Abstract
The issue of how progesterone affects mammary gland growth is controversial, and the mechanism governing the effects of the hormone remains mostly unknown. We have previously shown that G protein-coupled receptor 30 (GPR30) is a progestin target gene whose expression correlates with progestin-induced growth inhibition in breast cancer cells. In this study, we investigate the role of GPR30 in regulating cell proliferation and mediating progestin-induced growth inhibition. When progestin failed to inhibit the growth of MCF-7 cells and instead stimulated growth, GPR30 was down-regulated. In this way, the inhibitory or stimulatory affects that progestin has on proliferation correlated with the level of expression of GPR30. Transient expression of GPR30 resulted in a marked inhibition of cell proliferation independent of estrogen treatment. GPR30 antisense was used to evaluate the role of GPR30 expression in progestin-induced growth inhibition. A diminished GPR30 mRNA expression by the antisense stimulated growth. Interestingly, GPR30 antisense abrogated the growth inhibitory effect of progestin and progesterone. Indeed, progestin induced 1) a reduction in cell proliferation, 2) G1-phase arrest, and 3) down-regulation of cyclin D1 was diminished. These data suggest that the orphan receptor, GPR30, is important for the inhibitory effect of progestin on growth.
PROGESTERONE IS A SEX steroid hormone that regulates many reproductive functions. In the uterus, progesterone inhibits estrogen-induced growth, but in the mammary gland the way progesterone acts is not fully understood. Progesterone is required for the formation of the lobular-alveolar structures in a normal mammary gland during pregnancy, suggesting it plays a role in growth stimulation (1). It is also established that DNA synthesis and mitosis are increased in the late luteal phase of the menstrual cycle (2, 3) and progesterone levels in serum correlate with breast tissue proliferation (4). Furthermore, when postmenopausal women are treated with steroid hormones, addition of progestins to estrogens increases the rate of proliferation evidenced by antiproliferating cell nuclear antigen (anti-PCNA) and Ki67 measurement (5). Addition of a progestin to estrogen increases also mammographic density, presumably reflecting a greater number of cells present (6).
Progestin’s critical role in mammary gland carcinogenesis is established in mice (7). In humans, the addition of progestin to hormone replacement therapy increases the risk of breast cancer (6, 8–10). On the other hand, progestins inhibit the growth of breast cancer and are thus commonly used drugs in breast cancer treatment (11).
There is a number of in vitro studies showing that progestin inhibits growth in breast cancer cells, but results that contradict these studies are presented as well (12–16). When normal mammary gland epithelial cells are cultured, progestin and progesterone inhibit proliferation (17–19), whereas progesterone increases DNA synthesis in organ culture (20). In conclusion, the growth stimulatory effect of progestin is more an in vivo phenomenon than an in vitro phenomenon, and the suppressive effects seem to dominate in vitro. The explanation for this discrepancy is unknown but might be explained by the influence of the paracrine environment.
The mechanism whereby progestin regulates cell growth has been studied mainly by analyzing cell cycle-regulating proteins. Progestins have been shown to first stimulate the entry of breast cancer cells to the cell cycle and then arrest cells to the late G1 phase (21, 22). After up-regulation of the cell cycle regulatory molecules, progestin-induced growth arrest is accompanied by down-regulation of cyclin D1, D3, A and B (21, 23), which all control the cell-cycle progression by activating cyclin-dependent kinases (Cdk). Cyclin D1 has a causative role in breast cancer formation (24, 25), and it is overexpressed in several mammary carcinomas (26, 27). In a recent study, cyclin D1 overexpression was able to confer progestin resistance, which provides evidence of cyclin D1 having a critical function in progestin-induced growth inhibition in breast cancer cells (28). Based on the biphasic response to a pulse of progesterone and detailed analysis of cell cycle regulatory molecules, it is hypothesized that the effect of progestin is dependent on the exposure time of progestin: intermittently administered progestins would be growth stimulatory, but when continuously administered progestins are growth inhibitory (29). It is further proposed that progestin drives cells to a decision point at the G1/S boundary and thereafter induces cellular changes that determine the ultimate response of the cells.
Progestins can regulate cell growth through autocrine and paracrine factors (30). One of the best examples of the paracrine effect of progesterone is a chimeric mammary gland epithelium composed of PR-negative and PR wild-type cells that have a normal alveolar development to which PR-negative cells are contributed (31). Although progestins are known to inhibit cell growth, progestin-induced inhibitory growth factors or their receptors remain to be identified.
We have previously studied G protein-coupled receptor 30 (GPR30) regulation by different steroid hormones (32). In this current study, we characterized the role of the orphan transmembrane receptor GPR30 in the growth regulation of breast cancer cells. Transient expression of GPR30 inhibited growth and an antisense construct of GPR30 reversed the growth-inhibitory effect of progestin. These results reveal a growth inhibitory receptor, GPR30, whose expression is important for the growth inhibitory effect of progestin in MCF-7 breast cancer cells.
Materials and Methods
Hormones
17β-Estradiol, medroxyprogesteroneacetate (MPA), dexamethasone, and flutamide were provided from Sigma (St. Louis, MO); progesterone from Merck (Darmstadt, Germany); insulin from Sigma and promegestone (R5020) from Schering Aktiengesellschatt (Berlin, Germany). Mifepristone (RU486) was a gift from Roussel Uclaf (Paris, France).
Cell culture
MCF-7 cells were cultured in phenol red free DMEM/F12 medium supplemented with 5% dextran-coated and charcoal-stripped FBS and the hormones: 1 μg/ml insulin or with 10 ng/ml insulin and 1 nm estrogen. In transfection studies the cells were incubated with 2 μg/ml tetracycline from 2 d to two passages. If indicated, cells were serum-deprived overnight to 24 h.
Cell growth assay
Cells (1–3 × 103) were seeded in 96-well plates and allowed to attach for 2 d. Thereafter, appropriate steroid hormones in 100% ethanol or ethanol were added. Relative cell numbers were measured using the crystal violet nuclei staining method (33). Absorbance was measured at a wavelength of 590 nm using a Victor 1420 Multilabel counter (Wallac, Inc., Turku, Finland).
Northern blot
Total RNA (40 μg) was electrophoresed on a denaturing gel and blotted onto nitrocellulose membrane (Micron Separations, Inc., Westborough, MA) using overnight capillary transfer in 10× saline/sodium phosphate/EDTA (pH 7.7). In vitro-transcribed mRNA was used as a probe. Probes were made from the pBK/cytomegalovirus (CMV)-GPR30 plasmid digested with restriction enzymes (SmaI or SalI) as described by the manufacturer (Ambion, Inc., Austin, TX). Sense GPR30 was detected with the antisense probe and antisense GPR30 was detected with the sense probe. Hybridization was carried out as described (32). Filters were rehybridized with oligonucleotide complementary to 18S ribosomal RNA (Ambion, Inc.) for normalization. Radiolabeled filters were exposed on x-ray film [Kodak BioMax (Rochester, NY)].
Proliferation assay
The proliferation assay was carried out as previously described using bromodeoxyuridine (BrdU) and immunohistochemistry (34, 35). In the transfection assay, fixated cells were stained for 15 h with 1 ml X-Gal staining solution [1 mg/ml 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (Sigma), 10 μm MgCl2, 3 mm K4Fe(CN)6 × 3H2O, 3 mm K3Fe(CN)6]. The cells were incubated with monoclonal anti-BrdU (Sigma) 360 μg/ml or Ki-S5 (Sigma) 1 μg/ml in PBS containing 0.5% BSA and 100 U/ml Nuclease S1 (Promega Corp., Madison, WI) for 1 h, and immunochemically stained using a Histostain-plus Bulk Kit (Zymed Laboratories, Inc., South San Francisco, CA). Peroxidase activity was measured using AEC+ Large Volume High Sensitivity Substrate-Chromogen System as recommended (DAKO Corp., Carpinteria, CA).
Transient transfection
MCF-7 cells (105) were transfected 24 h with (2 μg) pBk-CMV (Stratagene, La Jolla, CA) with and without a GPR30 insert and (1 μg) pCMVβGal using 3 μl Lipofectamin 2000 (Life Technologies, Inc.) as the manufacturer recommended.
LightCycler RT-PCR analysis
One-step RT-PCR was performed with a LightCycler instrument (Roche, Mannheim, Germany) using 100 ng total RNA and LightCycler RNA Master SYBR Green I kit (Roche) as previously described (32). The primer pares result in PCR products of 240 bp (GPR30) and 243 bp (TBP control gene). Quantitative analysis of the LightCycler data were performed employing LightCycler analysis software. The final results, expressed as N-fold differences in GPR30 gene expression between untreated and MPA treated samples, were determined as follows: NGPR30 = (GPR30treated/TBPtreated)/(GPR30untreated/TBPuntreated).
Construction of GPR30 antisense expressing MCF-7 cells
An MCF-7 cell line (named PMCF-7) expressing constitutively tetracycline-regulated transactivator was established. Cells were cotransfected with plasmids pUHD15–1, which expressed the tetracycline-regulated transactivator (tTa), and tTA transactivator responsive promoter (tetO-CMV) plasmid using liposomes (36). tetO-CMV codes the puromycin resistance gene under the control of tTa-responsive promoter, which allows the positive selection of a primary cell line functionally expressing tTa (37). Single clones were further characterized by transiently transfecting a plasmid tetracycline response element-luciferase-plasmid (pTRE-Luc) (CLONTECH Laboratories, Palo Alto, CA) reporter. The luciferase activity was measured using transfected tetracycline-treated cells as a control.
The full-length cDNA clone of GPCR-Br/GPR30 was cloned in the antisense orientation into the NotI and BamHI site of the pCEP4-TET vector. pCEP4-TET is a derivative of the episomal expression vector pCEP4 (Invitrogen, Groningen, The Netherlands), in which the CMV immediate early enhancer/promoter has been replaced by the heptamerized Tet operator sequences upstream of a minimal CMV promoter (38). The PMCF-7 cell clone was stable transfected with a pCEP4-TET vector without insert or with pCEP4-TET containing GPR30 in the antisense orientation. Two days after transfection 200 μg/ml Hygromycin (Sigma) was added to a select pool of PMCF-7 cells which contained pCEP4-TET vector for a total of 3 wk of selection. Individual clones were not further separated.
Flow cytometry
MCF-7 cells (105) were plated in T25 flasks and harvested in trypsin: EDTA. The cells were washed with PBS, suspended in a 2 ml cell lysis buffer [0.1% Triton X-100, 0.1% Na-citrate, 50 μg/ml propidium iodide (Sigma) in Q-water], incubated overnight at 4 C, and filtered through 30- μm nylon mesh. The samples were analyzed in a flow cytometer using a FACScan cell sorter and the data analyzed using CELLFIT program software (Becton Dickinson and Co., Franklin Lakes, NJ).
Immunoblotting
Immunoblotting was carried out as previously described (39). Cells were harvested by a cell scraper and calculated in Burker’s cell chamber. Cell samples were mixed with 3 vol sodium dodecyl sulfate-sample buffer. The viscosity of the samples was reduced by drawing samples through a 23-gauge needle. Thereafter, samples were boiled for 5 min. Equal amounts of cells (3 × 105) were resolved in 12% polyacrylamide gel and transferred to nitrocellulose membrane with an electrophoretic transfer apparatus. After blocking the membranes were incubated overnight at 4 C with antibody cyclin D1-HD11 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or β-actin (Sigma), and washed. Peroxidase-conjugated goat antirabbit IgG (Capperl, West Chester, PA) was used as a secondary antibody. After washing, labeled proteins were detected by enhanced chemiluminescense.
Results
Transient expression of GPR30 inhibited proliferation of MCF-7 cells
To study the effect of GPR30 on cell growth, MCF-7 cells were cotransfected with GPR30 in pBk-CMV plasmid (or pBk-CMV plasmid) and pCMVβ-Gal reporter construct, which expresses β-galactosidase. After pulse labeling cellular DNA with BrdU for 2 h, cells were immunohistochemically stained with the anti-BrdU antibody. Transfected cells were visualized by β-galactosidase (produced in transfected cells) staining. The percentage of BrdU-positive cells in β-galactosidase-expressing cells was counted. Results showed that transient expression of GPR30 inhibited estrogen-stimulated proliferation 44%, 40%, and 39% at 6, 12, and 30 h, respectively (Fig. 1A).
Transient expression of GPR30 inhibited proliferation of MCF-7 cells. MCF-7 cells were cotransfected with GPR30 in pBk-CMV plasmid (or pBk-CMV) and with pCMVβ-Gal reporter construct expressing β-galactosidase for 24 h under serum-free conditions. After transfection (time point 0 h), the serum-containing culture medium and BrdU (for 2 h) were added. Thereafter, cell proliferation was measured at the indicated time points. Cells transfected with the plasmids were stained with X-Gal substrate. BrdU was stained using immunohistochemistry and AEC+ as a substrate. The rate of proliferation was calculated both in vector-transfected cells and in GPR30-transfected cells. A, Cells were continuously cultured with 1 nm estrogen. B, Cells were continuously cultured without steroid hormones (1 μg/ml insulin). Results are means of three replicates, repeated three times. Data were analyzed by the Student’s paired two-tailed t test. Differences of P < 0.05 were considered to be significant (*); P < 0.01 (**) and P < 0.001 (***), highly significant. The values were compared with vector-transfected control cells.
Progestin is shown to inhibit estrogen-stimulated growth in breast cancer cells, but equally, without estrogen growth is stimulated. Thus, we asked whether GPR30-induced growth inhibition is dependent on estrogen treatment. As shown in Fig. 1B, the transient expression of GPR30 inhibited proliferation in cells grown without estradiol: 55%, 65%, 85%, 86%, and 70% between 0- and 30-h time points. The result indicated that the growth inhibitory response of GPR30 is irrespective of estrogen treatment.
Despite that estrogen-induced higher degree of proliferation than insulin, without estrogen cells attached to cell culture bottles more rapidly and started to proliferate. This would explain the growth difference between control cells at 0 h in panels A and B.
GPR30 antisense-stimulated growth
To further study the role of GPR30 in cell growth regulation, an antisense construct of GPR30 under tetracycline regulation was stable-transfected in MCF-7 cells. PMCF-7 clone expressing tetracycline-regulated transactivator was characterized by transiently transfecting a luciferase reporter plasmid, and a clone showing a 50-fold increase in luciferase activity compared with tetracycline-treated control cells was transfected with the GPR30 antisense plasmid. The antisense mRNA was expressed in a PMCF-7 cells and this decreased the expression of GPR30 mRNA (Fig. 2A). The administration of tetracycline down-regulated the expression of GPR30 antisense and allowed up-regulation of GPR30 mRNA (control cells).
A decreased GPR30 expression by the antisense abrogated progestin-induced growth inhibition. Cells expressing tetracycline-regulated transactivator were stable-transfected with pCEP/TET-GPR30 antisense vector. The cells were continuously cultured with estrogen and the control cells were treated with 2 μg/ml tetracycline (tet) to suppress GPR30 antisense expression. Two days after plating, cells were treated with 100 nm MPA or progesterone (prog). A, Total RNA (40 μg) was isolated from the antisense-transfected cells at 48 h, and used for Northern blot analysis. As a probe, in vitro transcribed mRNA from the pBK/CMV-GPR30 plasmid was used in sense and antisense orientation to detect the transfected antisense and endogenous sense GPR30 mRNA, respectively. As a loading control, the ribosomal 18S mRNA were used. In GPR30 antisense-transfected cells the relative number of cells was measured at the indicated time points after MPA (B) or progesterone (C) addition. D, The growth rates were compared between PMCF-7 cells stable-transfected with GPR30 antisense and vector (pCEP4-tet). E, The relative rate of cell growth was measured in vector stable-transfected PMCF-7 cells treated with and without tetracycline. The data (B–E) represent four replicates, repeated three times. In statistical analysis, comparison was made with tetracycline-treated control cells (B and C) or vector-transfected (D and E) control cells using paired t test as indicated in Fig. 1. The statistical values are marked in parentheses when a comparison was made between antisense cells treated with MPA and control cells.
When growth was compared between GPR30 antisense-expressing cells and control cells, antisense GPR30 (decreased GPR30 expression) enhanced cell growth (Fig. 2, B and C). We also stable-transfected PMCF-7 cells with an empty pCEP4 vector. The cells grew more slowly than antisense GPR30 cells, as expected (Fig. 2D), and tetracycline-treatment did not markedly affect growth (Fig. 2E). The results suggested that up-regulation of GPR30 inhibits growth of MCF-7 cells. The relatively small growth difference between tetracycline-treated and nontreated cells might be due to basic low-level expression of GPR30 in antisense cells.
GPR30 antisense accelerated first cell cycle
Progestin is known first to stimulate the cell cycle and then arrest cells in the late G1 phase in T47-D cells. We studied whether progestin and GPR30 affects cell cycle in a similar manner. Cells were first arrested in G0/1-phase by serum deprivation and the percentage of Ki-67 positive cells was counted between 0 and 10 h. GPR30 antisense increased the amount of Ki67 positive cells (Fig. 3A), suggesting that GPR30 diminishes proliferation during the first cell cycle. Progestin treatment increased the amount of Ki67 positive cells independent of GPR30 expression. The result implied that progestin accelerates, but GPR30 extends, the time cells need to enter the cell cycle (exit from G0-phase). The result also suggested that GPR30 expression should not be necessary for the enhancement of first cell cycle by progestin in MCF-7 cells.
Proliferation of MCF-7 cells expressing GPR30 antisense. Cells treated with (control) and without tetracycline were placed on glass coverslips, growth arrested overnight, and subsequently cultured with 100 nm MPA. A, To study how progestin and GPR30 antisense have an effect on the first cell cycle, the percentage of Ki-67 positive cells was counted between 0 and 10 h. The data presented are the means of three replicates, repeated three times. B, To study how GPR30 antisense has an effect on the progestin-induced inhibition of proliferation, cells were treated for 2 h with BrdU between 48 and 72 h, and visualized by immunostaining with anti-BrdU antibody. The data presented are the means of three replicates, repeated three times. Statistical significance was calculated using the t test and the values were compared with tetracycline-treated control cells, as indicated in Fig. 1. Statistical values are marked in parentheses when comparison was made between antisense cells treated with MPA and control cells.
GPR30 antisense abrogated MPA-induced growth inhibition
The finding that GPR30 mRNA expression correlated with progestin-induced growth inhibition (32), implied that GPR30 is involved in progestin-mediated growth regulation, which led us to study whether the inhibition of GPR30 expression compromises progestin effects. In the control cells (antisense cells grown in the presence of tetracycline), MPA inhibited growth by 40% and 43% at 120 h and 144 h, respectively (Fig. 2B); and in vector-transfected PMCF-7 cells (Fig. 2D) by 47% and 53% at 120 h and 144 h, respectively. Interestingly, in cells that expressed a high level of GPR30 antisense, MPA inhibited cell growth by only 14% and 8% at the time points of 120 h and 144 h, respectively (Fig. 2B). The effect of progesterone (Fig. 2C) and R5020 (results not shown) was compromised by the antisense GPR30 expression in a similar manner. There was a statistically significant difference (P < 0.01) between antisense and control cells, which were both treated with 100 nm MPA (P < 0.01) or 100 nm progesterone at d 3–6 (values are not shown in the figures).
GPR30 antisense compromised progestin-induced inhibition of proliferation
To further characterize the growth effect of GPR30 antisense, we used BrdU incorporation assay and flow cytometry analysis. BrdU-assay was employed at 48, 72, and 96 h after progestin addition to study the number of cells in the S-phase. Proliferation was increased in the GPR30 antisense-expressing cells compared with the cells treated with tetracycline (Fig. 3B). In the antisense GPR30-expressing cells MPA (100 nm) inhibited the proliferation rate only by 9%, 6%, and 4% at 48, 72, and 96 h, respectively; whereas in the tetracycline-treated control cells, MPA inhibited the amount of S-phase 3–10 times more and by 31%, 41%, and 39% between 48- and 96-h time points.
In flow cytometry analysis cells were growth arrested by overnight serum deprivation. Thereafter, cells were treated with 100 nm MPA. When measured 48 h after progestin addition, progestin decreased the number of cells in S-phase by 24% in tetracycline-treated control cells, and increased the number of cells in G0/G1-phase by 25% (Table 1). In antisense GPR30 expressing cells, progestin diminished the S-phase 9.5% and accumulated 8% of cells to the G0/G1-phase.
Flow cytometric cell cycle analysis of MCF-7 cells expressing GPR30 antisense under tetracycline regulation
| Cell cycle | Treatments | |||
|---|---|---|---|---|
| Estrogen | Estrogen, MPA | Estrogen, tetracycline | Estrogen, tetracycline, MPA | |
| G0/G1 | 37.3 ± 2.0 | 40.9 ± 2.4 | 37.8 ± 1.5 | 47.4 ± 1.7 |
| G2/M | 18.7 ± 2.5 | 19.3 ± 1.4 | 20.1 ± 0.2 | 20.7 ± 1.3 |
| S | 43.9 ± 4.5 | 39.8 ± 1.0 | 42.1 ± 1.3 | 32.0 ± 0.4 |
| Cell cycle | Treatments | |||
|---|---|---|---|---|
| Estrogen | Estrogen, MPA | Estrogen, tetracycline | Estrogen, tetracycline, MPA | |
| G0/G1 | 37.3 ± 2.0 | 40.9 ± 2.4 | 37.8 ± 1.5 | 47.4 ± 1.7 |
| G2/M | 18.7 ± 2.5 | 19.3 ± 1.4 | 20.1 ± 0.2 | 20.7 ± 1.3 |
| S | 43.9 ± 4.5 | 39.8 ± 1.0 | 42.1 ± 1.3 | 32.0 ± 0.4 |
Flow cytometric cell cycle analysis of MCF-7 cells expressing GPR30 antisense under tetracycline regulation
| Cell cycle | Treatments | |||
|---|---|---|---|---|
| Estrogen | Estrogen, MPA | Estrogen, tetracycline | Estrogen, tetracycline, MPA | |
| G0/G1 | 37.3 ± 2.0 | 40.9 ± 2.4 | 37.8 ± 1.5 | 47.4 ± 1.7 |
| G2/M | 18.7 ± 2.5 | 19.3 ± 1.4 | 20.1 ± 0.2 | 20.7 ± 1.3 |
| S | 43.9 ± 4.5 | 39.8 ± 1.0 | 42.1 ± 1.3 | 32.0 ± 0.4 |
| Cell cycle | Treatments | |||
|---|---|---|---|---|
| Estrogen | Estrogen, MPA | Estrogen, tetracycline | Estrogen, tetracycline, MPA | |
| G0/G1 | 37.3 ± 2.0 | 40.9 ± 2.4 | 37.8 ± 1.5 | 47.4 ± 1.7 |
| G2/M | 18.7 ± 2.5 | 19.3 ± 1.4 | 20.1 ± 0.2 | 20.7 ± 1.3 |
| S | 43.9 ± 4.5 | 39.8 ± 1.0 | 42.1 ± 1.3 | 32.0 ± 0.4 |
GPR30 antisense prevented MPA-induced cyclin D1 down-regulation
There is increasing evidence of the critical role of cyclin D1 in progestin-mediated growth in breast cancer cells. Accordingly, we examined whether the regulation of cyclin D1 and other cell cycle regulatory molecules by progestin is dependent on GPR30 expression. As shown in Fig. 4, the expression of GPR30 antisense prevented MPA-induced down-regulation of cyclin D1. The level of cyclin A or cylinD3 was not changed by GPR30 or progestin (results not shown).
Regulation of cyclinD1 expression in MCF-7 cells expressing GPR30 antisense and in cells treated with tetracycline to suppress GPR30 antisense expression. After plating, cells were growth arrested overnight, treated with 100 nm MPA, and lysed at 48 h. CyclinD1 expression was analyzed by Western blotting. As a loading control, samples were hybridized with an antibody against β-actin. The representative results were established three times.
Collectively, these results indicated that GPR30 mRNA expression is important for the growth-inhibitory property of progestins and progesterone in MCF-7 breast cancer cells. In the cells in which GPR30 expression is decreased, progestin-induced growth inhibition mediated by 1) reduction in the numbers of cells in S-phase; 2) increasing in the numbers of cells in G0/G1-phase; and 3) reduction of cell cycle regulator cyclin D1 were compromised.
Progestin-induced growth response correlated with GPR30 expression level
The results suggested that GPR30 has a critical function in progestin-mediated growth inhibition in cells which are grown in the presence of estrogen. Next we asked whether MPA inhibits growth in the absence of estrogen and in the presence of low levels of progesterone receptor (40). In fact, the opposite was found to be the case; as shown in Fig. 5A, MPA (100 nm) enhanced cell growth by 16%, 34%, and 24% at 72, 96, and 120 h, respectively.
Progestin stimulated estrogen-independent growth. MCF-7 cells were cultured without added steroid hormones (1 μg/ml insulin; control, c). A, Cells were plated at 96 wells and allowed to attach. Thereafter, MPA (1 nm, 10 nm, 100 nm) or a vehicle was added and relative number of cells was measured at the indicated time points using the crystal violet method. B, MCF-7 cells were plated on glass coverslips on 8-well plates. The cells were treated with 100 nm MPA, and proliferating cells were visualized by using BrdU and immunohistochemistry. C, MCF-7 cells were plated, treated with steroid hormones: 100 nm MPA (M), R5020 (R5), progesterone (p), DHT, dexamethasone (DEX), 1 μm antiprogestin RU486 (RU) and antiandrogen flutamide (F), and the relative number of cells was measured at 120 h. The data represent means of three (B) or four (A, C) replicates, repeated three times. The statistical significance was calculated using the Student’s paired t test and the values were compared with insulin-treated control cells as indicated in Fig. 1. The statistical values in panel A are marked in parentheses when a comparison was made between cells treated with 10 nm MPA and the control cells.
BrdU incorporation and cell proliferation was enhanced at 6 h, but not at the other time points investigated (Fig. 5B). Interestingly, the stimulation of cell growth was accompanied by down-regulation of GPR30 mRNA expression 8%, 70%, and 55% at 12, 24, and 48 h, respectively (Fig. 6). The results showed that GPR30 down-regulation is associated with progestin-induced growth stimulation.
GPR30 mRNA was down-regulated by MPA in MCF-7 breast cancer cells in the absence of estrogen. MCF-7 cells were continuously cultured without steroid hormones (1 μg/ml insulin; control). The cells were treated with 100 nm MPA and RNA was extracted at the indicated time points. GPR30 mRNA expression was analyzed using a quantitative LightCyclerRT-PCR. The results were normalized with the expression of the tata-box binding protein (TBP). The results are the means of two independent experiments.
Growth stimulatory effects of the steroid hormones were independent on GPR30 expression
As shown above, GPR30 down-regulation correlated with progestin-induced growth stimulation. We used GPR30 antisense-expressing cells to study whether progestin-induced growth stimulation (in cells grown without estrogen) is dependent on GPR30 expression. Progestin enhanced cell growth by an equal amount both in antisense GPR30- expressing and also in control cells (tetracycline-treated) (Fig. 7A).
Estrogen- and progestin-induced growth stimulation was GPR30 independent in MCF-7 cells. A, GPR30 antisense cells were cultured without estrogen (1 μg/ml insulin, c=control) and treated with tetracycline to suppress GPR30 antisense expression. The cells were plated at 96-well plates and treated with 100 nm MPA (M). The relative number of cells was measured at 120 and 144 h. The data represent the means of four replicates. B, GPR30 antisense expressing cells were precultured in the absence of steroid hormones (10 ng/ml insulin, c=control) with and without 2 μg/ml tetracycline, and after plating treated with 1 nm estrogen (E). The relative number of cells was measured 120 h after hormone addition. For the statistical analysis, a comparison was made with tetracycline-treated control cells using the Student’s t test as indicated in Fig. 1.
GPR30 is shown to be involved in estrogen-mediated MAPK activation. Thus, we studied whether estrogen-induced growth stimulation is affected by the antisense. In antisense GPR30-expressing and in tetracycline-treated (control) cells, estrogen (1 nm) stimulated growth 4-fold (Fig. 7B). These results might suggest that the down-regulation of GPR30 expression by MPA or estrogen (results not shown) is not causative for progestin- or estrogen-induced growth stimulation.
Progestins stimulated growth through PR
MPA is known to bind to androgen and glucocorticoid receptors in addition to PR. Thus, it was studied whether other steroids can also stimulate growth. Progesterone, R5020 and DHT stimulated growth, but less than MPA (Fig. 5C). Dexamethasone did not stimulate growth in MCF-7 cells. The affect of progesterone and R5020 was decreased by antiprogestin, and the affect of MPA was decreased by flutamide and RU486. This result illustrated that all progestins induced growth through PR, although MPA achieved an additional effect through the AR receptors.
Discussion
GPR30 is an orphan transmembrane receptor that shows some degree of similarity to chemokine receptors and is expressed primarily in endocrine tissues. Some G protein-coupled receptors are known to be involved in growth regulation, but the function of GPR30 was not known. Recently, GPR30 has been shown to regulate ERK1/ERK2 activity by regulating the amount of cAMP and through EGF receptor (41, 42). In a current study, we show that GPR30 plays a critical role in progestin-mediated growth inhibition.
We have previously established that GPR30 is a progestin target gene, whose expression correlates with progestin-induced growth inhibition (1). In this present study, we further characterized the effect of GPR30 on growth using antisense and overexpression approaches. Transient expression of GPR30 resulted in a marked growth inhibition in MCF-7 cells that was independent on estrogen treatment. The cells that expressed the GPR30 antisense construct in a tetracycline-regulated manner (tet-off-system) were used to study the regulation of cell cycle progression by GPR30. Expression of the antisense construct enhanced cell growth and increased the cells in S-phase, as revealed by BrdU staining. A decreased GPR30 mRNA expression also accelerated the entry of G0 cells into the cell cycle, as demonstrated by Ki-67 staining. These data suggest that GPR30 is involved in the regulation of the entry of cells into the cycle. This is consistent with previous studies in which progestin has been shown to arrest cells in the G1-phase (21, 22). The ligand of GPR30 is not known. Because GPR30-regulated growth was independent of the presence of estrogen and progestin, the ligand is likely to be in the serum or be secreted to the medium by MCF-7 cells. It is, however, possible that GPR30 is a true orphan receptor (without an endogenous ligand) and constitutively active.
Progestin, but not GPR30, enhanced the entry of the serum-deprived MCF-7 cells from G0 into the cell cycle. This is probably due to the fact that progestin could not inhibit growth at early time points, but rather stimulated growth—a phenomenon that is confirmed by other studies (21, 22). This further supports the theory that GPR30 expression is specifically involved in growth inhibition. In MCF-7 cells, progestin treatment caused a marked inhibition of cell growth, indicated by 1) lower cell numbers; 2) inhibition of the number of cells in S-phase; and 3) the accumulation of cells in G1-phase. Apoptosis was not induced after progestin treatment (results not shown), which suggested that the major growth inhibitory effect of progestin is due to the decrease in the rate of cell proliferation. This is in line with previous studies in which progestin and progesterone have been shown to induce apoptosis only at supraphysiological concentrations by down-regulating protective bcl-2 proteins and up-regulating p53 (43–45) in breast cancer cells.
The effect of progestin was studied in the antisense cells to reveal the role of GPR30 in steroid-regulated growth in MCF-7 cells. Interestingly, in cells where the expression of GPR30 was down-regulated by the antisense, the growth inhibitory effects of progestin were decreased. The reduction in cyclin D1 expression by MPA was also compromised when GPR30 expression was down-regulated by the antisense, which suggested that GPR30 is required for the regulation of cyclin D1 expression by progestin. However, the expression of cyclin A or cyclin D3 was not regulated by progestin or by GPR30 (results not shown). Thus, cyclin D1 regulation may have a significant role in progestin-mediated growth inhibition in MCF-7 cells. Most of the data concerning cyclin regulation by progestin have been established in T47-D breast cancer cells (21, 23).
These results suggested that progestin-induced cell growth inhibition is, at least partly, mediated through the G protein-coupled receptor 30. The role or mechanism of GPR30 in growth regulation has not been previously established. However, it has been demonstrated that GPR30 together with estrogen, activates the tyrosine kinase receptor, leading to ERK activation in ER negative breast cancer cells (41). Interestingly, the progesterone receptor can directly interact with the SH3 domain of tyrosine kinases, and that interaction is required for the activation of the ERK pathway and possibly for growth inhibition by progestin (46). Thus, the mechanism whereby GPR30 affect on growth could be due to the modulation of the activity of the MAPK cascade. However, there is no evidence for causative association with the MAPK cascade activation and the regulation of cell proliferation.
Although several studies have shown that progestin inhibits the growth of breast cancer cells (12–14), contradictory results have also been presented (12, 15, 16). In vivo studies are even more controversial, and recent evidence indicates that progestin is a growth stimulatory agent, and has tumor promoting potential (5, 7–9, 47, 48). The reason for the discrepancy is not known. In our studies, progestins only inhibited estrogen-stimulated growth in MCF-7 cells. In contrast, progestin enhanced cell growth without estrogen, which was mediated through PR and partly through AR in the case of MPA. The mechanism of growth stimulation remains to be discovered, since cell proliferation was stimulated only temporarily at 6 h and the rate of apoptosis was not diminished by progestin (results not shown). Interestingly, progestin-enhanced cell growth was associated with GPR30 down-regulation, but the finding that GPR30 antisense was not able to abrogate MPA-induced growth stimulation suggested that the down-regulation of GPR30 is not causative for cell growth acceleration. Estrogen-induced proliferation was also independent of GPR30 expression, which suggested that GPR30 might not mediate steroid hormone-induced proliferation.
These results imply that GPR30 is a receptor whose expression is critical for the growth inhibitory effect of progestin, and that progestin could exert both stimulatory and inhibitory effects on proliferation, depending upon the level of expression of GPR30. Thus, the differential regulation of GPR30 receptor in various paracrine environments may explain some of the conflicting effects of progestin in the mammary gland. It can further be hypothesized that progestin, after it first stimulates the cell cycle, induces genes involved in membrane initiated signaling and results a growth inhibition. Data suggest a novel mechanism whereby steroids inhibit cell proliferation, and might serve as a target for therapeutic intervention when treating proliferative disorders or for countering the tumorigenic properties of steroid hormones.
Acknowledgments
We wish to thank Prof. K. Saksela for critical comments; and J. Hinkka, H. Syvälä, and H. Mäkinen for their help. Plasmid tetO-CMV was provided by Prof. P. Pognonec, and GPCR-Br/GPR30 by Dr. D. Thompson—our gratitude to them.
This work has been supported by the Medical Research Foundation of Tampere University Hospital, Biomed 2 Project PL 963433, and the Cancer Foundation in Pirkanmaa.
Abbreviations
- BrdU
Bromodeoxyuridine
- Cdk
cyclin-dependent kinases
- CMV
cytomegalovirus
- GPR30
G protein-coupled receptor 30
- MPA
medroxyprogesteroneacetate
- tTa
tetracycline-regulated transactivator
