Skip to main content
NCBI home page
Molecular Endocrinology logoLink to Molecular Endocrinology
. 2009 Feb 12;23(5):662–670. doi: 10.1210/me.2008-0468

IKKβ Mediates Cell Shape-Induced Aromatase Expression and Estrogen Biosynthesis in Adipose Stromal Cells

Sagar Ghosh 1, Ahsan Choudary 1, Sangeeta Ghosh 1, Nicolas Musi 1, Yanfen Hu 1, Rong Li 1
PMCID: PMC2675949  PMID: 19221050

Abstract

Aromatase (Cyp19) is a key enzyme in estrogen biosynthesis and an important target in breast cancer therapy. Within tumor microenvironment, tumor cells stimulate aromatase expression in adipose stromal cells (ASCs), which in turn promotes estrogen-dependent growth of estrogen receptor (ER)-positive tumor cells. However, it is not clear how aromatase transcription and estrogen biosynthesis are regulated in ASCs under a precancerous condition. Here we demonstrate that cell shape change alone is sufficient to induce aromatase expression in primary ASCs from cancer-free individuals. The activation of aromatase transcription is mediated by IκB kinase-β (IKKβ), a kinase previously known for its cancer-promoting activity in tumor cells. Activation of IKKβ leads to elevated expression of transcription factor CCAAT/enhancer-binding protein-β (C/EBPβ), which binds to and stimulates two breast cancer-associated promoters of the aromatase gene. We also show that shape-induced estrogen production in ASCs can stimulate estrogen-dependent transcription in ER-positive breast tumor cells. We suggest that IKKβ-dependent aromatase induction due to changes in cellular architecture in adipose tissue may contribute to the breast cancer risks associated with high mammagraphic density and obesity.

Cell shape change induces IKKβ-dependent aromatase transcription and estrogen biosynthesis in human adipose stromal cells.

Excessive estrogen exposure is a well-recognized critical risk factor for breast cancer (1). Ovaries are the major sites for estrogen synthesis in premenopausal women, which explains why early menarche and late menopause are associated with elevated breast cancer incidence (2). In postmenopausal women, adipose tissue is a significant source for local and circulating estrogen (3,4). Estrogens produced in distal sc adipose tissue and within the breast tissue microenvironment influence the growth of breast epithelial cells in an endocrine and paracrine fashion, respectively (4). The rate of estrogen biosynthesis in adipose tissue correlates with the degree of adiposity. For example, it has been shown that estrogen production in the adipose tissue of overly obese postmenopausal women could be 10-fold higher than the lean controls (5). Aberrant estrogen production at least partially accounts for the well-documented, obesity-associated breast cancer risk among postmenopausal women (6,7).

Aromatase P450 (CYP19), which converts androgens to estrogens, catalyzes the final and rate-limiting step in estrogen biosynthesis. The human aromatase gene is expressed in a restricted number of tissue and cell types, including ovarian granulosa cells and adipose stromal cells (ASCs) (8). Transcription of the aromatase gene is controlled by several tissue-specific promoters that span over a 90-kb genomic region upstream of the coding exons of the aromatase gene (8,9). Although the same coding exons are present in all aromatase transcripts from different tissues, a noncoding exon 1 is transcribed in a tissue-specific manner and spliced to the common coding exons (10). Because aromatase abundance is a critical determinant for local and circulating estrogen levels, regulation of tissue-specific aromatase expression has a significant impact on a variety of estrogen target tissues under both physiological and pathological conditions.

Intratumor breast adipose tissue has a striking increase in aromatase expression and activity (11,12,13). This is largely due to aberrant activation of the otherwise dormant ovary-specific promoters of the aromatase gene in ASCs in response to tumor-secreted factors such as prostaglandin E2 and cytokines (14,15). The elevated local estrogen production in turn stimulates the growth of estrogen-dependent tumor cells, thus creating a vicious cycle within the tumor microenvironment. Aromatase inhibitors, such as letrozole, inhibit the enzymatic activity of aromatase and thus dampen the estrogenic capability throughout the body. The clinically proven efficacy of aromatase inhibitors in treating postmenopausal breast cancer unequivocally demonstrates the important role of extragonadally expressed aromatase in breast cancer development (16).

Despite the evidence for aromatase induction in adipose tissue in response to tumor stimuli, little is known about regulation of aromatase transcription and estrogen biosynthesis under a cancer-free condition. This is a clinically important question, because high mammagraphic density (17) and obesity (6,7), two well-documented risk factors for breast cancer, are associated with significant restructuring of stromal tissue. Here we demonstrate that cytoskeletal dynamics is a critical determinant of aromatase expression in primary ASCs that have no prior exposure to any tumor milieu. Intriguingly, cell shape change of ASCs from cancer-free individuals results in activation of the same aromatase promoters that are aberrantly up-regulated in breast tumors. In addition, we found a novel mechanism by which IκB kinase-β (IKKβ) regulates estrogen biosynthesis in a nuclear factor-β (NFκB)-independent pathway.

Results

Cell shape change markedly induces aromatase expression in ASCs

Using primary ASCs from cancer-free individuals, we made the initial observation that high confluency increased aromatase mRNA level up to 500-fold (Fig. 1A and supplemental Fig. S1A, published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org). This striking elevation in aromatase transcript correlated with significant increases in aromatase protein (Fig. 1B) and its enzymatic activity (Fig. 1C). As expected, the cell density-induced aromatase activity can be obliterated by letrozole, an aromatase inhibitor (Fig. 1C). ASCs can be differentiated into mature adipocytes under the adipogenic condition. However, the confluent ASC population did not stain positive for Oil Red O, a marker for mature adipocytes, nor did it express adipocyte-specific transcription factors such as CCAAT/enhancer-binding protein-α (C/EBPα) or peroxisome proliferator-activated receptor-γ (PPARγ) (supplemental Fig. S1, B and C). Therefore, the effect of cell confluency on aromatase expression was not a consequence of adipogenesis. Because the ASCs used in our study had not been exposed to any tumor milieu or changes in culture medium, our findings indicate that high cell density alone can markedly induce aromatase expression in ASCs without priming from breast tumor cells.

Figure 1.

Figure 1

Open in a new tab

Cell confluency significantly induces aromatase expression in ASCs. Comparison of the aromatase mRNA (A), protein levels (B), and enzymatic activity (C) in subconfluent (sub-conf) and confluent (conf) ASCs (d 12). The representative images of ASCs at the two cell densities are shown in A. Lamin A/C was used as the loading control in the immunoblotting. Aromatase inhibitor letrozole (0.1 μm) was used in the enzymatic assay in C to ascertain the specificity of the assay. Unless otherwise indicated, all error bars represent sd.

Confluency affects multiple cellular properties including cell shape and cell-cell contact. To test whether change of cell shape alone was sufficient to induce aromatase transcription, ASCs were treated with latrunculin A or jasplakinolide, two chemical compounds that alter actin-mediated cytoskeletal dynamics. As shown in Fig. 2, A and B, subconfluent cells that were treated with either drug underwent significant morphological changes. Importantly, these changes were accompanied by robust induction of aromatase expression. As an alternative to the chemical-induced cell shape change, subconfluent cells were plated on a plastic surface that minimizes cell adherence and hence confers the round cell morphology (Fig. 2C). Again, aromatase expression in the ASC suspension culture was markedly higher than their counterparts that grew on regular tissue culture plates (Fig. 2C). However, not all changes in cell shape and cytoskeletal dynamics gave rise to elevated aromatase gene expression. Disruption of microtubule polymerization by nocodazole caused a similar morphological change in ASCs, yet the nocodazole-treated cells did not show any appreciable changes in aromatase mRNA expression (Fig. 2D). Taken together, these results strongly suggest that actin-mediated cytoskeleton dynamics are critical to regulation of aromatase expression in ASCs.

Figure 2.

Figure 2

Open in a new tab

Cell shape change alone is sufficient to induce aromatase transcription. A and B, Treatment of subconfluent ASCs with actin-disrupting compounds latrunculin A (LatA; 1 μm) and jasplakinolide (Jasp; 0.5 μm) significantly increases aromatase mRNA levels. C, ASCs plated in low-attachment plates (low-attach) overnight undergo substantial morphological changes and elevation of aromatase expression. D, Flow cytometry indicating the cell cycle distribution of ASCs growing in regular, low-attachment plates or in the presence of nocodazole. Cell morphology and aromatase mRNA levels under various culture conditions in ASCs are shown. E, Cell shape change does not affect cell viability.

IKKβ, but not NFκB, mediates the effect of cell shape change on aromatase transcription

Detachment of anchorage-dependent epithelial cells from extracellular matrix in vivo or plastic stratum in vitro results in anoikis, i.e. detachment-induced apoptosis (18). However, no obvious cell death was observed in ASCs growing in low-attachment plates (Fig. 2E), suggesting that anoikis is an unlikely cause for the cell shape-induced aromatase expression. Cell detachment can activate the NFκB-dependent cell survival pathway, thus allowing certain normal and tumor cells to escape from anoikis (19). Interestingly, ASCs in low-attachment plates exhibited elevated activity of IKK (Fig. 3A), a key player in the NFκB-dependent pathway (19). Indeed, expression of several known NFκB target genes was up-regulated in the suspended ASC population (supplemental Fig. S2, A–C). To explore a potential causal relationship between the NFκB pathway and cell shape-induced aromatase expression, ASCs that underwent cell shape changes were treated simultaneously with an IKK-specific inhibitor, BAY11-7082. As shown in Fig. 3B, BAY11-7082 significantly dampened the jasplakinolide-induced aromatase transcription. Interestingly, the same IKK inhibitor also reversed the round cell morphology (compare the middle and right images in Fig. 3B). A similar inhibitory effect of BAY11-7082 was also observed for aromatase transcription induced by low attachment (Fig. 3C) and latrunculin A treatment (data not shown). Furthermore, shape-induced aromatase expression can be abolished by parthenolide, another IKK inhibitor that is structurally unrelated to BAY11-7082 (Fig. 3D).

Figure 3.

Figure 3

Open in a new tab

IKKβ is a critical mediator of the cell shape-induced aromatase transcription. A, IKKβ kinase activity is elevated after cell shape change. ASCs were plated in regular and low-attachment plates at different time intervals (o/n for overnight), and analyzed for the IKK kinase activity. Also shown is the quantitation of the intensity of the bands of the phosphorylated substrate upon normalization with the total protein amount. B and C, Effect of IKK inhibitor BAY11-7082 (BAY11) on cell shape-induced aromatase expression. ASCs were treated with jasplakinolide (Jasp; 0.5 μm) or cultured in ultra-low-attachment plates with or without BAY11. D, ASCs were also treated with or without parthenolide under low-attachment conditions, and aromatase mRNA expression was detected. E, The effects of IKKα, -β, and -γ knockdown on cell shape-induced aromatase transcription. Control and IKK knockdown cells were plated in low-attachment plates overnight, followed by mRNA analysis. F, Effects of WT and KD mutant IKKβ on aromatase gene expression. Lentivirus-infected ASCs were plated in regular or low-attachment plates overnight.

NFκB activity is regulated by a trimeric complex consisting of IKKα, IKKβ, and IKKγ (19). To further investigate the role of IKK in shape-induced aromatase expression, individual subunits of the IKK complex were knocked down by small interfering RNA (siRNA). Reduction of IKKβ expression significantly impaired aromatase transcription induced by low attachment (Fig. 3E), cell confluency (supplemental Fig. S2D), and latrunculin A (supplemental Fig. S2E). In contrast, knockdown of IKKα or IKKγ resulted in further increases in aromatase mRNA levels (Fig. 3E). In a reciprocal experiment, wild-type (WT) and kinase-dead (KD) IKKβ were ectopically expressed in ASCs. As expected, WT, but not mutant, IKKβ displayed potent kinase activity (supplemental Fig. S3A) and a stimulatory effect on known NFκB target genes such as IL-6 (supplemental Fig. S3B) and superoxide dismutase 2 (SOD2) (supplemental Fig. S3C). Importantly, ectopic expression of WT, but not mutant, IKKβ increased aromatase mRNA levels (Fig. 3F). Therefore, these results clearly demonstrate a critical role of IKKβ in mediating the cell shape-induced aromatase expression.

The fact that IKKα and IKKγ are not required for cell shape-induced aromatase expression suggests that the canonical NFκB pathway may be dispensable for cell shape-triggered aromatase induction. In most cell types, inactive NFκB is sequestered in the cytoplasm by its repressor IκB, and phosphorylation of IκB by the IKK complex triggers IκB degradation and subsequent nuclear translocation of NFκB (19). Ectopic expression of a nondegradable version of IκB (super-repressor), which constitutively sequesters NFκB in the cytoplasm (20), significantly repressed the expression of SOD2, a known NFκB target gene (Fig. 4A). In contrast, neither basal nor cell shape-induced aromatase mRNA levels were affected by the super-repressor (Fig. 4B). Consistent with this finding, an NFκB-specific peptide inhibitor (21) did not affect aromatase transcription in ASCs (data not shown). Therefore, the IKKβ-mediated aromatase induction is independent of the canonical NFκB pathway. In fact, it is possible that IKKα or IKKγ knockdown might have released free IKKβ that went on to activate aromatase transcription, which could explain the increased aromatase expression observed in the IKKα or IKKγ knockdown cells.

Figure 4.

Figure 4

Open in a new tab

Effect of IKKβ on aromatase transcription is NFκB independent. IκB super-repressor was overexpressed in ASCs, and cells were cultured in regular and low-attachment plate overnight. Overexpression of IκB super-repressor attenuated the shape-induced transcription of SOD2 (A) but not aromatase transcription (B).

Cell shape change induces two breast cancer-associated aromatase promoters

In normal adipose tissue, low levels of aromatase mRNA are mainly contributed by the relatively weak, adipose tissue-specific I.4 promoter (Fig. 5A) (4,8). In adipose tissue adjacent to breast tumor, induction of aromatase expression is largely due to aberrant activation of two proximal promoters, I.3 and PII, that are otherwise dormant in adipose tissue (Fig. 5A). The switch in promoter use is thought to be an important mechanism for breast cancer-associated aromatase expression. To determine which promoter was responsible for the cell shape-induced aromatase transcription, we measured the abundance of the alternative exon 1 that is specific for PII, I.3, and I.4. As shown in Fig. 5B, I.3- and PII-specific transcripts were significantly elevated in those cells that grew in low-attachment plates (lane 2), in the presence of latrunculin A (lane 4), and under confluent culture conditions (lanes 6 and 7). In contrast, levels of the I.4-specific transcript were not significantly affected under any of these cell shape-changing conditions.

Figure 5.

Figure 5

Open in a new tab

Cell shape change leads to activation of two breast cancer-associated promoters of the aromatase gene. A, Diagram illustrating the multiple tissue-specific promoters of the aromatase gene. The vertical solid bars represent the alternative noncoding exon 1 associated with various promoters. The two proximal promoters (1.3 and PII) are aberrantly activated in breast cancer-associated adipose tissue. B, Aromatase RNAs from ASCs after various treatments was subjected to semiquantitative RT-PCR using exon 1-specific primers. C, ASCs integrated with a I.3/PII-driven GFP reporter construct were plated at subconfluent, confluent, or low-attachment conditions. Both fluorescent (1–3) and phase-contrast (4–6) images are shown.

To determine whether the I.3 and PII promoter sequences were sufficient to confer the cell shape-induced transcriptional activation, a genomic DNA fragment containing both proximal promoters were subcloned into a green fluorescent protein (GFP) gene cassette. ASCs integrated with this GFP reporter fluoresced when seeded at high density or in low-attachment plates (panels 2 and 3 in Fig. 5C). The high inducibility of these two breast cancer-associated promoters in ASCs from cancer-free individuals underscores the importance of cell shape and cytoskeletal dynamics in aromatase expression and estrogen synthesis.

Up-regulated C/EBPβ after cell shape change activates aromatase transcription

Several known transcription factors can bind to I.3 and PII and regulate their promoter activity (4,8,22,23,24). Protein level of one of these factors, C/EBPβ, was significantly elevated in ASCs that grew in low-attachment plates or in high confluency (Fig. 6A). In subconfluent cells, chromatin immunoprecipitation (ChIP) did not detect any appreciable association of C/EBPβ with the proximal region of the endogenous I.3/PII promoters (lane 2 in Fig. 6B). However, C/EBPβ signal was detected at the promoter-proximal region of the aromatase gene in confluent ASCs (lane 4 in Fig. 6B). The cell shape-induced increase in C/EBPβ protein level was not accompanied by a corresponding change in its mRNA transcript (supplemental Fig. S4), suggesting that the up-regulation of C/EBPβ protein level was due to a posttranscriptional mechanism. To ascertain the role of C/EBPβ in cell shape-induced aromatase transcription, control and C/EBPβ- knockdown cells were plated in regular and low-attachment plates. As shown in Fig. 6C, depletion of C/EBPβ in ASCs abolished aromatase transcription. Given the role of IKKβ in cell shape-induced aromatase expression, we also examined the effect of ectopically expressed IKKβ on C/EBPβ protein expression. Expression of WT, but not KD, IKKβ resulted in a higher level of C/EBPβ protein (Fig. 6D), suggesting that IKKβ acts upstream of C/EBPβ in the cell shape-triggered pathway.

Figure 6.

Figure 6

Open in a new tab

Increased C/EBPβ protein level contributes to cell shape-induced aromatase transcription. A, ASCs were subjected to various treatments as indicated. Protein levels of various transcription factors were examined by immunoblotting. B, ChIP indicating enhanced association of C/EBPβ with the promoter proximal region of the aromatase gene in confluent ASCs. A genomic region 2 kb upstream of the aromatase transcription start site was included as negative control. C, C/EBPβ knockdown abolishes cell shape-induced aromatase expression. D, Ectopic expression of WT IKKβ enhances C/EBPβ protein expression.

Impact of ASCs on ER-positive breast tumor cells

To examine the impact of cell shape-induced aromatase transcription on breast cancer cells, we cultured ASCs under subconfluent or confluent conditions in the presence of testosterone, a substrate of aromatase. The ASC-conditioned medium was then added to ER-positive ZR75-1 breast cancer cells that were deprived of estrogen. We found that conditioned medium from confluent, but not subconfluent, ASCs considerably stimulated transcription of a known estrogen-responsive gene pS2 in the ER-positive breast cancer cells (compare lanes 3 and 5; Fig. 7A). Upon normalization with the total cell numbers in the two ASC populations, the medium that was conditioned with confluent ASCs gave rise to 5-fold larger amount of pS2 mRNA transcript than did the medium with subconfluent ASCs. Although testosterone itself did not have any direct impact on pS2 transcription in breast cancer cells (compare lanes 1 and 2), its omission from the ASC-culturing medium completely abolished the stimulatory effect of ASC-conditioned medium on estrogen-dependent transcription (compare lanes 5 and 6; Fig. 7A). Thus, cell shape-induced estrogen production in ASCs can significantly stimulate estrogen-dependent transcription in ER-positive breast cancer cells, without priming by any tumor cell-secreted stimuli.

Figure 7.

Figure 7

Open in a new tab

The impact of aromatase induction on estrogen-dependent transcription in ER-positive breast cancer cells. A, Subconfluent and confluent ASCs were cultured in the presence of 50 nm testosterone, a substrate of aromatase. The conditioned medium was added to estrogen-deprived, ER-positive ZR75-1 breast cancer cells. The mRNA levels of the estrogen-responsive gene pS2 were measured by quantitative RT-PCR. B, A model for cell shape-induced aromatase transcription.

Discussion

Aromatase in adipose tissue, which is largely responsible for estrogen production in postmenopausal women, is a proven therapeutic target for treating ER-positive postmenopausal breast cancer. In addition, excessive estrogen synthesis in adipose tissue is considered as a significant contributing factor to the obesity-associated breast cancer risk among postmenopausal women. Therefore, a comprehensive understanding of aromatase gene regulation in adipose tissue will go a long way to reduce ER-positive breast cancer incidence and mortality. Intratumor ASCs have strikingly elevated levels of aromatase as compared with those isolated from tumor-free areas (11,12,13), suggesting an impact of tumor cells on the estrogen biosynthetic capability of ASCs. Furthermore, aromatase expression in cultured ASCs can be significantly stimulated by tumor cell- secreted cytokines (14,15), again indicating the reliance of ASCs on tumor cells for activation of aromatase expression. Our results show that in the absence of any tumor milieu, cytoskeletal changes are sufficient to augment the estrogen biosynthetic capability of ASCs. In line with these findings, ASCs that undergo cell shape change greatly stimulate estrogen-dependent transcription in ER-positive breast cancer cells. Mechanistic investigation supports a model whereby actin-mediated cytoskeletal changes activate IKKβ, which in turn leads to elevated expression of C/EBPβ. Ultimately, more C/EBPβ that binds to the breast cancer-associated aromatase promoters substantially activates aromatase transcription (Fig. 7B).

It has been increasingly appreciated that tensional forces due to altered architecture in the tissue microenvironment can significantly influence gene expression pattern, the invasive behaviors of breast tumor cells, and breast cancer incidence/mortality (25,26). For example, actin-mediated cell shape change in human mammary epithelial cells results in global chromatin condensation and reduction in gene expression (27). Breast tumors, which grow in a mechanically constrained microenvironment, are associated with elevated mechanical tension (28). Tension-induced cell structural and morphological changes may also affect homeostasis in normal tissue. High mammagraphic breast density, a well-recognized risk factor for breast cancer, is at least in part due to excessive connective tissue and stroma in the breast (17). Likewise, increases in the number and size of mature adipocytes as in the case of obesity could conceivably impose substantial tensional stress on resident cells such as ASCs in adipose tissue. Therefore, aberrant tissue architecture in these in vivo and in vitro scenarios may have profound effects on the physical forms of ASCs and hence their capability to express aromatase. Our findings provide a reasonable molecular explanation for the elevated estrogen levels associated with obesity and breast cancer (6,7).

Overexpression or deregulated activity of IKKs is frequently associated with breast cancer (29). For example, a recent study indicates that activated IKKβ in tumor cells promotes inflammation and angiogenesis, thus contributing to poor prognosis of the disease (30). Interestingly, IKK has been previously implicated in mechanical stretch-induced IL-6 expression in endothelial cells (31). Whether breast tumor cells and ASCs share the same IKKβ-dependent pathway remains to be determined. However, it is tempting to speculate that IKKβ may serve as an important transducer of cytoskeletal signals that ultimately activates the local and systemic levels of estrogen and other inflammatory factors. In this regard, it will be of interest to determine whether aromatase expression in other estrogen-producing cell types can be significantly activated by the same cytoskeletal stimuli. Inhibition of IKKβ activity in both tumor and stromal compartments may provide an attractive therapeutic strategy to simultaneously inhibit the invasive behavior of the tumor cells and tumor-promoting capability of stromal cells.

C/EBPβ is associated with mammary gland development and breast cancer (32). In particular, previous studies have implicated C/EBPβ in specifically regulating aromatase transcription from the I.3 and PII promoters (33,34). Our results show that C/EBPβ is an important effector of the cell shape-induced, IKKβ-mediated signal transduction pathway. Our data also indicate that the surge in the C/EBPβ protein level after the shape change is likely due to posttranscriptional regulation. The exact mechanism for the cell shape-induced C/EBPβ expression remains to be elucidated. Although mammalian target of rapamycin (mTOR) has been implicated in the translational control of C/EBPβ through eukaryotic translation initiation factors eIF-2α and eIF-4E (35), our preliminary data indicate that inhibition of mTOR by rapamycin does not significantly affect cell shape-triggered elevation of C/EBPβ protein level (Ghosh, S., and R. Li, unpublished data). Several C/EBPβ isoforms that arise from alternative translation initiation sites can be detected in the cells. These isoforms display opposing function in gene activation and cell proliferation (36). It has been suggested that alterations in the ratio of different isoforms may contribute to breast cancer development (37). This is supported by a recent study that demonstrates a role of specific isoform of C/EBPβ in metastasis of breast cancer (38). In this regard, it will be of interest to compare the impact of different C/EBPβ isoforms on aromatase gene expression in ASCs.

Our work uncovers a previously unappreciated link between cell shape and hormone biosynthesis. It will be of interest to determine whether the degree of aromatase induction and estrogen synthesis varies in the cancer-free population. A related question concerns the possible individual-based variation in the expression levels of IKKβ and C/EBPβ and other yet to be identified players in the same signal transduction pathway in adipose tissue. Ultimately, identification of cancer-free individuals whose ASCs exhibit a high propensity of cell shape-triggered aromatase inducibility may aid in the assessment of breast cancer susceptibility. Furthermore, pharmacological targeting of the IKKβ-mediated signal transduction pathway in adipose tissue might be an effective strategy to reduce breast cancer incidence among the high-risk population.

Materials and Methods

Cell culture and treatment

Primary human ASCs were isolated from patients undergoing elective surgical procedures at the University of Virginia, using methods previously published and approved by the University of Virginia’s Human Investigation Committee (39). The cells were cultured in DMEM-F-12 medium with 10% fetal bovine serum (FBS) and antibiotic-antimycotic solution, using previously described methods (39). HEK293T and ZR75-1 cells were originally obtained from the American Type Culture Collection (Manassas, VA) and cultured per provider’s instruction.

Subconfluent and confluent ASC cultures were plated at 0.75 × 105 and 2 × 105 per well in six-well plates, respectively, and 5 × 105 cells were plated in 10-cm ultra-low-attachment plates (Corning Inc., Corning, NY). Latrunculin A (1 μm), jasplakinolide (0.5 μm), BAY11-7082 (5 μm), and parthenolide (10 μm; EMD Chemicals, Inc., Gibbstown, NJ) were applied to the ASCs overnight (∼16 h). To examine the effect of ASC-conditioned medium on estrogen-dependent transcription in ZR75-1 cells, ASCs were grown for 5 d, washed three times with PBS, and treated with or without testosterone (50 nm) for another 24 h in FBS-free medium. Concurrently, ZR75-1 cells were starved in estrogen-free medium for 3 d and subsequently cultured in ASC-conditioned medium for 24 h before harvest for RNA analysis.

Antibodies

The following antibodies were purchased from Cell Signaling Biotechnology (Beverly, MA): IKKα (2682), IKKγ (2685), β-actin (4967), and β-catenin (9562). The following antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA): lamin A/C (sc-7292), CREB (sc-186), c-Jun (sc-45), ATF-1 (sc-243), p65 (sc-372), C/EBPβ (sc-150), and TFIIH (sc-293). In addition, the following antibodies were used in the various experiments: α-tubulin (CP06; Calbiochem), α-aromatase (MCA 20775; Serotec, Raleigh, NC), and IKKβ (1601-1, Epitomics, Burlingame, CA).

Lentivirus-based expression vectors

WT and KD mutant IKKβ (gifts from Dr. M. Karin) and IκB super-repressor (S36A) (a gift from Dr. B. Chatterjee) were subcloned into a modified pWPI vector (acquired from Dr. D. Trono via Addgene, Cambridge, MA) between NdeI and PmeI. To generate the I.3/PII-driven GFP reporter construct, the EF1-α promoter of the pWPXLd (Addgene) vector was replaced with a 1-kb genomic DNA fragment upstream of the aromatase gene. Lentiviruses were generated and amplified in 293T cells by cotransfection with the above-mentioned constructs and packaging vectors pMD2.G and psPAX2 (Addgene).

Aromatase activity assay

Aromatase activity assays were performed as previously described (40).

Immunoblotting

ASCs were harvested and lysed in Laemmli buffer with a cocktail of protease inhibitors. The total protein concentrations were quantified by the BCA protein assay (Pierce, Rockford, IL). An equal amount of total proteins was resolved by SDS-PAGE, and Western blottings were performed using specific antibodies.

Quantitative and semiquantitative RT-PCR

Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. RNA was reverse-transcribed using the ImPrompII kit (Promega, Madison, WI). Real-time PCR was carried out using the fluorescent dye SYBR-Green and an ABI 7900 Real-Time PCR System (Applied Biosystems, Foster City, CA). Semiquantitative PCR was performed using pairs of primers for different tissue-specific transcripts (see supplemental Table S1). GAPDH mRNA was used for normalization throughout the study. PCR condition was as follows: 94 C for 2 min, 94 C for 30 sec, 55 C for 30 sec, 72 C for 1 min (35 cycles), 72 C for 10 min, and 4 C overnight.

IKK activity assay

IKK activity assay was performed following previously published work (41) with modifications. Briefly, cells (1.5 × 106) were lysed in 250 μl lysis buffer [20 mm Tris (pH 7.5), 5 mm EDTA, 10 mm Na3PO4, 100 mm NaF, 2 mm Na3VO4, 1% Nonidet P-40, 10 μm leupeptin, 3 mm benzamidine, 10 μg/ml aprotinin, and 1 mm phenylmethylsulfonyl fluoride]. Cell lysates (250 μg) were immunoprecipitated with IKKβ antibody overnight at 4 C. Immunoprecipitates were washed twice in lysis buffer and twice with wash buffer [50 mm Tris (pH 7.4), 0.5 m NaCl, and 0.1 mm EGTA]. IKK activity was measured by adding 50 mm Tris (pH 7.4), 4 μg recombinant GST-IκB (UW9970; Biomol, Plymouth Meeting, PA), 1 mm dithiothreitol, 10 mm MgCl2, 0.1 mm EGTA, and 0.1 mm ATP (5 μCi [γ -32P] ATP) at 30 C for 40 min. The extracts were mixed with 4× Laemmli buffer, and equal amounts of samples were loaded into a 10% SDS-PAGE gel. The gel was stained overnight with Coomassie Blue and dried, and the radioactive intensity was measured using a phosphorimager.

siRNA

Gene-specific knockdown by siRNA oligonucleotides were conducted using Lipofectamine reagent RNAiMAX (Invitrogen). The knockdown experiment was performed as previously described (42). Briefly, cells at a density of 60% were first transfected with siRNA oligos at a final concentration of 10 nm. For the comparison of subconfluent and confluent cells, cells were trypsinized 24 h after transfection and plated in subconfluent and confluent densities. Cells were harvested for RNA analysis 72 h after plating. For low-attachment study, knockdown cells were trypsinized 72 h after transfection and replated in low-attachment plates overnight before harvesting. IKKα (M-003473-01-0005), IKKβ (M-003503-00-0005), IKKγ (M-003767-00-0005), C/EBPβ (M-006423-03), and control (D-001810-10) siRNA oligos were purchased from Dharmacon, Inc. (Lafayette, CO).

ChIP

ChIP was carried out as previously described (43). In short, ASCs were cross-linked with 1% formaldehyde in growth medium for 10 min at room temperature, followed by addition of glycine to terminate the cross-linking reaction. Cells were harvested in cell lysis buffer [5 mm HEPES (pH 7.9), 85 mm KCl, and 0.5% Triton X-100] and incubated on ice for 15 min. Cells were pelleted and resuspended in nuclei lysis buffer [50 mm Tris-HCl (pH 8.0), 10 mm EDTA (pH 8.0), and 1% SDS]. Chromatin was solubilized and sheared by pulse sonication in Bioruptor XL (Diagenode, Sparta, NJ), and clarified by centrifugation at 15,000 rpm for 30 min. Chromatin-containing fractions were diluted 10-fold in dilution buffer [50 mm Tris-HCl (pH 8.0), 2 mm EDTA (pH 8.0), 150 mm NaCl, and 1% Triton X-100], and 10% of the chromatin mixture was set aside as an input control. Material equivalent to 5 × 106 cells was used per immunoprecipitation with 2 μg IgG or the α-C/EBPβ antibody overnight at 4 C. Immunoprecipitates were processed as described (43). Primers used for the PCR amplification are listed in supplemental Table S1.

Flow cytometry analysis

For cell cycle analysis, ASCs were grown on regular or low-attachment plates, or treated with 100 ng/ml nocodazole for 24 h before harvesting for the flow cytometry analysis. Cells were harvested in PBS with 1% FBS and 1% sodium azide and were fixed in 1 ml ice-cold ethanol. Cell pellet was resuspended in 500 μl PBS that contains 25 μg/ml propidium iodide and 25 μg/ml ribonuclease A, incubated in 37 C for 30 min, and subjected to flow cytometry analysis.

Cell viability assay

ASCs were plated in either regular or low-attachment tissue culture plates overnight. Cells in regular plates were trypsinized and replated onto fresh regular plates. Unattached cells in low-attachment plates were carefully collected and replated onto regular plates. The replated cells were then harvested at various time points and stained with trypan blue solution (T8154; Sigma-Aldrich, St. Louis, MO). Dye-excluding cells were considered alive. Cell viability after reattachment was calculated as such: (no. of living cells/total no. of cells originally plated) ×100.

Supplementary Material

[Supplemental Data]
me.2008-0468_index.html (2.7KB, html)

Acknowledgments

We thank Drs. M. Karin, D. Trono, and B. Chatterjee for plasmids; Novartis Pharmaceutical for letrozole; and P. Hornsby for critical reading of the manuscript.

Footnotes

The work was supported by grants to R.L. (CA93506) and Y.H. (CA118578) from the National Institutes of Health.

Disclosure Summary: The authors have nothing to declare.

First Published Online February 12, 2009

Abbreviations: ASC, Adipose stromal cell; C/EBPα, CCAAT/enhancer-binding protein-α; ChIP, chromatin immunoprecipitation; FBS, fetal bovine serum; GFP, green fluorescent protein; IKKβ, IκB kinase-β; KD, kinase-dead; NFκB, nuclear factor-κB; siRNA, small interfering RNA; SOD2, superoxide dismutase 2; WT, wild type.

References

  1. Nass SJ, Davidson NE 1999 Advance in breast cancer therapy. Hematol Oncol Clin North Am 13:311–332 [DOI] [PubMed] [Google Scholar]
  2. Wooster R, Weber BL 2003 Breast and ovarian cancer. N Engl J Med 348:2339–2347 [DOI] [PubMed] [Google Scholar]
  3. Simpson ER, Clyne C, Rubin G, Boon WC, Robertson K, Britt K, Speed C, Jones M 2002 Aromatase: a brief overview. Annu Rev Physiol 64:93–127 [DOI] [PubMed] [Google Scholar]
  4. Bulun SE, Lin Z, Imir G, Amin S, Demura M, Yilmaz B, Martin R, Utsunomiya H, Thung S, Gurates B, Tamura M, Langoi D, Deb S 2005 Regulation of aromatase expression in estrogen-responsive breast and uterine disease: from bench to treatment. Pharmacol Rev 57:359–383 [DOI] [PubMed] [Google Scholar]
  5. Grodin JM, Siiteri PK, MacDonald PC 1973 Source of estrogen production in postmenopausal women. J Clin Endocrinol Metab 36:207–214 [DOI] [PubMed] [Google Scholar]
  6. Stoll BA 2002 Upper abdominal obesity, insulin resistance and breast cancer risk. Int J Obes Relat Metab Disord 26:747–753 [DOI] [PubMed] [Google Scholar]
  7. Rose DP, Komninou D, Stephenson GD 2004 Obesity, adipocytokines, and insulin resistance in breast cancer. Obes Rev 5:153–165 [DOI] [PubMed] [Google Scholar]
  8. Simpson ER, Davis SR 2001 Aromatase and the regulation of estrogen biosynthesis: some new perspectives. Endocrinology 142:4589–4594 [DOI] [PubMed] [Google Scholar]
  9. Kamat A, Hinshelwood MM, Murry BA, Mendelson CR 2002 Mechanisms in tissue-specific regulation of estrogen biosynthesis in humans. Trends Endocrinol Metab 13:122–128 [DOI] [PubMed] [Google Scholar]
  10. Bulun SE, Sebastian S, Takayama K, Suzuki T, Sasano H, Shozu M 2003 The human CYP19 (aromatase P450) gene: update on physiologic roles and genomic organization of promoters. J Steroid Biochem Mol Biol 86:219–224 [DOI] [PubMed] [Google Scholar]
  11. Bulun SE, Price TM, Aitken J, Mahendroo MS, Simpson ER 1993 A link between breast cancer and local estrogen biosynthesis suggested by quantification of breast adipose tissue aromatase cytochrome P450 transcripts using competitive polymerase chain reaction after reverse transcription. J Clin Endocrinol Metab 77:1622–1628 [DOI] [PubMed] [Google Scholar]
  12. Harada N 1997 Aberrant expression of aromatase in breast cancer tissues. J Steroid Biochem Mol Biol 61:175–184 [PubMed] [Google Scholar]
  13. Sasano H, Harada N 1998 Intratumoral aromatase in human breast, endometrial, and ovarian malignacies. Endocr Rev 19:593–607 [DOI] [PubMed] [Google Scholar]
  14. Meng L, Zhou J, Sasano H, Suzuki T, Zeitoun KM, Bulun SE 2001 Tumor necrosis factor α and interleukin 11 secreted by malignant breast epithelial cells inhibit adipocyte differentiation by selectively down-regulating CCAAT/enhancer binding protein α and peroxisome proliferator-activated receptor γ: mechanism of desmoplastic reaction. Cancer Res 61:2250–2255 [PubMed] [Google Scholar]
  15. Zhou J, Gurates B, Yang S, Sebastian S, Bulun SE 2001 Malignant breast epithelial cells stimulate aromatase expression via promoter II in human adipose fibroblasts: an epithelial-stromal interaction in breast tumors mediated by CCAAT/enhancer binding protein β. Cancer Res 61:2328–2334 [PubMed] [Google Scholar]
  16. Morales L, Neven P, Paridaens R 2005 Choosing between an aromatase inhibitor and tamoxifen in the adjuvant setting. Curr Opin Oncol 17:559–565 [DOI] [PubMed] [Google Scholar]
  17. Boyd NF, Guo H, Martin LJ, Sun L, Stone J, Fishell E, Jong RA, Hislop G, Chiarelli A, Minkin S, Yaffe MJ 2007 Mammographic density and the risk and detection of breast cancer. N Engl J Med 356:227–236 [DOI] [PubMed] [Google Scholar]
  18. Frisch SM, Screaton RA 2001 Anoikis mechanisms. Curr Opin Cell Biol 13:555–562 [DOI] [PubMed] [Google Scholar]
  19. Perkins ND 2007 Integrating cell-signalling pathways with NF-κB and IKK function. Nat Rev Mol Cell Biol 8:49–62 [DOI] [PubMed] [Google Scholar]
  20. Wang CY, Mayo MW, Baldwin Jr AS 1996 TNF- and cancer therapy-induced apoptosis: potentiation by inhibition of NF-κB. Science 274:784–787 [DOI] [PubMed] [Google Scholar]
  21. Hodgson L, Henderson AJ, Dong C 2003 Melanoma cell migration to type IV collagen requires activation of NF-κB. Oncogene 9:98–108 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Chen S, Ye J, Kijima I, Kinoshita Y, Zhou D 2005 Positive and negative transcriptional regulation of aromatase expression in human breast cancer tissue. J Steroid Biochem Mol Biol 95:17–23 [DOI] [PubMed] [Google Scholar]
  23. Kijima I, Ye J, Glackin C, Chen S 2008 CCAAT/enhancer binding protein δ up-regulates aromatase promoters I.3/II in breast cancer epithelial cells. Cancer Res 68:4455–4464 [DOI] [PubMed] [Google Scholar]
  24. Ishikawa H, Fencki V, Marsh EE, Yin P, Chen D, Cheng YH, Reisterd S, Lin Z, Bulun SE 2008 CCAAT/enhancer binding protein β regulates aromatase expression via multiple and novel cis-regulatory sequences in uterine leiomyoma. J Clin Endocrinol Metab 93:981–991 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Paszek MJ, Weaver VM 2004 The tension mounts: mechanics meets morphogenesis and malignancy. J Mammary Gland Biol Neoplasia 9:325–342 [DOI] [PubMed] [Google Scholar]
  26. Ingber DE 2008 Can cancer be reversed by engineering the tumor microenvironment? Semin Cancer Biol 18:356–264 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Le Beyec J, Xu R, Lee SY, Nelson CM, Rizki A, Alcaraz J, Bissell MJ 2007 Cell shape regulates global histone acetylation in human mammary epithelial cells. Exp Cell Res 313:3066–3075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Paszek MJ, Zahir N, Johnson KR, Lakins JN, Rozenberg GI, Gefen A, Reinhart- King CA, Margulies SS, Dembo M, Boettiger D, Hammer DA, Weaver VM 2005 Tensional homeostasis and the malignant phenotype. Cancer Cell 8:241–254 [DOI] [PubMed] [Google Scholar]
  29. Karin M 2008 The IκB kinase: a bridge between inflammation and cancer. Cell Res 18:334–342 [DOI] [PubMed] [Google Scholar]
  30. Lee DF, Kuo HP, Chen CT, Hsu JM, Chou CK, Wei Y, Sun HL, Li LY, Ping B, Huang WC, He X, Hung JY, Lai CC, Ding Q, Su JL, Yang JY, Sahin AA, Hortobagyi GN, Tsai FJ, Tsai CH, Hung MC 2007 IKKβ suppression of TSC1 links inflammation and tumor angiogenesis via the mTOR pathway. Cell 130:440–455 [DOI] [PubMed] [Google Scholar]
  31. Kobayashi S, Nagino M, Komatsu S, Naruse K, Nimura Y, Nakanishi M, Sokabe M 2003 Stretch-induced IL-6 secretion from endothelial cells requires NF-κB activation. Biochem Biophys Res Commun 308:306–312 [DOI] [PubMed] [Google Scholar]
  32. Grimm SL, Rosen JM 2003 The role of C/EBPβ in mammary gland development and breast cancer. J Mammary Gland Biol Neoplasia 8:191–204 [DOI] [PubMed] [Google Scholar]
  33. Chen S, Zhou D, Yang C, Okubo T, Kinoshita Y, Yu B, Kao YC, Itoh T 2001 Modulation of aromatase expression in human breast tissue. J Steroid Biochem Mol Biol 79:35–40 [DOI] [PubMed] [Google Scholar]
  34. Deb S, Zhou J, Amin SA, Imir AG, Yilmaz MB, Lin Z, Bulun SE 2006 A novel role of sodium butyrate in the regulation of cancer-associated aromatase promoters I.3 and II by disrupting a transcriptional complex in breast adipose fibroblasts. J Biol Chem 281:2585–2597 [DOI] [PubMed] [Google Scholar]
  35. Calkhoven CF, Müller C, Leutz A 2000 Translational control of C/EBPα and C/EBPβ isoform expression. Genes Dev 14:1920–1932 [PMC free article] [PubMed] [Google Scholar]
  36. Descombes P, Schibler U 1991 A liver-enriched transcriptional activator protein, LAP, and a transcriptional inhibitory protein, LIP, are translated from the same mRNA. Cell 67:569–579 [DOI] [PubMed] [Google Scholar]
  37. Zahnow CA 2002 CCAAT/enhancer binding proteins in normal mammary development and breast cancer. Breast Cancer Res 4:113–121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Gomis RR, Alarcón C, Nadal C, Van Poznak C, Massagué J 2006 C/EBPβ at the core of the TGFβ cytostatic response and its evasion in metastatic breast cancer cells. Cancer Cell 10:203–214 [DOI] [PubMed] [Google Scholar]
  39. Katz AJ, Tholpady A, Tholpady SS, Shang H, Ogle RC 2005 Cell surface and transcriptional characterization of human adipose-derived adherent stromal (hADAS) cells. Stem Cells 23:412–423 [DOI] [PubMed] [Google Scholar]
  40. Yue W, Brodie AM 1997 Mechanisms of the actions of aromatase inhibitors 4-hydroxyandrostenedione, fadrozole, and aminoglutethimide on aromatase in JEG-3 cell culture. J Steroid Biochem Mol Biol 63:317–328 [DOI] [PubMed] [Google Scholar]
  41. Reyna SM, Ghosh S, Tantiwong P, Meka CS, Eagan P, Jenkinson CP, Cersosimo E, Defronzo RA, Coletta DK, Sriwijitkamol A, Musi N 2008 Elevated toll-like receptor 4 expression and signaling in muscle from insulin-resistant subjects. Diabetes 57:2595–2602 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Hu YF, Ghosh S, Amleh A, Yue W, Lu Y, Katz A, Li R 2005 Modulation of aromatase expression by BRCA1: a possible link to tissue-specific tumor suppression. Oncogene 24:8343–8348 [DOI] [PubMed] [Google Scholar]
  43. Ding N, Zhou H, Esteve PO, Chin HG, Kim S, Xu X, Joseph SM, Friez MJ, Schwartz CE, Pradhan S, Boyer TG 2008 Mediator links epigenetic silencing of neuronal gene expression with X-linked mental retardation. Mol Cell 31:347–359 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[Supplemental Data]
me.2008-0468_index.html (2.7KB, html)
me.2008-0468_1.pdf (276.1KB, pdf)
me.2008-0468_2.pdf (55.7KB, pdf)
me.2008-0468_3.pdf (44.2KB, pdf)
me.2008-0468_4.pdf (37.7KB, pdf)
me.2008-0468_5.pdf (48KB, pdf)

Articles from Molecular Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES