Keywords
estrogens - obesity - inflammation - breast cancer
Our work on breast cancer concerns the role of local aromatase expression in the breast
as the source of estrogen driving breast cancer development in the postmenopausal
woman.[1]
[2] Following the menopausal transition, the ovaries cease to make estrogens, which
then becomes the responsibility of extragonadal sites such as breast, bone, and brain.
Work from our laboratory and others indicates that estrogen produced in these sites
acts locally rather than systemically in a paracrine and intracrine fashion.[3]
[4] Thus, in the case of the breast, our goal has been to define the mechanisms whereby
aromatase expression is regulated within the breast, with the translational goal of
developing new breast-specific inhibitors of aromatase expression as improved endocrine
therapy for breast cancer treatment and prevention.
Aromatase in the Breast, Aging, and Breast Cancer
Within the breast, aromatase expression occurs in the adipose mesenchymal cells and
increases fourfold in the presence of a tumor.[3] This occurs in conjunction with differential promoter usage such that the gonadal
proximal promoter PII dominates relative to the promoter I.4.[5]
[6]
[7] We and others have shown that this is because inflammatory factors such as prostaglandin
E2 produced by the tumorous epithelium activate aromatase expression in the surrounding
breast adipose stromal cells via the EP2 receptor, which results in stimulation of
adenylyl cyclase, and the EP1 receptor, which stimulates DAG and IP3 formation.[8]
[9] It is also an absolute requirement for a monomeric orphan member of the nuclear
receptor family to bind to a nuclear receptor half site downstream of the CREs on
the aromatase promoter PII, namely, LRH-1.[10]
[11] In addition, we have shown that a powerful coactivator of LRH-1 is PGC1α[12] and that the expression of PGC1α is stimulated 10-fold by the cyclic AMP pathway.
Hence, these stimulatory pathways work in concert to facilitate tumor-driven aromatase
expression within the breast mesenchyma, providing an important example of epithelial–mesenchymal
interactions working to facilitate tumor development ([Fig. 1]).
Fig. 1 Diagram of the role of PGE2 in the regulation of aromatase expression in human breast
adipose stromal cells.
Our recent major interest is to understand the link between obesity and breast cancer.
There is now substantial epidemiological evidence to support the conclusion that obesity
is linked to the increased risk of several forms of cancer such as colon, endometrial,
and breast cancer.[13]
[14] Given the obesity problem worldwide, the potential significance of this conclusion
is that tens of millions more women may develop breast cancer in their senior years
than was previously believed to be the case. The problem is compounded by the fact
that breast cancer risk increases with aging. In postmenopausal life, this appears
to be due primarily to an increased capacity of adipose tissue to synthesize estrogens
as a function of age[15]
[16]; however, the mechanism of this increase is not entirely understood at this time.[17]
[18] While it is facile to say that obesity may be reversed or prevented by healthy diet
and exercise, most individuals who are obese find it difficult to achieve permanent
loss of weight by these methods; hence, other therapeutic interventions are required
to stave off a global epidemic of breast cancer arising from the obesity pandemic.
At this time, the cellular and molecular mechanisms underlying the increased risk
of breast cancer associated with obesity and aging are incompletely understood. The
AMP-activated protein kinase (AMPK) is now recognized to be a master regulator of
energy homeostasis and the nexus for the convergence of endocrine signals including
leptin, adiponectin, estradiol, androgens, and phytoestrogens.[19]
[20]
[21] AMPK activity is regulated primarily through phosphorylation at T172(α) by the upstream
kinases LKB-1 and CaMKK; however, in most tissues LKB-1 appears to predominate. Furthermore,
phosphorylation of the α-catalytic subunit of AMPK at S485 (α1) or S491 (α2) by PKA
reduces its catalytic activity.[22] The possibility of a link between the LKB-1/AMPK pathway and aromatase expression
in the breast arose from an unexpected source, namely, the rare condition of Peutz–Jeghers
syndrome. Boys with this condition develop florid gynecomastia by the age of 6 or
7 years associated with the formation of Sertoli cell tumors. We studied such tumors
some years ago and showed that they have very high rates of aromatase expression driven
by promoter II thus explaining the gynecomastia in these boys.[23] The link with the LKB-1/AMPK pathway was revealed when it was shown that Peutz–Jeghers
syndrome was due to mutations in the LKB-1 gene.[24]
Recently, a new family of CREB coactivators called CREB-regulated transcription coactivators
(CRTCs; previously known as TORCs) has been shown to increase the expression of cyclic
AMP responsive genes. When AMPK is active, CRTCs are sequestered in the cytoplasm
due to phosphorylation by AMPK on S171 and binding to 14–3-3. In the absence of AMPK
activity, CRTC2 is dephosphorylated and translocates to the nucleus where it associates
with CREB and increases target gene expression.[25] We have shown that this is true in the case of aromatase in breast stromal cells.
This provides a mechanism whereby the LKB-1/AMPK pathway can inhibit expression of
aromatase in the breast. Furthermore, we have shown that leptin stimulates and adiponectin
inhibits aromatase expression in breast mesenchymal cells via this pathway. In the
case of leptin, this is associated with translocation of CRTC2 to the nucleus and
binding to the aromatase promoter PII, whereas with adiponectin CRTC2 is retained
in the cytoplasm and its binding to the promoter is decreased ([Fig. 2]).[26] Consistent with this, we have shown that the CRTC2 mutant S171A remains in the nucleus
and stimulates aromatase PII activity, whereas the S171D mutant remains in the cytoplasm
and cannot stimulate PII activity. (S171 is the site of phosphorylation by AMPK.)
We have also observed that Fsk/PMA treatment (to mimic PGE2) of these cells results
in phosphorylation of α1S485/α2S491 residues in AMPK, which results in inhibition
of AMPK activity. This is in turn associated with translocation of CRTC2 to the nucleus
and binding to the PII promoter ([Fig. 3]).
Fig. 2 Action of leptin to stimulate, and adiponectin to inhibit, aromatase expression in
human breast adipose stromal cells.
Fig. 3 Diagram of the role of obesity to stimulate breast cancer proliferation.
Moreover, we have shown that the ratio of estradiol to testosterone also regulates
the LKB1/AMPK pathway in adipose tissue.[19] In mouse adipose tissue and in 3T3L1 cells, testosterone or dihydrotestosterone
inhibits, and estradiol stimulates, LKB1 expression, and ERα binds to the LKB1 promoter
in the presence of estradiol.[19] Thus, we predict that when the circulating ratio of testosterone to estradiol is
high, as in the postmenopausal years, then aromatase expression in the breast is stimulated.
Obesity, Inflammation, and Breast Cancer
With obesity now recognized to be an inflammatory condition, research activity has
turned to the role of inflammatory mediators as drivers of aromatase expression in
the breast and breast cancer risk. PGE2 is such a mediator and as indicated above
is a major driver of aromatase expression in human breast adipose stromal cells. We
have previously shown that inflammatory cytokines also drive aromatase expression
in these cells, notably IL-6 and TNFα. Recent work from Dannenberg's group[27] has shown that the lipid-laden adipocytes in the breasts of obese women are frequently
surrounded by macrophages, indicative of an inflammatory condition. Moreover, they
showed that aromatase expression in the breasts of these women was higher than in
nonobese women, and correlated with increased levels of breast COX2 (the first enzyme
on the pathway to prostaglandins including PGE2) and also PGE2 itself. In recent work,
we have shown that PGE2 acts in a similar fashion to leptin to inhibit the LKB1/AMPK
pathway, presumably by activation of PKA as indicated above. Thus, a second mechanism
whereby PGE2 stimulates aromatase expression is provided.[26]
AMPK—The Master Regulator
Thus, we see that AMPK plays a central role in the mechanisms whereby inflammation
and obesity regulate aromatase expression in the breast, namely, as an inhibitory
factor. These results immediately provide a new and commanding explanation for the
link between obesity, aging, and breast cancer risk. First, obesity, whether premenopausal
or age-related, results in a decrease in circulating adiponectin and increase in leptin.
Second, postmenopausally when the ovaries cease to make estrogen, the ratio of testosterone
to epitestosterone increases. Third, obesity results in the formation of inflammatory
mediators, notably PGE2. Each of these would in turn result in a decrease in activity
of the LKB1/AMPK pathway in breast adipose, resulting in increased expression of aromatase.
The resulting increase in estrogen formation in the breast would lead to increased
breast cancer proliferation ([Fig. 3]).
Hence, factors that stimulate AMPK have the potential to be a new generation of breast
cancer therapeutics. Moreover, since promoter II drives aromatase expression in the
breast and promoter I.4 drives expression in bone, and AMPK is inhibitory solely of
promoter II–driven aromatase expression but not promoter I.4–driven expression, such
factors should be breast-specific. Thus, they would not inhibit aromatase in other
sites where estrogens have important roles, such as bone, and thus should not give
rise to the contraindications, such as arthralgia and bone loss, which cause many
women to abandon endocrine therapy for breast cancer. One such factor is the antidiabetic
drug metformin, which acts by stimulating AMPK and which we have shown is inhibitory
of PGE2-stimulated aromatase expression in adipose stromal cells.[28] Several studies have indicated that metformin is protective against breast cancer
and inhibits the growth of breast cancer cells in culture. However, the action of
metformin to stimulate AMPK is unclear and is indirect, so there is currently much
interest to develop specific agonists of AMPK.
