Epidemiological evidence supports a correlation between obesity and breast cancer in women. AMP-activated protein kinase (AMPK) is recognized to be a master regulator of energy homeostasis. One of its actions is to phosphorylate and inhibit the actions of cAMP-responsive element binding protein (CREB)-regulated transcription coactivator 2 (CRTC2). In postmenopausal women, the CREB-dependent regulation of aromatase is a crucial determinant of breast tumor formation through local production of estrogens. We report here that the regulation of aromatase expression in the breast by AMPK and CRTC2, in response to the altered adipokine milieu associated with obesity, provides an important link between obesity and breast cancer risk. Cancer Res; 70(1); 4–7
The concept of a relationship between dysregulated metabolism and carcinogenesis was first enunciated by Otto Warburg more than 80 years ago. This idea lay dormant for decades and was only revived relatively recently, following renewed interest in metabolism resulting from the developments in methodology which enabled studies of its regulation. An example of this concept is the relationship of obesity to breast cancer. There is now a body of evidence that supports a link between obesity, the metabolic syndrome, and insulin resistance with increased risk of several cancers including those of colon and the breast. Thus there is growing interest in exploring the possibility that antidiabetic therapies could decrease cancer incidence or cancer-related deaths. This possibility is highlighted by the fact that a number of studies have shown that the use of metformin, an oral antidiabetic drug that has been used for many years, is associated with decreased cancer risk (1).
There is growing evidence that the action of metformin is mediated primarily by stimulation of AMP-activated protein kinase (AMPK), in association with interaction with its upstream kinase LKB1. AMPK is now recognized to be a master regulator of energy homeostasis, stimulating pathways of energy production and inhibiting those of energy use (reviewed in ref. 2). Thus, for example, it stimulates glycolysis and fatty acid β-oxidation but inhibits gluconeogenesis as well as fatty acid and cholesterol biosynthesis. AMPK inhibits the expression of SREBP1c, which in turn leads to suppression of the expression of fatty acid synthase and acetyl-CoA carboxylase (ACC). AMPK also phosphorylates ACC leading to its inhibition and thus blocks the formation of malonyl-CoA, the first committed molecule in the pathway of fatty acid synthesis. Malonyl-CoA is also a powerful allosteric inhibitor of carnitine palmitoyl transferase-1. Decreased levels of malonyl-CoA thus lead to increased uptake of fatty acid residues by the mitochondria and thus increased β-oxidation.
According to a number of published studies, AMPK also markedly inhibits the proliferation of both malignant and nonmalignant cells (reviewed in ref. 3). One way in which it achieves this is by inhibition of lipid biosynthesis as outlined above, because rapidly proliferating cells require higher rates of de novo lipogenesis to support rapid membrane synthesis. However, AMPK also acts to inhibit the cell cycle as well as protein synthesis. AMPK stimulates expression of p53, which in turn stimulates expression of the cell cycle inhibitors p27Kip1 and p21Cip1, leading to inhibition of cell cycle progression. However AMPK also inhibits protein synthesis by inhibition of the mammalian target of rapamycin (mTOR) complex. AMPK forms a complex with sestrin 1 and 2 and TSC2 that promotes phosphorylation of TSC2 and activation of TSC1/TSC2 GAP activity toward Rheb, thereby inhibiting mTORC1, and consequently, cell growth and proliferation. Metformin has been shown to inhibit cell proliferation by these mechanisms, consistent with this role to activate AMPK (4).
So what, then, are the factors that link obesity to increased risk of cancer and, in particular, breast cancer? Two adipokines, leptin and adiponectin, have been examined in this regard. Leptin synthesis and plasma levels increase with obesity, and recent work has shown that higher leptin levels were significantly associated with an increase in breast cancer (5). Moreover there is a report that leptin stimulates aromatase expression in MCF-7 cells (6). By contrast adiponectin levels in the serum decrease with increased obesity, and three reported epidemiological studies of adiponectin have all shown an inverse association between serum adiponectin levels and breast cancer risk (7). Studies have also shown inhibition of growth of MCF-7 cells by adiponectin. Furthermore the adiponectin receptor AdipoR1 is highly expressed in human adipose tissue, and both AdipoR1 and AdipoR2 are expressed in MCF-7 and MDA-MB-231 breast cancer cells (8).
In a recent article, Brown and colleagues (9) present evidence for a third mechanism whereby AMPK can decrease breast cancer risk in postmenopausal women, namely by the inhibition of aromatase expression within the breast. Following the menopausal transition, the ovaries cease to make estrogen, yet the incidence of breast cancer increases with aging. A body of evidence exists to support the conclusion that it is local estrogen produced within the breast that drives breast cancer formation in postmenopausal women (reviewed in ref. 10). Estrogen formation is catalyzed by the cytochrome P450 enzyme aromatase, which is present in breast adipose stroma and epithelium. Inflammatory factors such as prostaglandin E2 stimulate aromatase expression in breast adipose mesenchymal cells via the proximal promoter, promoter II, by a pathway involving cyclic AMP and cAMP-responsive element binding protein (CREB), which binds to two CREs on the aromatase promoter II. The link between the LKB1/AMPK pathway and breast aromatase expression came from a study of boys with Peutz Jeghers syndrome, who, in addition to hamartoma tumors, developed florid gynecomastia at an early age (11). This development was consistent with the high rates of aromatase expression via promoter II in the Sertoli cell tumors in these boys. More recently it was shown that LKB1 is commonly mutated in Peutz Jeghers syndrome, immediately suggesting an association with aromatase expression (12).
Our study identifies three key findings that shed light on a possible mechanism by which obesity is linked to the increased incidence of estrogen-dependent breast cancer in postmenopausal women, via increased aromatase expression. Although they are overviewed here, the interested reader is encouraged to refer to the original publication for more details (9).
First, CREB-regulated transcription coactivator 2 (CRTC2) acts as a regulator of aromatase promoter PII activity. Endogenous binding of CRTC2 to promoter PII in primary human breast adipose stromal cells was shown using chromatin immunoprecipitation, and this interaction was shown to be accompanied by a significant increase in promoter activity. The importance of the coactivation of CREB by CRTC2 in the context of the transcriptional regulation of aromatase is highlighted by the fact that the activation of the majority of CREB-target genes has been shown to require the coactivation of CREB by CRTCs (13).
Secondly, the LKB1/AMPK pathway is inhibitory of aromatase expression in human breast adipose stromal cells, and this pathway is itself inhibited by tumor-derived inflammatory factors such as PGE2. Moreover AMPK directly phosphorylates CRTC2, which inhibits its activity by causing a phosphorylation-dependent interaction with 14-3-3 proteins, resulting in its sequestration in the cytoplasm (14). Thus PGE2 by inhibiting AMPK activity causes the nuclear translocation of CRTC2, leading to increased binding of CRTC2 to aromatase promoter PII, which contributes to an increase in aromatase expression. Most noteworthy, this phenomenon is prevented when AMPK is activated using AICAR or when LKB1 is overexpressed. This finding has significant therapeutic implications as activation of the LKB1/AMPK pathway in cases of estrogen-dependent breast cancer would theoretically be accompanied by a breast-specific inhibition of aromatase expression, circumventing adverse side-effects currently observed with the use of global phase III aromatase inhibitors.
Thirdly, the adipokines leptin and adiponectin act in opposing manners to regulate the LKB1/AMPK pathway in human breast adipose stromal cells. Although the exact mechanisms by which leptin and adiponectin govern changes in AMPK activity remain to be elucidated, we present evidence of the transcriptional regulation of LKB1 by these adipokines in human breast adipose stromal cells. Notably, we show that leptin downregulates LKB1 expression, accompanied by a decrease in AMPK phosphorylation, increased nuclear translocation of CRTC2, and a resulting increase in aromatase expression. As leptin expression and secretion are directly correlated with obesity, it raises the question of whether obese individuals with breast cancer have higher levels of aromatase expression and hence of estrogens within the adipose compared with their slim counterparts, and more specifically, whether this is sufficient to cause an increase in proliferation of breast epithelial cells. Conversely, adiponectin was shown to increase expression of LKB1, resulting in the increased phosphorylation of AMPK, thus preventing CRTC2 from entering the nucleus, and hence inhibiting aromatase expression.
The current so-called pandemic of obesity has the potential to lead to a corresponding increase in breast cancer particularly in elderly women, with the possibility that many more women may be at risk world-wide than was previously thought to be the case. Thus there is a clear and compelling need to understand the mechanisms whereby the spectrum of syndromes collectively known as the metabolic syndrome are associated with carcinogenesis and with breast cancer in particular. A substantial body of evidence, as outlined here, indicates that increased risk of carcinogenesis is associated with decreased activity of AMPK. This relationship, in turn, is associated with the pattern of adipokine synthesis and secretion occurring with increased adiposity, namely increased leptin and decreased adiponectin, although other factors including insulin and insulin-like growth factor-I (IGF-I) are doubtless also involved. As outlined here, evidence has been established that three independent mechanisms may be involved, namely increased lipogenesis; increased cellular proliferation due to stimulation of cell cycle activity and protein biosynthesis; and the stimulation of aromatase expression and hence estrogen production locally within the breast itself. Most obese individuals find it difficult or impossible to permanently reduce weight via diet and exercise. Thus therapeutic intervention may offer the best hope at least in the short term for preventing the obesity pandemic developing into a breast cancer epidemic. These results are summarized in Fig. 1.
Phase III aromatase inhibitors are proving highly successful as endocrine therapy for breast cancer in postmenopausal women, replacing tamoxifen as frontline, second line, and neoadjuvant therapy. However long-term contraindications such as bone loss, arthralgia, adverse cardiovascular effects, and possibly cognitive defects may limit the use of these compounds in the prevention setting. Moreover as outlined here, aromatase expression may be only one of several pathways whereby dysregulation of energy homeostasis influences breast cancer development. On the other hand metformin shows promise in breast cancer treatment and prevention on the basis of the results of observational and preclinical studies, which certainly support further investigations of its efficacy in this context. However the mechanisms whereby metformin act, and in particular how it stimulates AMPK, are poorly understood, and clearly there is a need to develop compounds that stimulate AMPK activity and/or expression with higher affinity and specificity. The problem is complicated by the number of subunits and isoforms of these subunits that together compose the AMPK molecule, and the fact that different combinations of these isoforms are present in different tissues. However, given the potential magnitude and significance of this problem worldwide, such studies are surely justified.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant Support: Victorian Breast Cancer Consortium Inc. and by Project Grant 494819 and Program Grant 494802 from the National Health and Medical Research Council (NH and MRC). K.A. Brown was supported by the Terry Fox Foundation through an award from the National Cancer Institute of Canada.
- Received June 18, 2009.
- Revision received September 21, 2009.
- Accepted September 23, 2009.
- ©2010 American Association for Cancer Research.