-E2 mice was coincident with comparable reductions in mean adipocyte size, suggesting that estrogen effects on adipose mass are mediated in large part by estrogen effects on adipocyte size (Fig. 2C). These data demonstrate that chronic estrogen treatment promotes smaller adipocytes and reduced adipose mass independently of reduced energy intake in OVX mice.
Estrogen Treatment Increases Plasma FFA and Decreases Plasma Glucose in Fasted Mice—Because both estrogen treatment and body fat mass may influence plasma FFA, insulin, and glucose, we measured these metabolites in our fasting animals. Plasma insulin values were not different between groups but OVC-E2 mice had significantly lower serum glucose values as compared with the OVX-C, suggesting differences in glucose/insulin homeostasis (Fig. 3). Interestingly, plasma FFAs in OVX-E2 mice were approximately twice that of OVX-C mice. These results were surprising in light of the fact that elevated circulating FFA are believed to promote peripheral insulin resistance by accumulation of fatty acid derivatives in muscle and liver (
). To better understand this apparent paradox, we examined genomic and signaling pathways that may allow peripheral tissues to better accommodate (i.e. utilize) increased circulating FFA (discussed later in the article) as well as potential factors contributing to elevated FFA.
Estrogen Treatment Increases Adipocyte Perilipin Expression and Enhances Lipolytic Response to Catecholamines—To determine whether elevated circulating FFA in fasting OVX-E2 mice reflected increased rates of catecholamine-stimulated lipolysis, we measured lipolysis in isolated periovarian adipocytes from OVX-C and OVX-E2 mice in the absence (basal) and presence of the catecholamine, epinephrine. Adipocytes isolated from E2-treated mice demonstrated lower rates of basal lipolysis (in the absence of catecholamines), but an approximate doubling of lipolytic rate following stimulation by epinephrine (Fig. 4A). Similar results were found when lipolysis was expressed per mg of lipid, per adipocyte surface area, and when lipolysis was stimulated by isoproterenol, a selective β-adrenergic receptor agonist (data not shown).
Enhanced catecholamine-stimulated lipolysis was not associated with increases in HSL, the predominant lipase in adipocytes (Fig. 4B). However, levels of the lipid droplet-associated protein perilipin were elevated (>3-fold) in adipocytes of OVX-E2 mice. Increased expression of perilipin has been shown to suppress basal lipolysis and increase stimulated lipolysis (
), a pattern observed in OVX-E2 adipocytes in the present study. Interestingly, there was no difference in perilipin mRNA expression (Fig. 5), suggesting that perilipin protein expression is regulated, at least in part, post-translationally.
These results suggest that chronic E2 treatment enhances the lipolytic rate by enhancing levels of perilipin protein. This enhancement may be related to E2-induced reductions in adipocyte size (
Estrogen Regulates Adipose Tissue Gene Expression to Prevent TG Accumulation—We used real-time PCR to elucidate the underlying molecular mechanism(s) by which E2 decreases adipose tissue mass. These studies revealed dramatic E2-induced down-regulation of genes that promote adipocyte lipid storage (Fig. 5). These included lipoprotein lipase (LPL), which promotes adipocyte uptake of circulating lipids for TG formation, and the lipogenic genes, acetyl-CoA carboxylase-1 (ACC-1) and fatty acid synthase (FAS). E2-mediated reductions in ACC-1 and FAS gene expression are likely the result of down-regulation of liver X receptor α (LXR-α) and SREBP-1c, a downstream target of LXR-α that transcriptionally regulates ACC-1 and FAS gene expression (Fig. 5). It was recently demonstrated that E2 can directly regulate adipose tissue LXR-α expression through an ERE in the LXR-α promoter (
In contrast to the down-regulation of genes promoting lipid storage, E2 had no effect on markers of adipocyte differentiation (PPAR-γ, preadipocyte factor 1, and CCCAT/enhancer binding protein-α), or on genes regulating thermogenesis (uncoupling protein-1 (UCP-1)), fat oxidation (pyruvate dehydrogenase kinase 4, acyl-CoA oxidase (ACOX)), or oxidative metabolism-related transcription factors (PPAR-α, PPAR-γ coactivator protein 1α, and estrogen-related receptor α) (data not shown). Collectively, data for adipocytes suggest that estrogen promotes leanness in part by reducing adipocyte size through reduced uptake of fatty acids (down-regulated LPL), reduced lipogenesis (down-regulated ACC-1 and FAS), and elevated catecholamine-stimulated lipolysis.
Novel Non-genomic Actions of Estrogen Directly Activate (Phosphorylate) AMPK in Muscle—AMPK is a critical energy sensor in metabolic tissues. Activation of this kinase increases glucose uptake and facilitates fat oxidation by promoting FFA uptake into mitochondria. Activation of AMPK is believed to be an important mediator of the beneficial metabolic effects of exercise (
). To assess the potential role of AMPK activation in mediating the metabolic effects of estrogen, we assessed AMPK phosphorylation (threonine 172), a hallmark of AMPK activation in muscle, liver, and adipose tissue of OVX-C and OVX-E2 mice. Western blot analysis revealed a dramatic increase (∼5-fold) in AMPK phosphorylation in the skeletal muscle of OVX-E2 mice (Fig. 6). No differences were seen in liver and adipose tissue AMPK phosphorylation (data not shown). These results suggest that chronic estrogen treatment selectively up-regulates AMPK activity in skeletal muscle of ovariectomized mice.
The adipokines leptin and adiponectin have been shown to activate AMPK, whereas resistin is proposed to decrease AMPK activation (
). To evaluate if elevated muscle AMPK activation in OVX-E2 mice reflected E2-induced alterations in adipokine levels, we compared plasma and adipose tissue mRNA levels of adiponectin, leptin, and resistin in OVX and OVX-E2 mice (Fig. 7). Levels of AMPK-activating adipokines, leptin, and adiponectin, were reduced, presumably reflecting smaller adipocytes (
), respectively, thus likely do not contribute to increased AMPK activation in OVX-E2 mice. Whereas levels of the AMPK inhibitor, resistin, were also reduced and therefore may result in increased AMPK activity, there are no reports of resistin inhibiting AMPK in muscle, only liver (
). Overall, these results do not support the hypothesis that altered adipokine levels account for AMPK activation in OVX-E2 mice.
Alternatively, estrogen could directly activate AMPK independently of adipokines. We investigated the ability of estrogen to directly activate AMPK in C2C12 myocytes, a well established mouse muscle cell line that exhibits many characteristics of muscle, including contraction. Remarkably, estrogen rapidly (within minutes) activated AMPK in a time- and dose-dependant manner (Fig. 8, A and B). AMPK phosphorylation was increased on average 4-fold (at maximal E2 concentration 10 μm). Importantly, this effect was completely inhibited by ICI 182,780, a selective antagonist of estrogen receptors (ER-α and ER-β). These results demonstrate that estrogen can stimulate the rapid and robust phosphorylation of AMPK, and this likely occurs through non-genomic activation of membrane-bound ERs.
Estrogen Treatment Down-regulates Muscle and Liver SREBP-1c and Up-regulates Muscle PPAR-δ—To further examine estrogen-mediated changes that may promote decreases in adipose tissue mass, we investigated the expression of several genes involved in the regulation of lipogenesis and oxidation in the muscle and liver of OVX-C and OVX-E2 mice.
Similar to observations in adipose tissue (Fig. 5), estrogen decreased expression of the lipogenic genes, SREBP-1c, and its downstream targets, ACC-1 and FAS, in muscle and liver (Fig. 9, A, upper panel, and B). However, in contrast to adipose tissue, LXR-α expression was not down-regulated by estrogen in skeletal muscle or liver. In muscle, estrogen significantly up-regulated transcription factor, PPAR-δ, and its downstream targets, LPL, pyruvate dehydrogenase kinase 4, ACOX, and the UCPs, UCP2, and UCP3 (Fig. 9A, lower panel) (
). Similar to adipose tissue, estrogen did not alter the expression of oxidative metabolism-related transcription factors PPAR-α, PPAR-γ coactivator protein 1α, or estrogen-related receptor α in muscle or liver (data not shown).
This is the first study to demonstrate that estrogen reduces adiposity in OVX rodents, which is not confounded by differences in food intake between the E2-treated and control animal cohort. Thus, the reduced adiposity and adipocyte size observed in E2-treated mice (Fig. 2) can be more clearly understood within the context of E2 effects on peripheral metabolism (see below). E2-associated reductions in adipose tissue mass in pairfed OVX mice are consistent with the phenotypes of both estrogen receptor-α (ERKO) knock-out and aromatase (and thus estrogen)-deficient mice, both of which exhibit increased adiposity with no reported differences in food intake (
). Our studies demonstrate novel genomic and non-genomic mechanisms by which E2 promotes a reduction in adipose tissue mass and adipocyte size. Interestingly, reductions in adiposity were statistically significant in intra-abdominal but not subcutaneous fat depots (Fig. 2B), consistent with the observation in menopausal women that hormone replacement therapy primarily acts on intra-abdominal depots, thereby preventing central adiposity (
Our data suggests that E2 reduces adiposity, at least in part, in OVX mice by promoting the use of lipid as fuel. This effect of E2 on fuel partitioning reflects: 1) enhancement of pathways that promote fat oxidation in muscle; 2) inhibition of fat storage (lipogenesis) in adipose, liver, and muscle; and 3) enhanced rates of adipocyte lipolysis.
We identified two novel actions of E2 to promote fat oxidation in muscle. First, estrogen up-regulated the expression of the transcription factor, PPAR-δ, and a number of its downstream targets (Fig. 9) (
) observed that estrogen promoted lipid oxidation in PPAR-α-deficient mice.
Our evidence suggests E2, through up-regulation of PPAR-δ expression, acts in muscle to alter fuel partitioning and oxidative capacity to enhance muscle oxidation. In fact, this effect may be potentiated by PPAR-δ-mediated up-regulation of LPL, which may promote uptake of endogenous ligands that activate PPAR-δ (
). In addition, increased oxidative capacity in muscle, through regulation of ACOX, PDHK, UCP2, and UCP3 expression, would allow for enhanced FA uptake without lipid accumulation as the substrate could be immediately oxidized.
In addition to the genomic effects of estrogen, we demonstrated another means by which E2 may enhance fat oxidation: through activation of AMPK. While this article was being prepared, Shulz et al. (
) demonstrated E2-activation of AMPK in endothelial cells. AMPK, by regulating intracellular signals, acts as a “fuel gauge” regulating fat oxidation, FA synthesis, and glucose uptake. Activation of AMPK results in inactivation of ACC, which prevents the synthesis of malonyl-CoA, a metabolite necessary for TG synthesis. This, in turn, increases carnitine palmityltransferase-1 activity and allows for long chain FA transport to the mitochondria for oxidation. Here, we demonstrate for the first time that estrogen rapidly activates AMPK in skeletal muscle. Further studies are warranted to determine whether the effect of E2 in skeletal muscle is mediated through ER-α, ER-β, the newly described estrogen-regulated receptor, G protein-coupled receptor 30 (
In addition to promoting fat oxidation, E2 appears to act globally on all metabolic tissues (adipose tissue, muscle and liver) to prevent lipogenesis. Estrogen treatment decreased expression of SREBP-1c in adipose tissue, muscle, and liver, and adipose tissue expression of LXR-α, a positive regulator of SREBP-1c (
). Whereas E2-mediated reductions of LXR-α expression likely contribute to a decrease in adipose tissue SREBP-1c expression, LXR-α is not reduced in muscle or liver in response to E2. This suggests an alternate mechanism by which E2 regulates expression of SREBP-1c and its downstream targets. It is possible that E2 directly regulates SREBP-1c, which contains an ERE in its promoter region (
), down-regulated muscle SREBP-1c expression in the present study may reflect the novel non-genomic effects of E2 on AMPK activation (Fig. 6). Similarly, reduced expression of lipogenic genes may reflect the partitioning of fuel toward fat oxidation thereby reducing the availability of substrate for TG synthesis.
In either case, reduced expression of SREBP-1c target genes is expected to reduce de novo lipogenesis, with concomitant reductions in stored lipid and adipocyte size. In addition, E2-dependent repression of adipose tissue LPL expression (Fig. 5) would further inhibit lipid accumulation in adipocytes by decreasing the uptake of FAs from the circulation for TG synthesis.
Last, TG breakdown (i.e. lipolysis) may also contribute to E2-mediated reductions in adipose tissue mass and adipocyte size. Acute estrogen administration has previously been shown to increase the lipolytic rate in vivo (
) demonstrated that adipose tissue deinnervation reduces E2-mediated reductions in adipose tissue mass by 23%. Here, we identify a novel mechanism by which estrogen potentially promotes chronically elevated rates of adipocyte lipolysis. Specifically, we demonstrate that E2 treatment increases levels of the lipid droplet protein, perilipin, a key regulator of lipolysis (Fig. 4) (
). Adipocytes with higher perilipin expression exhibited considerably enhanced responsiveness to catecholamine stimulation (Fig. 4). This, coupled to E2-mediated increases in sympathetic activity, would further enhance the effects of E2 on reducing adipose tissue TG accumulation.
In summary, we demonstrate for the first time that E2 decreases adiposity and adipocyte size in OVX mice independent of differences in energy intake. We identify novel genomic and non-genomic pathways that would promote fat oxidation, prevent TG accumulation, and enhance TG breakdown (lipolysis). These metabolic effects reflect the genomic actions of estrogen on SREBP-1c and PPAR-δ-regulated gene expression, as well as novel, non-genomic estrogen actions on AMPK.