Estrogens and obesity: Is it all in our heads?

Abstract

Estrogens have preventative effects on weight gain and associated comorbidities, but the tissuespecific targets remain unknown. Here, Xu et al. demonstrate that ablation of estrogen signaling in two populations of hypothalamic neurons leads to weight gain and subsequent metabolic dysregulation and could be important target sites of estrogen actions.

The obesity epidemic continues to pose one of the largest worldwide threats to overall health and currently there are few effective preventative or therapeutic options. Determining the factors that mediate both weight gain and its metabolic effects is a priority, as this understanding may both allow for stratification of populations at risk and the discovery of pharmacologic targets for eventual therapies. Estrogen receptors are one such potential target, as estrogen signaling has been shown to have powerful metabolic effects in both clinical and experimental models. Selective estrogen receptor modulators (SERMs) may be of benefit, having the ability to activate estrogen receptors in specific tissues while remaining inert or leading to inactivation in others. In this issue of Cell Metabolism, Xu et al. identify key targets within the central nervous system that may represent an important step towards developing SERMs designed to counter the obesity crisis.

In humans, population-based studies of post-menopausal women have shown that the risks of insulin resistance, fatty liver disease, and cardiovascular disease are increased compared to age matched controls (Carr, 2003). Body composition changes parallel these risks, specifically a redistribution of fat towards the visceral compartment. Estrogen replacement therapy (HRT) has a protective effect against many of these changes. However, the increased cardiovascular and cancer risks associated with HRT make estrogen replacement itself an unsuitable option for preventative use (Rossouw et al., 2002), hence the need for more targeted therapies.

Despite this growing need, currently little is known regarding the specific sites of action necessary for estrogen to exert its protective effects or its cellular mechanism of action. The impact on obesity has been shown to be mediated primarily by signaling through estrogen receptor-alpha (ERα). Nearly two decades ago, we described a male patient with an ERα loss of function mutation (Smith et al., 1994). By age 24, this patient had overt type 2 diabetes and hyperinsulinemia with a BMI in the obese range. Animal studies later corroborated these findings, demonstrating that global knockout of ERα is sufficient to produce excess weight gain, increased visceral fat, glucose intolerance, insulin resistance, and dyslipidemia (Ribas et al., 2010). However, as ERα is expressed in a number of tissues, including muscle, liver, adipocytes, reproductive organs, and the central nervous system, this discovery did not identify the precise site of action for estrogen and ERα signaling.

The study by Xu et al. represents an important step in the narrowing of this vast array of possibilities. By using a nestin-Cre construct to greatly reduce ERα expression within the entire central nervous system, Xu et al. were able to demonstrate abnormal weight gain in both male and female mice, with associated glucose intolerance. The females also had increased body fat and visceral fat compared to wild-type controls. Interestingly, both hyperphagia and reduced energy expenditure were observed in these conditional knockouts. While each could certainly contribute to the increased weight gain, this observation differs from previous studies examining metabolic parameters in whole body ERα knockouts (Park et al., 2011; Ribas et al., 2010) that showed decreased energy consumption without increased food intake. One potential explanation for this discrepancy is that high levels of circulating estradiol were seen in the nestin-Cre conditional knockouts while peripheral ERα signaling remained largely intact, a major difference in these models.

To further delineate the regions of the CNS involved, Xu et al. focused on the importance of ERα signaling in two populations of hypothalamic neurons known to be involved in regulating metabolism and energy balance. Within the brain, steroidogenic-factor 1 (SF-1) is expressed exclusively in neurons of the ventromedial hypothalamus. Signaling in SF-1 neurons is essential for the compensatory increase in energy expenditure seen after exposure to a hypercaloric diet (Xu et al., 2010). By suppressing ERα expression specifically in SF-1 expressing cells, Xu et al. demonstrated an important role for estrogen signaling in the impact of these neurons on energy homeostasis and body fat partitioning, particularly in females. The latter of these effects is particularly intriguing in that adipocyte size and triglyceride content were elevated only in gonadal (visceral) adipose tissue, the compartment most associated with impaired glucose tolerance and dyslipidemia. Further work to elucidate how these central signals modulate the observed peripheral effects is an important next step.

The second population of neurons investigated were neurons in the arcuate nucleus of the hypothalamus expressing proopiomelanocortin (POMC). POMC neurons are important for regulation of food intake through the secretion of α-melanocyte stimulating hormone, which signals through the melanocortin-4 receptor (MC4R) to induce satiety. In humans, MC4R mutations are the most common genetic cause of non-syndromic early-onset obesity, and such mutations are present in 0.5 – 5% of severely obese individuals (Farooqi et al., 2003). Xu et al, through conditional deletion of ERα in POMC-expressing neurons, determined that in female mice, signaling through ERα is important for the effects these neurons have on food intake.

This study represents the first demonstration of a tissue-specific suppression of ERα leading to profound effects on energy balance, fat partitioning, and metabolic parameters. However, peripheral estrogen signaling – through ERα in the liver, myocytes, adipocytes, or other tissues -- may still be important in modulating these metabolic effects and needs to be considered. In fact, recently it was demonstrated that tissue-specific deletion of ERα in myeloid or hematopoietic cells led to increased adipocyte mass coupled with glucose intolerance and insulin resistance (Ribas et al., 2011). In future studies of animals with SF-1 or POMC-specific ERα deletions, a comparison group of animals with wholebody ERα deletions would be helpful in determining to what extent these conditional deletions recapitulate the metabolic phenotype of animals with a global whole-body deficiency.

While studies of the actions of ERα in other tissues should continue, this does not discount the importance of the findings in this study. Identification of specific targets, by which estrogen signaling impacts food intake and energy expenditure, is an essential step towards the development of targeted therapies. Further investigation of the coactivators and signaling mechanisms involved in ERα-dependent activation of the SF-1 and POMC neurons should be undertaken with the hopes of eventually developing a pharmacologic treatment for the burgeoning obesity epidemic and its comorbidities.

Figure 1. Effects of hypothalamic estrogen signaling on body weight.

A schematic illustrating three outcomes of estrogen signaling mediated through ERα via two populations of hypothalamic neurons as demonstrated in the following study by Xu et al., all of which lead to maintaining a proper body weight.