K. Sonne-Hansen, A.E. Lykkesfeldt1
Department of Tumor Endocrinology, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49, DK-2100 Copenhagen, Denmark.
1A.E. Lykkesfeldt; E-mail: al@cancer.dk.
Abstract
Estrogens produced within breast tumors may play a pivotal role in growth stimulation of the breast cancer cells. However, it is elusive whether the epithelial breast cancer cells themselves synthesize estrogens, or whether the surrounding tumor stromal cells synthesize and supply the cancer cells with estrogen. The aromatase enzyme catalyzes the estrogen production, aromatizing circulating androgens into estrogens. The aim of this study was to investigate aromatase expression and function in a model system of human breast cancer, using the estrogen responsive human MCF-7 breast cancer cell line. Cells were cultured in a low estrogen milieu and treated with estrogens, aromatizable androgens or non-aromatizable androgens. Cell proliferation, expression of estrogen-regulated proteins and aromatase activity were investigated. The MCF-7 cell line was observed to express sufficient aromatase enzyme activity in order to aromatize the androgen testosterone, resulting in a significant cell growth stimulation. The testosterone-mediated growth effect was completely inhibited by the aromatase inhibitors letrozole and 4-hydroxy-androstenedione. Expression studies of estrogen-regulated proteins confirmed that testosterone was aromatized to estrogen in the MCF-7 cells. Thus, the results indicate that epithelial breast cancer cells possess the ability to aromatize circulating androgens to estrogens.
Keywords: Aromatase; Aromatase inhibitor; Breast cancer; Estrogen receptor; Androgen receptor; Progesterone receptor; Estrogen depletion; MCF-7.
Introduction
Estrogens are known to play a pivotal role in the devel-opment and promotion of human breast cancer. The highest frequency of breast cancer is found among postmenopausal women. These women have low levels of circulating estrogens, however, local synthesis of estrogens takes place in peripheral tissues, including the breast [1]. Estrogens are synthesized via aromatization of circulating C19 androgens, a process catalyzed by the aromatase enzyme. For the last decades, antiestrogen therapy with tamoxifen has been the preferred treatment of estrogen responsive breast cancer. However, there is a shift towards treatment with aromatase inhibitors (AI), after third-generation inhibitors have shown superiority to tamoxifen [2-7]. However, as for other endocrine treatments, many patients with advanced disease will develop resistance to treatment after a period with response [6]. The mechanisms responsible for development of acquired resistance are elusive and thus, studies are urgently required to fully understand the significance of local aromatase expression in breast cancer and the cellular and molecular consequences of AI treatment.
The aromatization of androgens has been observed to be more pronounced in malignant than normal breast tis-sues [8-10] and a locally increased estrogen production may stimulate proliferation of estrogen responsive breast cancer cells. It is still elusive which cell types that are responsible for intratumoral estrogen production, as aromatase protein has been immunohistochemically detected in both epithelial breast cancer cells and surrounding tumor stromal cells [11-15]. Further, in vitro cell culture studies have shown aromatase activity in both epithelial breast cancer cell lines [16-19] and breast-derived stromal fibroblasts from normal breast tissue and breast tumors [20-23]. The basal aromatase activity in both cell types is low, but the activity in fibroblasts can be highly stimulated with factors as IL-6, PGE2, TNFa and dexamethasone [20-23]. Only a moderate stimulation of aromatase enzyme activity has been reported in epithelial breast cancer cells [18,19]. Whether the modest activity observed in the cancer cells is sufficient to produce enough estrogen to cause intracellular biological effects remains controversial. Some studies have observed androgenmediated cell growth stimulation via aromatization in epithelial breast cancer cells [17,24,25], and induction of the estrogen-regulated gene pS2 with testosterone [16], indicating that estrogen synthesis in the cancer cells is sufficient to induce biological effects. However, others have reported a growth-inhibitory effect of testosterone [26,27] and inhibition of E2 upregulation of progesterone receptor (PR) protein expression [28], contradicting the biological significance of aromatization in epithelial breast cancer cells.
In order to further study the aromatase activity and function in human breast cancer cells, we have developed a model system with the estrogen responsive MCF-7 breast cancer cell line, grown in a low estrogen milieu. The focus of the present study was to investigate the biological relevance of endogenous aromatase activity in breast cancer cells. Further, it was examined whether androgens affected cancer cell growth by direct interaction with the estrogen receptor (ER) or the androgen receptor (AR). MCF-7 cells were treated with estrogens, aromatizable androgens or non-aromatizable androgen in a low estrogen milieu, alone or in combination with aromatase inhibitor (4-hydroxy-androstenedione or letrozole (Femara®)), the pure anti-estrogen ICI 182,780 (fulvestrant, FaslodexTM) or the anti-androgen bicalutamide (Casodex®). Cell growth, estrogen receptor a (ERα) and PR protein expression and aromatase activity were examined.
Materials and methods
2.1. Hormones and inhibitors
Testosterone, androstenedione (Adione), estradiol (E2), estrone (E1) and 4-OH-androstenedione (4-OH-A) were pur-chased from Sigma-Aldrich, St. Louis, MO. Dihydrotestos-terone (DHT) was purchased from Merch, Germany. Letro- zole (Femara®) was a gift from Novartis Pharma AG, Basel, Switzerland. ICI 182,780 (fulvestrant, FaslodexTM) and bicalutamide (Casodex®) were gifts from AstraZeneca, London, UK.
2.2. Cell culture
The MCF-7 cell line was obtained from the Human Cell Culture Bank, Mason Research Institute (Rockville, MD) and adapted to grow in a low serum concentration (1%) to obtain an estradiol concentration (approximately 1 pM) resembling postmenopausal concentrations of estradiol [29]. The AROM-1 cell line (MCF-7 cells stably transfected with the aromatase gene (CYP19)) was a gift from Dr. Mitch Dowsett’s laboratory [30]. MCF-7 cells were main-tained in FCS medium (DMEM/F12 medium (Gibco, In- vitrogen, CA) without phenol red, 1% heat-inactivated fetal calf serum (FCS) (Life Technologies, Bethesda, MD), 6 ng/ml bovine insulin (Novo Nordic, Bagsvsrd, Denmark), 2.5 mM L-glutamine). AROM-1 was maintained in AROM- 1 medium (RPMI medium (Gibco) with phenol red, 10% heat-inactivatedFCS, 10 μg/ml insulin, 2.5 mM L-glutamine, 600 μg/ml G418). Cells were sub-cultivated by trypsination once a week. In all experiments with MCF-7 cells, FCS was replaced with 5% newborn calf serum (NCS) (Life Tech-nologies) and the medium was supplemented with 2.5 x 105 U penicillin and 31.25 μg/L streptomycin (NCS medium). NCS was used for experiments, as it contains low levels of growth factors and steroid hormones, facilitating detection of any stimulatory effects of supplemented hormones. Cells were incubated in a 5% CO2 humidified incubator at 37 °C.
2.3. Cell proliferation assays
MCF-7 cells (2.5 x 104) were seeded into 24-well plates (2 cm2 wells) and left to attach for 24 h in FCS medium, fol-lowed by 24 h starvation in NCS medium. The NCS medium was changed and supplemented with hormones and inhibitors (day 0) as indicated in the figures. Vehicle (ethanol) was added to the control culture. Culturing was sustained for 5 days, with replacement of medium, hormones and inhibitors on day 3. On day 5, the cells were rinsed inPBS and a crystal violet colorimetric assay, staining DNA, was used to obtain an indirect measure of the cell number [31]. The obtained optical density for each sample was expressed as a relative value in percent of the appropriate control.
2.4. Aromatase assay
MCF-7 cells and AROM-1 cells (2 x 105) were seeded in T25 flasks in triplicate. MCF-7 cells were cultured for 4 days in FCS medium, followed by 48 h culture in either FCS or NCS medium. AROM-1 cells were cultured for 6 days in AROM-1 medium. Medium was changed every second day. Prior to the assay, medium was removed and the cells were rinsed twice in serum-free DMEM/F12 medium. 2.5 ml of serum-free DMEM/F12 medium with 0.25 μCi 1β-3H- androstenedione (Perkin-Elmer Life Sciences Inc., Boston, MA) was added to each culture flask, as well as control flasks without cells for measurements of background radioactivity. Hundred nanomolar letrozole (MCF-7 cells) or 10 μM letro- zole (AROM-1 cells) was concurrently added to pre-selected flasks. Incubation was sustained for 24 h, the assay medium was collected and the cells were counted in a Biuker-Tiuk chamber. Aromatase activity was assayed by measuring the amount of 3H present in the water phase after aromatization of 1 β-3H-androstenedione according to a protocol developed by Dr. A. Purohit and Dr. M. Reed [32]. In brief, steroid was removed from the medium by extraction with diethylether and dextran activated charcoal. Duplicate measures of 3H in the water phase of each sample were obtained with a liquid scintillation counter. The obtained background reading from the control flasks was subtracted from the sample values and the values were corrected for loss of 3H2O during processing (approximately 20%, A. Purohit, personal communication). The aromatase activity was calculated as pmol androstene- dione converted/106 cells/24 h.
2.5. Western analysis
MCF-7 cells (3 x 105) were seeded in T25 flasks and cul-tured for 4 days in FCS medium, with medium change every second day. The cells were starved in NCS medium for 24 h and cultured in NCS medium with hormones ± inhibitors for 24 h as indicated in the figure legends. Vehicle (ethanol) was added to the control culture. Cells were harvested by trypsina- tion, counted and divided into microtubes (5 x 105 cells/tube) and lysed withRIPA buffer (100 nM NaCl, 20 mM Tris base, 1% Triton X-100, 0.5% sodium deoxycolate, 0.1% SDS, 1 nM EDTA, pH 8). Protein concentration was determined using the Bio-Rad Protein Assay kit (Bio-Rad Laboratories, Munich, Germany). Ten micrograms of total protein per sample was separated on precast 4-12% Bis-Tris gels in MOPS buffer, using the X-Cell sureLock module (Invit- rogen) and proteins were transferred using the X-Cell II blot module (Invitrogen) according to the manufacturer’s instructions to PVDF membranes (Immobilon-P, Millipore, Bedford, MA). Membranes were stained with Ponceau S to confirm equal loading and transfer of samples. Non-specific binding of antibodies was blocked by incubation of the membrane in 5% non-fat dry milk (1 h RT). Membranes were exposed for 1 h at RT to monoclonal rabbit estrogen receptor a (ERα 1:10,000), progesterone receptor (PR 1:2000) antibody (both LabVision, Fremont, CA) or monoclonal mouse cytokeratin-7 (K7 1:500,000) antibody. Specific binding was visualized by incubation for 1 h at RT with species-specific peroxidase-conjugated secondary antibodies (Dako Cytoma- tion, Glostrup, Denmark) followed by visualization using the ECLPLUS kit (Amersham Pharmacia, Buckinghamshire, UK). Chemiluminiscence was detected on Hyper FilmTM (Amersham Pharmacia Biotech). Expression of K7 was used as loading control. The K7 antibody was kindly provided by Dr. Jiri Bartek (Danish Cancer Society, Copenhagen, Denmark).
2.6. Statistics
Levene’s test was used to analyze for homogeneity of variance in all experiments and log transformation of data was performed in case of unequal variance. A linear model of analysis of variance, followed by two sided pair wise t-tests with Bonferroni’s correction was used to detect differences between treatments. Results were considered significant when p < 0.05. Calculations were performed using SAS, version 8.2 (SAS institute, Cary, NC).
Results
3.1. Growth of MCF-7 cells in FCS and NCS supplemented medium
The MCF-7 cells cultured in FCS supplemented medium had a high growth rate and adding estradiol (E2, 100 pM) did not increase cell growth significantly (Fig. 1). Culturing cells in NCS supplemented medium significantly reduced the growth rate compared to cells cultured in FCS medium (p < 0.0001). Addition of 100 pM E2 to the NCS medium significantly stimulated cell growth compared to the NCS control (p < 0.0001), restoring growth to the level observed for the FCS culture. Thus, MCF-7 cells cultured in the NCS model system were dependent on E2 to retain their normal growth rate. The pure anti-estrogen ICI 182,780 (ICI, 10 nM) in a 100-fold excess concentration completely abolished the stimulatory effect of E2 (p < 0.0001), bringing cell growth to the NCS culture level. No effect of the aromatase inhibitor letrozole (100 nM) was observed on E2 stimulated growth, indicating that letrozole did not have unspecific growth inhibitory effects on the cells in the applied concentration. Four independent quadruplicate experiments gave similar results.
Figure 1. Growth of MCF-7 cells in FCS and NCS supplemented medium. E2; estradiol, ICI; ICI 182,780, Le; letrozole. Black bars illustrate cells cultured in FCS medium ± E2(100pM). Gray bars illustrate cells cultured in NCS medium ± E2 (100 pM) ± inhibitors (ICI 10 nM, Le 100 nM). Cells were cultured for 5 days with hormones and inhibitors and cell number was estimated by a colorimetric assay. Values are mean ± S.D. from a representative quadruplicate experiment. Four independent quadruplicate experiments were performed. (a) Significantly different from the FCS culture, (b) significantly different from the NCS culture, and (c) significantly different from NCS + E2 culture, p < 0.0001.
3.2. Growth stimulation of MCF-7 cells with steroid hormones
Dose-response growth curves for MCF-7 cells were es-tablished with the estrogens E2 and estrone (E1) and for the androgens testosterone, androstenedione (Adione) and dihydrotestosterone (DHT). Fig. 2 shows a representative quadruplicate experiment for each hormonal treatment. All applied concentrations of E2 from 1 pM significantly stimulated cell growth (p < 0.0001). A plateau with maximal stimulation approximately 10-fold above control level was reached at a concentration of 10 pM E2. E1 significantly stimulated cell growth with concentrations from 10 pM (p < 0.0001), with maximal growth stimulation obtained at a concentration of 100 nM. Thus, E1 was approximately 10 times less potent than E2. All applied concentrations of testosterone significantly stimulated cell growth compared to the NCS control (p < 0.0001). A plateau with maximal growth stimulation was reached with a concentration of 100 nM testosterone. DHT only stimulated cell growth significantly at concentrations from 100 nM (p < 0.0001), and maximal effect was not established with the applied concentrations of DHT. Adione stimulated cell growth approximately 2.5-fold above the level of the NCS control culture at concentrations between 10-100 nM (p < 0.0001). The effect of Adione leveled off at 1 μM, which was repeatedly observed in the majority of performed experiments. Thus, all androgens tested were observed to stimulate MCF-7 cell growth, but with different potency. Four independent quadruplicate experiments were performed for each steroid, and comparable dose-response curves were obtained, even though the fold stimulation varied between experiments. Concentrations of 100 pM E2,100 nM testosterone and 100 nM DHT were selected for further experiments. Testosterone and DHT were selected in order to study effects of bothanaromatizable androgen (testosterone) and a non-aromatizable androgen (DHT). The androgen concentration was based on the testosterone dose-response curve.
Figure 2. Growth stimulation of MCF-7 cells with steroid hormones. E2, estradiol; E1, estrone; T, testosterone; A, androstenedione; DHT, dihydrotestosterone. Cells were cultured for 5 days in NCS medium supplemented with steroid hormones and cell number was estimated by a colorimetric assay. Values are mean ± S.D. from a representative quadruplicate experiment for each steroid hormone. At least four independent quadruplicate experiments were performed for each steroid. Treatment with all steroid hormones, except 1 and 10 nM DHT, and 1 nM and 1 pM A, resulted in a significantly higher MCF-7 cell growth compared to the NCS control culture (p < 0.0001).
3.3. Inhibition of androgen-mediated MCF-7 cell growth
To elucidate the mechanism behind the androgenmediated cell growth, the aromatase inhibitor letrozole, the anti-estrogen ICI and the anti-androgen bicalutamide were added to cells treated with the aromatizable androgen testosterone orthe non-aromatizable androgenDHT (Fig. 3). Letro- zole and ICI completely abolished the effect of testosterone (p < 0.0001), whereas bicalutamide had no effect. Only ICI inhibited DHT-mediated cell growth (p < 0.0001). Fig. 3 shows a representative quadruplicate experiment. Four in-dependent quadruplicate experiments gave comparable re-sults. Thus, the estrogen receptor, but not the androgen re-ceptor, appeared to be involved in testosterone and DHT- mediated growth. Aromatization of testosterone to estrogen was required for growth-stimulation with testosterone, whereas DHT may have interacted directly with the estrogen receptor.
Figure 3. Inhibition ofandrogen-mediated MCF-7 cell growth. T; testosterone, DHT; dihydrotestosterone, ICI; ICI 182,780, Le; letrozole, BC; bicalu- tamide. Cells were cultured for 5 days in NCS medium supplemented with 100 nM androgen ± inhibitors (10 nM ICI, 100 nM Le, 100 nM BC). Cell number was estimated by a colorimetric assay. Values are mean ± S.D. from a representative quadruplicate experiment. Four independent quadruplicate experiments were performed. (a) Significantly different from the T- treated culture and (b) significantly different from the DHT-treated culture, p < 0.0001.
3.4. Inhibition of aromatization of testosterone with aromatase inhibitors
Increasing concentrations of the non-steroidal aromatase inhibitor letrozole and the steroidal inhibitor 4-hydroxy- androstenedione (4-OH-A) were added to testosterone (100 nM) treated cultures of MCF-7 cells. Fig. 4 shows a representative quadruplicate experiment. All concentrations of letrozole, from 1 nM, reduced testosterone-mediated growth (p < 0.0001), with complete inhibition down to NCS control level obtained at a concentration of 100 nM letrozole. 4-OH- A reduced testosterone-mediated growth at concentrations from 10 nM (p < 0.0001), with complete inhibition to control level obtained at a concentration of 1 μM 4-OH-A. Thus, both the steroidal and the non-steroidal aromatase inhibitor were able to completely abolish the growth-stimulatory effect of testosterone. Letrozole was approximately 10-fold more po-tent than 4-OH-A. Four independent experiments, each in quadruplicate, gave comparable results. Thus, testosterone-mediated growth stimulation appears to be dependent upon aromatization to estrogen.
Figure 4. Inhibition of testosterone-mediated MCF-7 growth with aromatase inhibitors. T; testosterone, Le; letrozole, 4-OH-A; 4-hydroxy- androstenedione, AI; aromatase inhibitor. Cells were cultured for 5 days in NCS medium with 100nM testosterone plus increasing doses of aromatase inhibitor as indicated in the figure. Cell number was estimated by a colorimetric assay and expressed as percent of the testosterone-treated culture. Values are mean ± S.D. from one quadruplicate representative experiment. At least four independent quadruplicate experiments were performed. All applied concentrations of the aromatase inhibitors, except for 1 nM 4-OH- A, resulted in inhibition of testosterone-mediated growth (p < 0.0001).
3.5. Aromatase activity in MCF-7 andAROM-1 cells
The activity of the aromatase enzyme was measured using a tritium release assay [32], with 1 (3-3H-androstenedione as the substrate in serum-free medium. A significant aro-matase activity could be measured in MCF-7 cells (Fig. 5), and was significantly higher (p < 0.0001) in cells primed in NCS supplemented medium (0.085 pmol androstene- dione converted/106 cells/24 h) compared to cells primed in FCS supplemented medium (0.011 pmol androstene-dione converted/106 cells/24 h). Higher activity (0.4 pmol an- drostenedione converted/106 cells/24 h) was observed in the AROM-1 cells (MCF-7 cells stably transfected with the aromatase gene), which were used as a positive control in the assay. Thus, the amount of labeled substrate was not limiting for aromatization in the MCF-7 cells. Addition of 100 nM letrozole to MCF-7 cells cultured inNCS medium completely inhibited the aromatase activity (p < 0.0001). Tenmicromolar letrozole inhibited the aromatase enzyme in AROM-1 cells (p < 0.0001). Experiments with 1,2β-3H-labeled testosterone as substrate were also performed, but the results were confounded by cell-independent release of 3H from the substrate (data not shown).
Figure 5. Aromatase enzyme activity in MCF-7 and AROM-1 cells. Prior to the assay, MCF-7 cells were cultured for 4 days in FCS and 2 days in FCS or NCS medium. AROM-1 cells were cultured for 6 days in AROM-1 medium. Cells were incubated with 0.25 pCi 1β-3H-androstenedione ± letrozole (Le) in serum-free medium for 24 h. The letrozole concentration was 100 nM for MCF-7 cells and 10 pM for AROM-1 cells. Medium was assayed for produced 3H2O and aromatase activity was calculated as pmol androstenedione (Adione) converted per 106 cells per 24 h. Values are mean ± S.D. of a representative triplicate experiment out of four independent experiments. (a) Significantly different from the FCS-treated MCF-7 culture, (b) significantly different from the NCS-treated MCF-7 culture, and (c) significantly different from the FCS-treated AROM-1 culture, p < 0.0001.
3.6. PR andERα protein expression in MCF-7 cells treated with estradiol or testosterone
The MCF-7 cell line is known to express PR and ERα, and only extremely low levels of the estrogen receptor p (ERp) [33]. PR is expressed in two forms, PR-A andPR-B. PR is an estrogen-regulated gene [34] and the expression is often used as a marker for the presence of functional ERα. ERα protein is significantly reduced after estrogen treatment of MCF-7 cells, both due to instability of the receptor in the presence of estrogen, and due to downregulation of the mRNA [35]. Accordingly, ERα and PR protein expression was examined in order to investigate whether testosterone stimulated MCF-7 cells via the ERα, both directly and/or after aromatization to estrogen. As shown in Fig. 6, ERα expression was downregu- lated in cells treated with E2 and testosterone. PR expression was highly upregulated after exposure to E2 and testosterone. Letrozole abolished the testosterone-mediated regulation of both ERα and PR, but not the regulation exerted by E2. ICI inhibited the effects of both E2 and testosterone on PR expression, whereas the level of ERα remained low as ICI in itself downregulates ERα protein expression [35]. As letro- zole completely inhibited the testosterone-mediated effect, aromatization of testosterone to estrogen was required for testosterone to affect the protein expression of these estrogen responsive genes. Further, regulation of ERα and PR expression was mediated via estrogen-interaction with ERα as ICI abolished the effect of aromatized testosterone.
Figure 6. PR and ERα protein expression in MCF-7 cells. PR-A and B; progesterone receptor A and B form, ERα; estrogen receptor a, E2; estradiol (100pM), T; testosterone (100nM), ICI; ICI 182,780 (10nM), Le; letrozole (100nM). Expression of cytokeratin 7 (K7) was used as loading control. Prior to treatment, cells were cultured for 4 days in FCS medium and starved for 2 days in NCS medium. Cells were treated with hormones ± inhibitors for 24 h. Protein was extracted and ERα and PR expression measured by Western analysis. Numbers below lanes are protein expression levels normalized to the K7 expression and expressed as fold ofthe NCS culture. Three independent experiments gave comparable results.
Discussion
Locally produced estrogen is suggested to play a major role for proliferation of estrogen responsive breast cancer (recently reviewed in [36]). Intratumoral estrogens are produced from circulating androgens, catalyzed by the aromatase enzyme. Whether aromatase is localized in cancer cells, the surrounding stromal cells or both, has been debated since the first published immunohistochemical study of aromatase in breast carcinomas [37] and it still remains controversial [11-15]. The present study was undertaken in order to examine the presence of aromatase activity and its possible biological significance in human MCF-7 breast cancer cells.
A low but reproducible aromatase activity could be measured in the MCF-7 cells, which is in agreement with other reports [16-19]. The androgen testosterone was observed to stimulate MCF-7 cell growth significantly at concentrations as low as 1 nM, which is within the physiological range for this steroid hormone. The stimulation could be inhibited with aromatase inhibitors, indicating that aromatization of testosterone was responsible for the testosterone-mediated growth. Thus, in the present model system, the epithelial breast cancer cell line MCF-7 was observed to express sufficient aromatase activity in order to stimulate cell growth via aromatization of testosterone to E2. Expression of ERα and PR confirmed that testosterone was aromatized to estrogen, which then affected expression of these estrogen-regulated genes. Testosterone did not appear to interact directly with ERα. Further, the anti-androgen bicalutamide had no effect onT-stimulated growth, suggesting that the androgen receptor (AR) was not involved in the testosterone-mediated effects. Conflicting reports of testosterone-mediated effects on MCF-7 cells exist inthe literature. Cell growth [17,24] andpS2 expression [16] have previously been observed to be stimulated as a consequence of aromatization of testosterone to estrogen. Further, testosterone has been reported to stimulate MCF-7 cell growth via direct interaction with ER [38]. Direct interaction of testosterone with ER has been reported at concentrations from 100 nM [39,40], which could not be confirmed in the present study, where testosterone only exerted effects after aromatization to estrogen. However, testosterone has also been observed to inhibit MCF-7 growth [26,27] and inhibit E2-mediatedupregulation of PR protein expression [28], pre-sumably via AR signaling [27,28].
Compared to testosterone, Adione only stimulated MCF- 7 cell growth slightly, however, significantly. This was surprising as Adione based on circulating substrate availability, is believed to be the major substrate for the aromatase enzyme in peripheral tissues [41,42]. Yet, the concentration of Adione and the product from aromatization of Adione, E1 , has been observed to be low in breast tumors, compared to the concentration of testosterone and E2 [43]. This could indicate that the preferred substrate for the aromatase enzyme in breast cancer tissue is testosterone, produced in peripheral tissue from conversion of circulating Adione via type 5 17β- hydroxysteroid dehydrogenase (HSD) activity [1]. The low effect of Adione may thus indicate that the HSD activity is insufficient for the conversion of Adione to testosterone in the MCF-7 cells used in the present study. To our knowledge, the activity of type 5 HSD has not been studied in MCF-7 cells. The product from aromatization of Adione, E1, is a less potent estrogen than E2, and it could also be speculated that the amount of E1 produced from aromatization of Adione was insufficient to stimulate MCF-7 growth, compared to the E2 produced from testosterone. Yet, we found that E1 stimulated MCF-7 cell growth at concentrations as low as 10 nM and the small growth stimulation obtained with Adione may thus not be ascribed to lack of E1 potency. Other studies have also shown that aromatization of testosterone elicited estrogenic responses at concentrations where Adione was inactive [17,40], but also here was the reason unclear. The low aromatase activity obtained with 1β-3H-labeled androstene- dione as substrate in the present study is in correspondence with the low growth stimulation obtained with androstene- dione. It could be speculated that the aromatase activity would be higher, if labeled testosterone served as substrate. To investigate this, 1,2β-3H-labeled testosterone was used as substrate in the aromatase assay as 1 β-3H-labeled testosterone is not commercially available. However, the results were con- foundedby cell-independent release of 3H from the substrate, resulting inhighbackground measures (data not shown). This may have been due to spontaneous release of 3H from the 2 β- position, which is not involved in aromatization of androgens to estrogens.
In addition to ERα, the MCF-7 cells were observed to express AR (data not shown). The non-aromatizable andro-gen DHT has a high affinity for the AR and it could be speculated that DHT mediated cell growth via AR. How-ever, high concentrations, from 100 nM DHT, were needed to stimulate MCF-7 cell growth, which could indicate that the stimulation was not mediated by AR interaction. The inhibition study confirmed this, as the anti-androgen bica- lutamide had no effect on DHT-mediated growth. On the contrary, ICI completely blocked the effect, indicating that DHT may have mediated MCF-7 growth by direct ERα in-teraction. DHT has earlier been observed to bind to ERα in MCF-7 cells and elicit estrogenic actions, but only at phar-macological concentrations [38-40], which is in correspon-dence to our growth studies. Another study has reported that both physiological (1-10 nM) and pharmacological concentrations (100-1000 nM) of DHT stimulated proliferation after 48 h incubation, whereas longer time incubation with DHT inhibited proliferation [44]. In the present study, the effect of DHT was assessed after 5 days treatment and the inhibitory effect of DHT beyond 48 h incubation was not confirmed. In conclusion, only pharmacological concentrations of DHT had a stimulatory effect on MCF-7 cell growth, possibly mediated by direct interaction with ERα.
Whether androgens play a role in human breast cancer, besides being substrates for aromatization, is debated. An-drogens have been suggested both directly to promote or to inhibit breast cancer cell growth, or to act differentially de-pending on the estrogen milieu, acting as inhibitors in a high estrogen milieu, but as promoters in a low estrogen milieu (reviewed in [45,46]). The hypothesis of androgens acting directly stimulatory in a low estrogen milieu could be of rel-evance for treatment benefit and disease progression in breast cancer patients deprived of estrogen via aromatase inhibitor treatment. However, the hypothesis was not supported by the present findings, as testosterone had no direct stimulatory effect but only affected the examined endpoints after arom- atization, despite the low estrogen milieu. DHT may have stimulated MCF-7 cell growth via ERα interaction, however, only at pharmacological concentrations.
The reason for the conflicting results found in the litera-ture regarding androgen-mediated growth effects on MCF-7 cells is unclear, but could be related to experimental design or differences between MCF-7 strains. Several comparative studies have shown that direct comparison between studies can be conflicting due to independent progression of MCF-7 strains [47-52]. Whether strain differences in regard of re-ceptor content, E2 responsiveness or aromatase enzyme ex-pression may contribute to the differences can be speculated. One study has previously reported different aromatase activity in two MCF-7 strains [17]. Estrogen responsiveness is required for aromatized androgens to act as stimulators and may thus affect the observed effects of androgens. Our experience is that MCF-7 strains differ significantly with respect to estrogen responsiveness. In one study where testosterone inhibited MCF-7 cell growth, the cells appeared to be estrogen independent as the cells had a high growth rate in serum-free medium [26], possibly affecting the results. The MCF-7 cells used in the present study have been adapted to grow with a low serum concentration and may supposedly have become less dependent of, or hypersensitive to estrogens and growth factors. Hypersensitivity to E2 could explain the growth stimulatory effects of testosterone if just a small amount was aromatized to E2. As seen in Fig. 2, our MCF-7 strain was indeed sensitive to E2, however, the sensitivity was comparable to that of MCF-7 strains routinely maintained in 10% FCS (maximal growth stimulation at 100 pM E2) and not to that of hypersensitized MCF-7 cells (maximal growth stimulationat 100fME2), long-term deprived of estrogen (L- TED) [53,54]. Also, other studies have measured aromatase activity in MCF-7 cells only shortly deprived of estrogen and thus not hypersensitized [16,17]. Yet, the MCF-7 cells used in the present study may have acquired other changes from the long-term culture in a low estrogen milieu, low growth factor milieu (1% FCS), rendering this strain able to aromatize testosterone. Indeed, the aromatase activity was observed to be significantly higher in cells cultured in steroid poor NCS serum than FCS serum, which could indicate that growth in steroid poor serum further stimulated the aromatase activity in the cells. The concentration of estrogens in the NCS serum used in our laboratory has been routinely measured by specific radio-immune assay (RIA) and was comparable to theconcentrationinsteroid-strippedFCS serum (E2 < 40pM, E2-sulphate < 200 pM, E1 = 100 pM, E1 -sulphate = 400 pM). Also, MCF-7 cells adapted to grow on serum-free medium, yet retaining estrogen-responsiveness, have been observed to be growth-stimulated with androgens through conversion to estrogens [25]. It would be of great interest to repeat the experiments in different L-TED breast cancer cell lines, to see whether growth in a low estrogen milieu per se enables breast cancer cells to aromatize androgens. It is possible that a similar effect of estrogen deprivation in breast cancer patients may render the cancer cells able to produce endogenous estrogen.
In conclusion, a significant aromatase activity was de-tected in the MCF-7 cells. The activity was sufficient for the cells to aromatize testosterone to estrogen, resulting in significant cell growth stimulation at testosterone concentra-tions from 1 nM, as well as induction of estrogen responsive proteins. The results indicate that breast cancer cells, adapted to grow in a low estrogen milieu, possess the ability to synthesize estrogens from circulating androgens. Accord-ingly, estrogen may be an important intracrine factor in postmenopausal human breast cancer, where circulating estrogen levels are low.
Acknowledgements
We acknowledge Novartis Pharma AG, Basel, Switzer-land, for providing letrozole, and AstraZeneca, UK, for pro-viding ICI 182,780 andbicalutamide. Mitch Dowsett’s labo-ratory, Institute of Cancer Research, Royal Cancer Hospital, UK, is acknowledged for their donation of AROM-1 cells and Jiri Bartek, the Danish Cancer Society, Denmark, for the donation of K7 antibody. Mike Reed and Alan Purohit, Imperial College, UK, are acknowledged for providing the aromatase assay protocol together with expert advice. Nils Brunner, Royal Veterinary and Agricultural University, Denmark, is acknowledged for invaluable discussions of the experimental design and Ib Jarle Christensen, Hvidovre University Hospital, Denmark, for statistical assistance. Inger Heiberg and Birgit Reiter are acknowledged for excellent technical assistance. The work was supported by grants from the Medical Faculty at University of Copenhagen, the Danish Cancer Research Foundation, the Danish Cancer Society’s Scientific Committee (DNLU), Karen A. Tolstrup’s Fond, LEO Pharma Research Foundation, Beckett Fonden, Ingenior August Frederik Wedell Erichsens Legat and Svend Coles Frederiksens og Hustrus Fond.
References
F. Labrie, V. Luu-The, C. Labrie, J. Simard, DHEA and its transformation into androgens and estrogens in peripheral target tissues: intracrinology, Front. Neuroendocrinol. 22 (3) (2001) 185-212.
Arimidex Study Group, J.M. Nabholtz, A. Buzdar, M. Pollak, W. Harwin, G. Burton, A. Mangalik, M. Steinberg, A. Webster, M. von Euler, Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: results of a North American multicenter randomized trial, J. Clin. Oncol. 18 (22) (2000) 3758-3767.
J. Bonneterre, A. Buzdar, J.M. Nabholtz, J.F. Robertson, B. Thurli- mann, M. von Euler, T. Sahmoud, A. Webster, M. Steinberg, Anastrozole is superior to tamoxifen as first-line therapy in hormone receptor positive advanced breast carcinoma, Cancer 92 (9) (2001) 2247-2258.
H. Mouridsen, M. Gershanovich, Y Sun, R. Perez-Carrion, C. Boni, A. Monnier, J. Apffelstaedt, R. Smith, H.P. Sleeboom, F. Janicke, A. Pluzanska, M. Dank, D. Becquart, P.P. Bapsy, E. Salminen, R. Snyder, M. Lassus, J.A. Verbeek, B. Staffler, H.A. Chaudri-Ross, M. Dugan, Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: results of a phase III study of the International Letrozole Breast Cancer Group, J. Clin. Oncol. 19 (10) (2001) 2596-2606.
M. Baum, A.U. Budzar, J. Cuzick, J. Forbes, J.H. Houghton, J.G. Klijn, T. Sahmoud, Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial, Lancet 359 (9324) (2002) 2131-2139.
H. Mouridsen, M. Gershanovich, Y. Sun, R. Perez-Carrion, C. Boni, A. Monnier, J. Apffelstaedt, R. Smith, H.P. Sleeboom, F. Jaenicke, A. Pluzanska, M. Dank, D. Becquart, P.P. Bapsy, E. Salminen, R. Snyder, H. Chaudri-Ross, R. Lang, P. Wyld, A. Bhatnagar, Phase III study of letrozole versus tamoxifen as first-line therapy of advanced breast cancer in postmenopausal women: analysis of survival and update of efficacy from the International Letrozole Breast Cancer Group, J. Clin. Oncol. 21 (11) (2003) 2101-2109.
R.C. Coombes, E. Hall, L.J. Gibson, R. Paridaens, J. Jassem, T. De- lozier, S.E. Jones, I. Alvarez, G. Bertelli, O. Ortmann, A.S. Coates, E. Bajetta, D. Dodwell, R.E. Coleman, L.J. Fallowfield, E. Mick- iewicz, J. Andersen, P.E. Lonning, G. Cocconi, A. Stewart, N. Stuart, C.F. Snowdon, M. Carpentieri, G. Massimini, J.M. Bliss, A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer, N. Engl. J. Med. 350 (11) (2004) 1081-1092.
M.J. Reed, A.M. Owen, L.C. Lai, N.G. Coldham, M.W. Ghilchik, N.A. Shaikh, V.H. James, In situ oestrone synthesis in normal breast and breast tumour tissues: effect of treatment with 4- hydroxyandrostenedione, Int. J. Cancer 44 (2) (1989) 233-237.
W.R. Miller, P. Mullen, P. Sourdaine, C. Watson, J.M. Dixon, J. Telford, Regulation of aromatase activity within the breast, J. Steroid Biochem. Mol. Biol. 61 (3-6) (1997) 193-202.
A.A. Larionov, L.M. Berstein, W.R. Miller, Local uptake and synthesis of oestrone in normal and malignant postmenopausal breast tissues, J. Steroid Biochem. Mol. Biol. 81 (1) (2002) 57-64.
A. Brodie, Q. Lu, J. Nakamura, Aromatase in the normal breast and breast cancer, J. Steroid Biochem. Mol. Biol. 61 (3-6) (1997) 281-286.
Y. Yamamoto, J. Yamashita, M. Toi, M. Muta, S. Nagai, N. Hanai, A. Furuya, Y. Osawa, S. Saji, M. Ogawa, Immunohistochemical analysis of estrone sulfatase and aromatase in human breast cancer tissues, Oncol. Rep. 10 (4) (2003) 791-796.
P.C. de Jong, M.A. Blankenstein, J.W. Nortier, P.H. Slee, J. van de Ven, J.M. van Gorp, J.R. Elbers, M.E. Schipper, G.H. Blijham, J.H. Thijssen, Q. Lu, D. Jelovac, A.M. Brodie, The relationship between aromatase in primary breast tumors and response to treatment with aromatase inhibitors in advanced disease, J. Steroid Biochem. Mol. Biol. 87 (2-3) (2003) 149-155.
H. Sasano, D.P Edwards, T.J. Anderson, S.G. Silverberg, D.B. Evans, R.J. Santen, P. Ramage, E.R. Simpson, A.S. Bhatnagar, W.R. Miller, Validation of new aromatase monoclonal antibodies for im- munohistochemistry: progress report, J. Steroid Biochem. Mol. Biol. 86 (3-5) (2003) 239-244.
Z. Zhang, H. Yamashita, T. Toyama, Y. Hara, Y. Omoto, H. Sugiura, S. Kobayashi, N. Harada, H. Iwase, Semi-quantitative immunohistochemical analysis of aromatase expression in ductal carcinoma in situ of the breast, Breast Cancer Res. Treat. 74 (1) (2002) 47-53.
W.E. Burak Jr., A.L. Quinn, W.B. Farrar, R.W. Brueggemeier, Androgens influence estrogen-induced responses in human breast carcinoma cells through cytochrome P450 aromatase, Breast Cancer Res. Treat. 44 (1) (1997) 57-64.
C. Palma, M. Criscuoli, A. Lippi, M. Muratori, S. Mauro, C.A. Maggi, Effect of the aromatase inhibitor MEN 11066, on growth of two different MCF-7 sublines, Eur. J. Pharmacol. 409 (2) (2000) 93-101.
J.A. Richards, T.A. Petrel, R.W. Brueggemeier, Signaling pathways regulating aromatase and cyclooxygenases in normal and malignant breast cells, J. Steroid Biochem. Mol. Biol. 80 (2) (2002) 203-212.
S. Catalano, S. Marsico, C. Giordano, L. Mauro, P Rizza, M.L. Panno, S. Ando, Leptin enhances, via AP-1, expression of aromatase in the MCF-7 cell line, J. Biol. Chem. 278 (31) (2003) 28668-28676.
Y. Zhao, V.R. Agarwal, C.R. Mendelson, E.R. Simpson, Transcriptional regulation of CYP19 gene (aromatase) expression in adipose stromal cells in primary culture, J. Steroid Biochem. Mol. Biol. 61 (3-6) (1997) 203-210.
A. Singh, A. Purohit, M.W. Ghilchik, M.J. Reed, The regulation of aromatase activity in breast fibroblasts: the role of interleukin-6 and prostaglandin E2, Endocr. Relat. Cancer 6 (2) (1999) 139-147.
A.L. Quinn, W.E. Burak Jr., R.W. Brueggemeier, Effects of matrix components on aromatase activity in breast stromal cells in culture, J. Steroid Biochem. Mol. Biol. 70 (4-6) (1999) 249-256.
A. Purohit, A. Singh, M.W. Ghilchik, O. Serlupi-Crescenzi, M.J. Reed, Inhibition of IL-6+IL-6 soluble receptor-stimulated aromatase activity by the IL-6 antagonist, Sant 7, in breast tissue-derived fibroblasts, Br. J. Cancer 88 (4) (2003) 630-635.
M. Schmitt, K. Klinga, B. Schnarr, R. Morfin, D. Mayer, Dehy- droepiandrosterone stimulates proliferation and gene expression in MCF-7 cells after conversion to estradiol, Mol. Cell Endocrinol. 173 (1-2) (2001) 1-13.
J. Jensen, J.W. Kitlen, P Briand, F. Labrie, A.E. Lykkesfeldt, Effect of antiestrogens and aromatase inhibitor on basal growth of the human breast cancer cell line MCF-7 in serum-free medium, J. Steroid Biochem. Mol. Biol. 84 (4) (2003) 469-478.
J. Ortmann, S. Prifti, M.K. Bohlmann, S. Rehberger-Schneider, T. Strowitzki, T. Rabe, Testosterone and 5 alpha-dihydrotestosterone inhibit in vitro growth of human breast cancer cell lines, Gynecol. Endocrinol. 16 (2) (2002) 113-120.
S. Ando, F. De Amicis, V. Rago, A. Carpino, M. Maggiolini, M.L. Panno, M. Lanzino, Breast cancer: from estrogen to androgen receptor, Mol. Cell Endocrinol. 193 (1-2) (2002) 121128.
J.H. MacIndoe, L.A. Etre, An antiestrogenic action of androgens in human breast cancer cells, J. Clin. Endocrinol. Metab. 53 (4) (1981) 836-842.
P. Briand, A.E. Lykkesfeldt, Effect of estrogen and antiestrogen on the human breast cancer cell line MCF-7 adapted to growth at low serum concentration, Cancer Res. 44 (3) (1984) 1114-1119.
V.M. Macaulay, J.E. Nicholls, J. Gledhill, M.G. Rowlands, M. Dowsett, A. Ashworth, Biological effects of stable overexpression of aromatase in human hormone-dependent breast cancer cells, Br. J. Cancer 69 (1) (1994) 77-83.
B.K. Lundholt, P Briand, A.E. Lykkesfeldt, Growth inhibition and growth stimulation by estradiol of estrogen receptor transfected human breast epithelial cell lines involve different pathways, Breast Cancer Res. Treat. 67 (3) (2001) 199-214.
M.J. Reed, L. Topping, N.G. Coldham, A. Purohit, M.W. Ghilchik, V.H. James, Control of aromatase activity in breast cancer cells: the role of cytokines and growth factors, J. Steroid Biochem. Mol. Biol. 44 (4-6) (1993) 589-596.
P. de Cremoux, C. Tran-Perennou, B.L. Brockdorff, E. Boudou, N. Briinner, H. Magdelenat, A.E. Lykkesfeldt, Validation of real-time RT-PCR for analysis of human breast cancer cell lines resistant or sensitive to treatment with antiestrogens, Endocr. Relat. Cancer 10 (3) (2003) 409-418.
P Kastner, A. Krust, B. Turcotte, U. Stropp, L. Tora, H. Gronemeyer, P. Chambon, Two distinct estrogen-regulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B, EMBO J. 9 (5) (1990) 1603-1614.
B.L. Jensen, J. Skouv, B.K. Lundholt, A.E. Lykkesfeldt, Differential regulation of specific genes in MCF-7 and the ICI 182780-resistant cell line MCF-7/182R-6, Br. J. Cancer 79 (3-4) (1999) 386-392.
E.R. Simpson, Sources of estrogen and their importance, J. Steroid Biochem. Mol. Biol. 86 (3-5) (2003) 225-230.
J.M. Esteban, Z. Warsi, M. Haniu, P Hall, J.E. Shively, S. Chen, Detection of intratumoral aromatase in breast carcinomas. An immunohistochemical study with clinicopathologic correlation, Am. J. Pathol. 140 (2) (1992) 337-343.
M. Maggiolini, O. Donze, E. Jeannin, S. Ando, D. Picard, Adrenal androgens stimulate the proliferation of breast cancer cells as direct activators of estrogen receptor alpha, Cancer Res. 59 (19) (1999) 4864-4869.
D.T. Zava, W.L. McGuire, Human breast cancer: androgen action mediated by estrogen receptor, Science 199 (4330) (1978) 787-788.
J.C. Le Bail, F. Marre-Fournier, J.C. Nicolas, G. Habrioux, C19 steroids estrogenic activity in human breast cancer cell lines: importance of dehydroepiandrosterone sulfate at physiological plasma concentration, Steroids 63 (12) (1998) 678-683.
E.R. Simpson, M.S. Mahendroo, G.D. Means, M.W. Kilgore, M.M. Hinshelwood, S. Graham-Lorence, B. Amarneh, Y. Ito, C.R. Fisher, M.D. Michael, Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis, Endocr. Rev. 15 (3) (1994) 342-355.
M.J. Reed, A. Purohit, Breast cancer and the role of cytokines in regulating estrogen synthesis: an emerging hypothesis, Endocr. Rev. 18 (5) (1997) 701-715.
J.H. Thijssen, Local biosynthesis and metabolism of oestrogens in the human breast, Maturitas 49 (1) (2004) 25-33.
S.R. Aspinall, S. Stamp, A. Davison, B.K. Shenton, T.W. Lennard, The proliferative effects of 5-androstene-3 beta,17 beta-diol and 5 alpha-dihydrotestosterone on cell cycle analysis and cell proliferation in MCF7, T47D and MDAMB231 breast cancer cell lines, J. Steroid Biochem. Mol. Biol. 88 (1) (2004) 37-51.
C. Dimitrakakis, J. Zhou, C.A. Bondy, Androgens and mammary growth and neoplasia, Fertil. Steril. 77 (Suppl. 4) (2002) S26-S33.
D.J. Liao, R.B. Dickson, Roles of androgens in the development, growth, and carcinogenesis of the mammary gland, J. Steroid Biochem. Mol. Biol. 80 (2) (2002) 175-189.
I.H. Hamelers, R.F. Van Schaik, J.S. Sussenbach, P.H. Steenbergh, 17beta-Estradiol responsiveness of MCF-7 laboratory strains is dependent on an autocrine signal activating the IGF type I receptor, Cancer Cell Int. 3 (1) (2003) 10.
J.L. Gooch, D. Yee, Strain-specific differences in formation of apoβ- totic DNA ladders in MCF-7 breast cancer cells, Cancer Lett. 144 (1) (1999) 31-37.
H. Bahia, J.N. Ashman, L. Cawkwell, M. Lind, J.R. Monson, PJ. Drew, J. Greenman, Karyotypic variation between independently cultured strains of the cell line MCF-7 identified by multicolour fluorescence in situ hybridization, Int. J. Oncol. 20 (3) (2002) 489-494.
C. Jones, J. Payne, D. Wells, J.D. Delhanty, S.R. Lakhani, A. Kortenkamp, Comparative genomic hybridization reveals extensive variation among different MCF-7 cell stocks, Cancer Genet. Cytogenet. 117 (2) (2000) 153-158.
S. Cassanelli, J. Louis, D. Seigneurin, Progesterone receptor heterogeneity in MCF-7 cell subclones is related to clonal origin and kinetics data, Tumour Biol. 16 (4) (1995) 222-229.
S. Tamir, S.S. Kadner, J. Katz, T.H. Finlay, Regulation of antitrypsin and antichymotrypsin synthesis by MCF-7 breast cancer cell sublines, Endocrinology 127 (3) (1990) 1319-1328.
C.M. Chan, L.A. Martin, S.R. Johnston, S. Ali, M. Dowsett, Molecular changes associated with the acquisition of oestrogen hypersensitivity in MCF-7 breast cancer cells on long-term oestrogen deprivation, J. Steroid Biochem. Mol. Biol. 81 (4-5) (2002) 333-341.
R. Santen, M.H. Jeng, J.P. Wang, R. Song, S. Masamura, R. McPherson, S. Santner, W. Yue, W.S. Shim, Adaptive hypersensitivity to estradiol: potential mechanism for secondary hormonal responses in breast cancer patients, J. Steroid Biochem. Mol. Biol. 79 (1-5) (2001) 115-125.