|Home | About | Journals | Submit | Contact Us | Français|
In non-small cell lung cancer (NSCLC) cells, 17β-estradiol increases transcription, activates MAPK, and stimulates proliferation. We hypothesize that estrogen receptor β (ERβ) mediates these responses because it, but not ERα, is detected in our NSCLC cell lines. To test this, we determined the effects of the ERβ-selective agonists genistein (GEN) and 2,3-bis(4-hydroxyphenyl)propionitrile (DPN) and the ERα-selective agonist 4,4’,4”-(4-propyl-[1H]-pyrazole-l,3,5-triyl)trisphenol (PPT) in 201T cells. The cells were transfected with either an ERα or an ERβ expression vector and an estrogen response element (ERE)-tk-luciferase reporter construct. PPT increased luciferase activity in cells expressing ERα but not ERβ GEN and DPN selectively increased luciferase activity in ERβ-transfected cells at concentrations ≤ 10 nM. Fulvestrant blocked the GEN and DPN-mediated increases, indicating that transcription was ER-dependent. GEN but not PPT mediated a significant 1.5-fold increase in reporter activity upon transfection with ERE-tk-luciferase alone, demonstrating that endogenous ERβ activates transcription. PPT and DPN increased MAPK phosphorylation (2.5-fold and 3.7-fold, respectively). However, only DPN stimulated 201T growth in vitro (p = 0.008) and in vivo (p = 0.05). We conclude that ERβ mediates genomic and non-genomic responses to estrogen in 201T cells and that activation of both pathways may be necessary for increased proliferation of these cells.
Lung cancer is the leading cause of cancer mortality worldwide, with greater than 160,000 deaths expected in 2007 . Although smoking remains the primary cause of this disease, approximately 25% of patients who develop lung cancer are lifetime never smokers . A disproportionate number of these patients are females, which is suggestive of a potential role for estrogens in supporting the development or growth of lung cancer. Although epidemiologic studies have not shown a consistent association between use of exogenous estrogens and lung cancer risk (reviewed in ), recent reports point to a detrimental effect of estrogen on lung cancer outcome. In an analysis of the SEER database, we found that women diagnosed with either squamous cell carcinoma or bronchiolalveolar carcinoma who were between 55 and 59 years old (and were presumably post-menopausal) had significantly better survival than women who were 40–49 years old (and were presumably pre-menopausal) . In a retrospective analysis of SWOG clinical trials, Albain et al. identified a survival benefit for women diagnosed with advanced non-small cell lung cancer (NSCLC) compared to men . However, the survival advantage was observed only in women who were age 60 or older. In a third study, Ross et al. quantified free estradiol levels in serum samples collected from males enrolled in phase III clinical trials in advanced NSCLC . These investigators observed that men with high free estradiol levels had significantly poorer survival than men with lower estradiol levels.
To elucidate the role of estrogen in lung cancer, the effects of 17β-estradiol (E2) have been directly evaluated using pre-clinical models. In an animal model in which lung adenocarcinomas were induced by K-ras activation and p53 deletion, E2 promoted tumor progression: Both tumor burden and differentiation were affected by estrogen . In established NSCLC cell lines, E2 significantly increased cell proliferation in vitro and in vivo, increased endogenous gene expression and promoted VEGF secretion; and induced rapid phosphorylation/activation of p44/p42 mitogen activated protein kinase (MAPK) [8–12]. Conversely, agents that either antagonized ER function (fulvestrant) or blocked endogenous E2 synthesis (exemestane) significantly inhibited the growth of lung tumor xenografts [10, 11, 13]. The pro-proliferative/pro-survival responses elicited by E2 in these experiments provide an explanation for the poorer clinical outcomes observed in NSCLC patients who are either predicted or documented to have high estrogen levels.
The cellular response to estrogens is mediated by estrogen receptors ERα and ERβ, which belong to the nuclear steroid hormone receptor superfamily. Classically, these proteins function as sequence specific, ligand-dependent transcription factors: Upon E2 binding, ERα and ERβ increase the transcription of genes whose promoters contain an estrogen response element (ERE). The two receptors are encoded by distinct genes and display differential tissue distributions. Whereas ERα is expressed primarily in the uterus, liver, kidney and heart, ERβ expression occurs primarily in the ovary, prostate, lung, gastrointestinal tract, bladder, hematopoietic and central nervous systems . Analysis of murine knock-out models uncovered an important role for ERβ in normal lung biology. Targeted inactivation of ERβ results in lung abnormalities in female mice (at 3 months of age) including a decrease in the number of alveoli and altered surfactant homeostasis . By 5 months of age, both male and female ERβ-deficient mice exhibit signs of significant lung dysfunction . No similar role for ERα has been described.
A central question to be answered with regard to the estrogen signaling pathways that are operative in lung cancer cells is which of the ER proteins mediates the response. To address this issue, immunohistochemistry studies have been conducted in which the relationship between ER expression in tumor tissue and disease outcome was analyzed. Nuclear localization of ERβ was observed in 45%-69% of NSCLC cases [17–20]. In each of four recent studies, nuclear ERβ expression was found to be a favorable prognostic indicator, although this was observed only in males in two of the studies [18, 20]. Although ERβ immunoreactivity was also detected in the cytoplasm, it was observed primarily in tumor specimens that also express a nuclear pool of the receptor. The analysis and interpretation of ERα staining patterns in NSCLC is considerably more complex. Nuclear ERα immunostaining is either never, or only rarely, detected in primary lung tumors [17–20]. ERα is detected in the cytoplasm of NSCLC cases, but the proportion of positive cases varies widely from approximately 3%  to 73% . In two studies where the prognostic significance of ERα expression was reported, it was found to have either no correlation with survival  or to correlate with poor prognosis .
Because ERβ is expressed and functional in normal lung tissue and is consistently detected in primary lung cancers, we hypothesized that that this receptor mediates the response of NSCLC cells to estrogen. To ascertain whether activation of endogenous ERβ is sufficient to induce the full range of estrogenic responses described in lung cancer cells, we determined the effects of the ERβ subtype selective agonists genistein (GEN) and 2,3-bis(4-hydroxyphenyl)propionitrile (DPN) on gene transcription, MAPK activation and cell growth in 201T NSCLC cells. GEN has a 26-fold binding affinity preference for ERβ versus ERα and activates ERβ with nearly 7-fold greater potency than ERα . DPN has a 70-fold binding affinity preference for ERβ versus ERα and activates ERβ with 80-fold greater potency than ERα . As a control, we also evaluated the effects of the ERα-selective agonist 4,4’,4”-(4-propyl-[lH]-pyrazole-l,3,5-triyl)trisphenol (PPT) in our studies, which has a > 400-fold binding affinity preference for ERα versus ERβ .
201T cells were produced previously from primary non-small cell lung cancer (NSCLC) tumors as described . The human estrogen-dependent breast cancer cell line MCF-7 and the human NSCLC cell lines H23 and A549 were purchased from ATCC (Manassas, VA). 201T and A549 cells were maintained in Basal Medium Eagles (GIBCO Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) (HyClone Laboratories, Inc., Logan, UT), 2 mM L-glutamine (GIBCO) and 100-units/mL penicillin-streptomycin (GIBCO). MCF-7 and H23 cells were maintained in RPMI 1640 medium (Mediatech) supplemented with 10% FBS and 100-units/mL penicillin-streptomycin. All cultures were maintained at 37°C in a humidified 5% CO2 atmosphere and were mycoplasma-free. For all experiments involving ER ligands, cells were washed in sterile PBS and then cultured in phenol red-free medium with either charcoal-stripped serum (CSS) or no serum for at least 24 hr prior to addition of ligand.
E2 and GEN were purchased from Sigma Aldrich (St. Louis, MO). PPT, DPN and fulvestrant were purchased from Tocris (Ellisville, MO). The ligands were prepared either as l×l0−2M stocks in 100% ethanol (E2, PPT, DPN and fulvestrant) or as a l×l0−1 M stock in DMSO (GEN) and stored at –20°C. Prior to use in an experiment, each of the ligands was further diluted into tissue culture medium so as to keep vehicle concentrations constant across all treatment groups. CMV-ERα contains the full-length coding sequence of human ERα (595 amino acids). CMV-ERβ contains the full-length coding sequence of human ERβ (530 amino acids). ERE-tk-luciferase contains a single ERE cloned upstream of the thymidine kinase promoter and the luciferase gene (gift of Dr. Don DeFranco, Department of Pharmacology and Chemical Biology, University of Pittsburgh). CMV-SPORT-βgalactosidase was purchased from Invitrogen (Carlsbad, CA).
Cells were grown to 85% confluence in T75 flasks in complete tissue culture media. 201T and A549 nuclear and cytosolic extracts were prepared using the Nuclear Extraction Kit (Panomics, Redwood City, CA) according to the manufacturers’ instructions. H23 nuclear, membrane, and cytosolic extracts were prepared using the Proteoextract Subcellular Proteome Kit (Calbiochem, San Diego, CA) according to the manufacturers’ instructions. Proteins were quantified using the BCA Protein Assay kit (Pierce Biotechnology, Rockford, IL). Equivalent amounts of protein from each sample were separated by electrophoresis through pre-cast 10% Tris-HCl polyacrylamide gels (BioRad, Hercules, CA) and transferred to PVDF membranes. Nonspecific binding sites were blocked by incubation in 1 X TBS-T (0.05% Tween 20, 150 mM NaCl, 10 mM Tris, pH 7.4)/5% milk for 1 h at room temperature, followed by incubation overnight at 4°C with a 1:1000 dilution of anti-ERα antibody (sc-544 from Santa Cruz Biotechnology, Santa Cruz, CA), a 1:500 dilution of anti-ERα antibody (sc-543 from Santa Cruz Biotechnology), or a 1:1000 dilution of anti-ERβ antibody (05–824 from Upstate, Lake Placid, NY). The blots were then washed in 1 X TBS-T and then incubated with horseradish peroxidase-conjugated anti-rabbit IgG (Amersham, Piscataway, NJ). Immune complexes were detected using ECL Chemiluminescent substrate (Pierce Biotechnology), followed by exposure to autoradiography film. Blots were stripped with Restore Stripping Buffer (Pierce Biotechnology) and reprobed with a 1:1000 dilution of anti-PARP antibody (sc-7150 from Santa Cruz Biotechnology) or a 1:10,000 dilution of a pan actin antibody (MAB1501 from EMD Biosciences San Diego, CA).
To study MAPK phosphorylation, 201T cells were grown to 75% confluence in T75 flasks. Cells were washed two times with 1X PBS followed by serum deprivation for 48 h in phenol red-free media. E2, PPT, or DPN was added for the times indicated. Whole cell protein extracts were prepared using ice-cold radioimmunoprecipitation buffer as described previously . Equal amounts of protein (35 µg) for each sample were separated by size on a pre-cast 10% SDS-tricine polyacrylamide gel (Invitrogen). Nonspecific binding sites were blocked as described above. The primary antibody was a 1:1000 dilution of anti- phospho-p44/p42 mitogen-activated protein kinase (MAPK) monoclonal antibody (Cell Signaling Technology, Beverly, MA) and secondary antibody was a 1:2000 dilution of horseradish peroxidase conjugated anti-mouse IgG (GE Healthcare Life Sciences, Piscataway, NJ). Washes and detection were performed as described above. Blots were stripped and reprobed for total MAPK protein. Quantitation of the signals was done by densitometric scanning and ImageQuaNT analysis (Molecular Dynamics ImageQuaNT software version 5.2, Sunnyvale, CA).
For transient transfection assays in which the activity of exogenous receptor was measured, cells were plated in 6-well dishes at a density of 4×105 cells per well in phenol red-free medium containing 10% CSS. The next day, the cells were transfected using Lipofectamine reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. Transfections contained 0.75 µg of ERE-tk-luc reporter plasmid, 0.1 µg of CMV-ER expression vector, and 0.5 µg of CMV-β-gal. After 16 h, the medium was replaced and the cells were treated with vehicle or ligand for an additional 24 h. Cells were harvested and assayed for luciferase as indicated below.
For transient transfection assays in which the activity of endogenous receptor was measured, cells were plated in 6-well dishes and transfected using Lipofectamine, as above. Transfections contained 1.5 µg of ERE-tk-luc reporter plasmid, 0.5 µg of CMV-β-gal, and 0.5 µg of empty vector. After 5 h, the medium was replaced with fresh medium containing vehicle or ligand. The next day, cells were harvested and assayed for luciferase as indicated below.
Cells were harvested and washed twice with PBS. The washed cell pellet was lysed in 100 µL of l× Reporter Lysis Buffer (Promega, Madison, WI). Complete cell lysis was achieved using one freeze-thaw cycle that consisted of a 10 min incubation at −80°C followed by a rapid thaw at 37°C. Cell extracts were transferred to a microcentrifuge tube and clarified by centrifugation for 3 min at 12,000 × g at room temperature. The supernatants were transferred to a fresh tube and assayed using the Luciferase Assay System (Promega). For each assay, 30 µL of extract was diluted with 70 µL of l× Reporter Lysis Buffer. Luminescence was read using an AutoLumat LB953 luminometer (Berthold, Pforzheim, Germany).
The luciferase activity was normalized to the β-gal activity to control for transfection efficiency and cell recovery. β-gal activity was measured by diluting a fixed volume of cell extract in 1× Reporter Lysis Buffer to a final volume of 0.1 mL. The resulting samples were mixed with 0.1 mL of 2× β-gal Assay Buffer (200 mM NaPO4 buffer pH 7.3, 2 mM MgCl2, 100 mM β-mercaptoethanol, 1.33 mg/mL ortho-Nitrophenyl-β-galactoside) and incubated at 37°C until yellow color developed. The reactions were terminated by addition of 0.7 mL of 1 M Na2CO3 solution, and the optical density was measured at 420 nm.
201T cells were seeded in complete tissue culture medium in 96-well plates. After overnight attachment, the tissue culture medium was removed. The adherent cells were washed twice with sterile PBS, and the cells were cultured for an additional 48 h in phenol-red free tissue medium supplemented with 0.1% CSS. The cells were then treated with the indicated ligands for a total of 48 h. The effect of treatment on cell growth was determined by MTT assay. The MTT assay kits were purchased from ATCC, and assays were conducted essentially as described by us previously . The % Cell Growth was calculated using the following equation: % Cell Growth = 100[(O.D. treatment-O.D.blank)/O.D.vehicle-O.D.blank)]
Female ovariectomized nude mice (10 per group) were treated with 500 µg PPT, 250 µg DPN or vehicle control by s.c. injection 5 times per week. Ligand doses were chosen based on those reported by Frasor J. et al . All compounds were dissolved in DMSO then diluted 1:10 in corn oil for injection in 100 µl volume. Following 1 week of ligand treatment, 201T NSCLC cells (2×106) were injected into the flank of each animal. Ligand treatment was continued for 4 additional weeks. Tumor volume was measured two times per week using digital calipers and reported as an average relative tumor volume (mm3). Animal studies were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh and were conducted in strict compliance with institutional guidelines.
Data were graphed and analyzed using GraphPad Prism 4.0 software. The Student’s t-test or ANOVA was used for the statistical analysis of the in vitro studies. All values are expressed as the mean ± SD. Significance tests were performed with two-sided significance level 0.05. For the in vivo study, we used a repeated measures (mixed effects regression) model to analyze the data. Based on the Bayes Information Criterion, the best model fit was achieved when tumor volume was logarithmically transformed, and when predictor variables included a linear term for the day and the interaction between day and treatment group. We also evaluated a quadratic model using the raw outcome value without transformation.
As a first step toward dissection of the relative roles of ERα and ERβ in mediating genomic and non-genomic signaling, fractionation studies were conducted to determine the subcellular localization of each receptor in NSCLC cells. We used 201T adenocarcinoma cells, H23 adenocarcinoma cells, and A549 bronchioloalveolar cells for these studies because their in vivo growth is increased by E2 and decreased by therapeutic agents that inhibit estrogen signaling [11–13, 27]. MCF-7 breast cancer cells which express both ERα and ERβ were included as a positive control for receptor expression. Cells were grown to near confluence and then were harvested and fractionated. Expression of ERα and ERβ was analyzed in each fraction by immunoblot. Full-length ERα protein (66 kD) was detected in the nuclear fraction of MCF-7 cells but not in 201T or A549 cells (Fig. 1A). Full-length ERα was also not detected in H23 cells, even when a different fractionation procedure and two antibodies that recognize distinct ERα epitopes (hinge and C-terminus) were utilized (Fig. 1B). Smaller immunoreactive bands of ≈ 42 kD and ≈ 54 kD were observed in the membrane and cytosolic fractions of H23 cells, respectively. Although we cannot exclude the possibility that these represent variant forms of ERα, they do not share common epitopes or subcellular localization and may represent non-specific cross-reactive bands. In contrast to ERα, full-length ERβ protein (59 kD) was detected in the nuclear and cytosolic fractions of 201T and A549 cells. The localization of ERβ to both the nuclear and cytosolic fractions suggests it has the potential to impact both genomic and non-genomic signaling pathways in lung cancer cells.
We planned to use well-characterized ER subtype selective agonists to determine the functional consequences of selective activation of endogenous ERβ (or ERα) in lung cancer cells. We first confirmed that the ligands displayed the expected receptor selectivity in 201T cells by measuring transcriptional responses in transient transfection assays. To do this, 201T cells were co-transfected with an expression vector that encodes either ERα or ERβ and an ERE-tk-luciferase reporter construct. Following transfection, the cells received vehicle (controls) or were treated with increasing concentrations of E2, the ERα-selective agonist PPT, or the ERβ-selective agonists DPN or GEN (Fig. 2). In these assays, luciferase reporter gene activity serves as a measure of ER-mediated transcription. E2 stimulated reporter gene activity in 201T cells in the presence of both ERα and ER β, whereas PPT activated transcription only in the presence of ERα. A maximal ERα driven transcription response was observed at 10 nM PPT, although receptor specificity was maintained up to the highest concentration of PPT tested (100 nM). DPN and GEN selectively increased ERβ transcription at ligand concentrations ≤ 10 nM. However, when the ligand concentration was increased to 100 nM, both DPN and GEN modestly increased ERα transcription. These data indicate that PPT functions as an ERα-selective agonist in 201T cells at ligand concentrations up to 100 nM, whereas DPN and GEN function as ERβ-selective agonists in 201T cells at concentrations ≤ 10 nM.
To confirm that the effects of GEN and DPN on ERE-mediated transcription were dependent on the classical ER, we examined the ability of the pure anti-estrogen fulvestrant  to block transcription activation. As above, 201T cells were co-transfected with an ERβ expression vector and the ERE-tk-luciferase reporter construct. The next day, cells were untreated or were treated with E2 ± fulvestrant, GEN ± fulvestrant, or DPN ± fulvestrant (Fig. 3). As expected based on the results presented in Fig. 2, E2, GEN, and DPN treatment induced a statistically significant increase in reporter gene activity compared to vehicle controls. The stimulatory effect of each ligand was abrogated in the presence of fulvestrant. This result confirms that the transcriptional response to GEN and DPN in 201T cells is mediated via the classical ER.
We next determined the effect of the subtype-selective ligands on transcriptional responses mediated by the endogenous ER. To do this, 201T cells were transiently transfected with ERE-tk-luciferase alone. The cells were left untreated or were treated with increasing concentrations of E2, PPT, or GEN (Fig. 4). The concentrations of each ligand that were utilized for these endogenous receptor activation studies were based on the results presented in Fig. 2. Luciferase activity was significantly increased by 10 nM E2 (p = 0.0085 versus vehicle) and 10 nM GEN (p = 0.0068 versus vehicle) but not by PPT. At a final ligand concentration of 10 nM, reporter gene activity increased a mean 1.6-fold in the presence of E2 and 1.5-fold in the presence of GEN.
In addition to signaling via the classical transcription pathway, the ER has been implicated in a non-classical signaling pathway involving the rapid, non-genomic activation of MAPK . This non-genomic pathway allows for cross-talk between the ER and growth factor signaling pathways that contribute to NSCLC cell growth [12, 13]. To determine whether ERα or ERβ mediates non-genomic signaling, 201T cells were cultured for 48 h in phenol-red free, serum-free medium. Then, the cells were incubated for increasing times (0–240 min) with either vehicle, 10 nM E2, 10 nM DPN, or 100 nM PPT. Whole cell extracts were prepared and analyzed by immunoblot for MAPK phosphorylation (Fig. 5A). As shown in Fig. 5B, the magnitude of induction of MAPK phosphorylation was greatest in 201T cells treated with E2 (7.5-fold). However, DPN and PPT also increased MAPK activation (3.7-fold and 2.5-fold, respectively), and the kinetics of activation were similar for E2, DPN and PPT (Fig. 5B). Phosphorylation of MAPK was maximal within 5 min of treatment and returned to near-baseline levels within 30 min. Similar results were obtained in A549 NSCLC cells treated with E2, PPT and DPN and in 201T cells treated with GEN (data not shown).
Finally, we examined the effect of the receptor subtype selective ligands on the growth of 201T cells. For in vitro studies, we determined whether the ligands stimulated cell growth as single agents and whether they cooperated with EGF to increase cell growth. To do this, the cells were cultured for 48 h in phenol-red free medium with 0.1% CSS. Then the cells were treated with vehicle, 10% FBS (as a positive control), EGF, each of the ER ligands alone, or each of the ER ligands plus EGF. Effects on growth were determined 48 h post-treatment by MTT assay (Fig. 6). Compared to vehicle control, FBS and EGF significantly increased cell growth, but E2 alone, DPN alone, and PPT alone had no consistent stimulatory effect (Fig. 6A). When combined with EGF, E2 and DPN but not PPT mediated a significant, further increase in cell growth (Fig. 6B). These data support the conclusion that ERβ cooperates with EGF to promote the proliferation of 201T cells. We also tested the ability of PPT and DPN to promote lung tumor growth compared to control treated animals in an in vivo xenograft model (Fig. 6C). When 201T cells were implanted into female, nude, ovariectomized mice, tumor growth was stimulated by DPN but not PPT (p = 0.05; DPN versus vehicle control). By day 33 of ligand treatment, a 1.5-fold increase in tumor volume was observed in mice treated with DPN versus control treated mice.
Although the role of estrogen as a proliferative stimulus in breast cancer is well established, its possible role in lung cancer has only more recently been studied. Several biological responses to E2 have been observed in lung cancer cells including growth stimulation, alteration of gene expression, and phosphorylation/activation of MAPK [7, 8, 10, 12. In addition, the estrogen signaling pathway has been found to interact with the epidermal growth factor receptor signaling pathway [12, 13] or with tobacco carcinogens to stimulate the proliferation of small airway epithelial cells  and lung cancer cells  The growth-promoting effects of estrogen on lung cancer precursors and lung cancer cells provide a rationale for the further evaluation of agents that disrupt estrogen signaling pathways in lung cancer chemoprevention and treatment models.
Theoretically, more precise therapeutic targeting could be achieved by identification of the ER subtype that mediates the effects of estrogen in lung cancer. However, ERα and ERβ expression in lung cancer cells has been variably reported, making it difficult to identify the receptor that is responsible. To assist in the identification of the receptor responsible for genomic and non-genomic signaling in NSCLC cells, we used immunoblot assays to examine the subcellular distribution of each receptor and studied the effect of ligands that selectively activate either ERα or ERβ. Consistent with our previously published data using whole cell extracts , we did not observe expression of full-length ERα protein in nuclear or cytoplasmic extracts prepared from NSCLC cells (Fig. 1). In contrast, ERβ was detected in both the nuclear and cytoplasmic fractions of two different E2-responsive cell lines, suggesting that it may participate in both genomic and non-genomic signaling. Consistent with this idea, we find that ERβ-selective ligands significantly stimulate reporter gene transcription and increase MAPK phosphorylation. Together, this data leads us to conclude that endogenous levels of ERβ are sufficient to generate both the genomic and non-genomic response to estrogen in 201T lung cancer cells.
To evaluate the ability of endogenous ER to mediate genomic signaling, we used a luciferase reporter plasmid that contains a single ERE in the promoter region. With this construct, we observed a significant and reproducible increase in transcription in response to E2 addition (Fig. 4 and data not shown). The magnitude of E2 induction of reporter gene activity in 201T cells was modest (1.5 fold), but was consistent with what was observed in a different lung cancer cell line that expresses only ERβ . In contrast to our findings, Zhang et al. recently reported that E2 does not activate transcription from an ERE-luciferase reporter plasmid in NSCLC cells that express ERβ but not ERα . We speculate that the greater fraction of nuclear ERβ protein we observe in our cell lines may (A) contribute to the disparate findings and (B) be more representative of the clinical situation because 45%-69% of lung cancer cases (analyzed across 4 studies that included at least 100 patients each) express a nuclear pool of the receptor [17–20].
We are unable to make any conclusions regarding the contribution of endogenous ERα signaling to the cellular response to estrogen because our cell lines do not express detectable levels of ERα protein (Fig. 1). However, one other study examined the relative roles of endogenous ERα or ERβ in supporting lung cancer cell proliferation by transiently transfecting siRNA duplexes targeting each receptor into H23 cells . Unlike our findings, their H23 cells express both ERα and ERβ. Complete siRNA-mediated suppression of mRNA corresponding to either receptor resulted in approximately 20% inhibition of cell growth, suggesting that ERα and ERβ contribute equally to the growth of these H23 cells. The ability of the siRNA to attenuate either the genomic or non-genomic responses to E2 was not examined, so that it is not possible to state whether the two receptors have redundant or independent functions with regard to these activities.
Surprisingly, we observed that PPT induced MAPK activation in 201T cells despite the absence of any detectable full-length ERα protein. Because we used two antibodies that were raised against distinct epitopes of ERα to analyze protein expression (Fig. 1A and Fig. 1B), we consider it unlikely that there is a small functional pool of ERα protein that escaped detection in our immunoblot studies. Although PPT is an ERα-selective ligand, it binds ERβ with low affinity . Possibly, this binding is sufficient to nominally activate non-genomic signaling from ERβ by dissociating the receptor from the repressive HSP90 complex, allowing for some MAPK phosphorylation. Although PPT may bind ERβ and activate non-genomic signaling, it clearly does not activate transcription from an ERE in the presence of ERβ (Fig. 2B and Fig. 4, , ). PPT also does not cooperate with EGF to support 201T cell growth in vitro (Fig. 6A) nor does it stimulate the growth of 201T cells in vivo (Fig. 6C). A model consistent with all of our data is that activation of non-genomic signaling (without a corresponding genomic response) is insufficient to induce a proliferative response to estrogen in these lung cancer cells.
In conclusion, our data provide evidence that endogenous levels of ERβ are sufficient to generate genomic and non-genomic responses to estrogen in 201T cells and suggest that activation of both of these pathways may be necessary to support cell proliferation. To determine the in vivo significance of these findings, studies are in progress to quantify the effect of ER subtype-specific ligands on gene activation and MAPK phosphorylation in the 201T xenograft model.
The authors wish to thank members of the University of Pittsburgh Cancer Institute writing group for reviewing this manuscript in preparation for its submission. L.P. Stabile was supported by a Young Clinical Scientist Award from the Flight Attendant Medical Research Institute.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1This work was supported by grant P50 CA90440 and, in part, under a grant with the Pennsylvania Department of Health. The Department specifically disclaims responsibility for any analyses, interpretations or conclusions.