[CANCER RESEARCH 52, 6539-6546, December 1, 1992] Estrogenic Potential of Progestins in Oral Contraceptives to Stimulate Human Cell Proliferation 1

Meei-Huey Jeng, Christopher J. Parker, and V. Craig Jordan 2 Department of Human Oncology, University of Wisconsin Comprehensive Cancer Center, Madison, Wisconsin 53792

ABSTRACT conflicting evidence regarding the increased risk of breast can- cer in women taking PC (4-12). Also, the inconsistency of Most oral contraceptives (OC) contain a progestin in combination laboratory evidence has made the interpretation of the relation- with an , and the progestin component in OC includes one of the following 19-nortestosterone derivatives: norethynodrel; norethindrone; ship between the use of PC to breast cancer risk more difficult or norgestrel (ievonorgestrel). It is well known that promote (13-16). the growth of breast cancer. However, progestins have recently also It is well known that estrogens promote the growth of breast been implicated in the development of breast cancer. We have compared cancer (17). However, progestins have also been implicated and contrasted the ability of synthetic progestins to stimulate the pro- in the development of breast cancer (4, 5, 18, 19), but the liferation of cultured human breast cancer cells and examined their underlying mechanism is not known. Most of the progestin possible mechanism of action. components of PC include the 19-nortestosterone derivatives We found that some progestins used in OC were able to stimulate the norethynodrel, norethindrone, and norgestrel () growth of -positive (ER § MCF-7 and T47DA18 hu- (1, 20-22) (Fig. 1), all of which have been shown to bind to PR man breast cancer cells but not ER- MDA-MB-231, BT-20, and with high affinity (23-25) and have potent progestational ac- T47DC4 human breast cancer cells. However, two other progestins, MPA and R5020, which are not used in OC, were either not able to tivity. However, we have recently demonstrated that one of the stimulate or only slightly stimulated growth. The potency of norethyno- progestin components used in PC, norethindrone, can stimu- drel [median effective dose (ECso) = 4 x 10 -s M] and norethindrone late breast cancer cell growth and decrease transforming growth (ECso = 3 X 10 -s M) was greater than norgestrel (ECso = 2 X 10 -7 M) factor/32 and/~3 mRNA levels (26), actions generally ascribed in MCF-7 cells. E2 (ECso = 8 X 10 -13 M) was an even more potent to estrogen action. It is therefore important to determine stimulator of growth. More importantly, the progestin-induced growth whether the progestin components of PC in general can initiate stimulation was blocked by the antiestrogens 4-hydroxytamoxifen and cell replication. ICI 164,384 but not the antiprogestin 17~-hydroxy-ll~-(4-dimethy- The action of hormones is mediated by the steroid's laminophenyl)-17a-(1-propynyl)-estra-4,9-dien-3-one (RU486). To de- cognate receptors located in the nucleus and the hormone re- termine whether the proliferative action of progestins was mediated sponse elements in the target genes. Antiestrogens (17, 27-29) through the ER, cells were transfected with a chloramphenicol acetyl- transferase reporter gene containing an estrogen response element de- and antiprogestin (30-32) are proving to be powerful tools in rived from vitellogenin 2A gene. The progestins which stimulated the the understanding of gene regulation mediated by steroid hor- growth of breast cancer cells also increased chloramphenicol acetyl- mones because of their ability to specifically block the actions of transferase activity. The induction of chloramphenicol acetyltransferase the relevant steroid hormones. activity was blocked by the addition of the antiestrogens 4-hydroxyta- In this study, we investigated whether the commonly used moxifen and ICI 164,384 but not the antiprogestin RU486. progestins in PC could stimulate the growth of human breast This study provides direct evidence that the 19-nortestosterone de- cancer cells and determined whether these progestins could act rivatives in OC have estrogenic properties and suggests that activation through the ER. We found that not only norethindrone but also of ER, but not receptor, is the growth-stimulatory mech- the other 19-nortestosterone derivatives, norethynodrel and anism for these synthetic progestins. Our results may help to explain norgestrel, can stimulate the growth of ER + but not ER- hu- the conflicting evidence linking OC and breast cancer risk. A rigorous evaluation of the "total" estrogenic potential of OC might produce a man breast cancer cells. More importantly, we demonstrated better correlation with breast cancer risk. that these progestins induced the activity of a CAT reporter gene containing an ERE, and this induction could be blocked by the addition of antiestrogens but not antiprogestin. INTRODUCTION PC 3 have been used by women to control ovulation, and MATERIALS AND METHODS most PC contain an estrogen in combination with a proges- tin (1). There is a concern that PC may increase the risk of Cell Culture. The MCF-7 cell line (33) was originally obtained from breast cancer because most of the known risk factors for breast Dr. Dean Edwards (Department of Medicine, Health Sciences Center cancer are related to steroid hormones (2, 3). However, there is at San Antonio, San Antonio, TX), who obtained the stock from the Michigan Cancer Foundation. The MCF-7 cells were between passages 470 and 510. MCF-7 cells were grown in minimum essential medium Received 5/18/92; accepted 9/17/92. containing 5% (v/v) calf serum supplemented with 0.29 mg/ml The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accord- L-glutamine, 100 units/ml penicillin plus 100 ~tg/ml streptomycin, 6 ance with 18 U.S.C. Section 1734 solely to indicate this fact. ng/ml bovine insulin (Sigma Chemical Co., St. Louis, MO), 0.35 g Supported by Grant RO1 CA32713. NaHCO3/liter, and 25 mu N-hydroxyethylpiperazine-N'-2-ethane sul- 2 To whom requests for reprints should be addressed, at Department of Human fonic acid. Oncology, University of Wisconsin Comprehensive Cancer Center, 600 Highland Avenue, Madison, WI 53792. The T47DA18 and T47DC4 cells (34) were cloned from T47D cells 3 The abbreviations used are: PC, oral contraceptives; PR, progesterone recep- originally obtained at passage 81 from the American Type Culture tor, 4-OHT, 4-hydroxytamoxifen; RU486, (17~-hydroxy-1 l~-(4-di- Collection (Rockville, MD). The T47DA18 and T47DC4 cells were methylaminophenyl)-17a-(1-propynyl)-estra-4,9-dien-3-one); ER, estrogen recep- between passages 200 and 220. T47DA18 cells were cultured in RPMI tor; CAT, chloramphenicol acetyl transferase; ERE, estrogen response element; E2, 17/~-; MPA, acetate; PRE/GRE, progesterone re- 1640 supplemented with 10% (v/v) fetal bovine serum (heat inacti- sponse element/ response element; ECso, median effective concen- vated), 0.29 mg/ml L-glutamine, 100 units/ml penicillin plus 100/~g/ml tration. streptomycin, 6 ng/ml bovine insulin, and 0.35 g NaHCOa/liter. 6539 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGENICITY OF PROGESTINS OH (SLM Instruments, Inc., American Instrument Company, Urbana, IL). --C E CH All points for each DNA measurement represent a mean _+ SE of three 0~CiCH sampled wells. Transient Transfection and CAT Assay. Cells were plated into O 10-cm-diameter dishes at a density of 8 x 105 cells in phenol red-free medium containing stripped serum. Cells were then deprived of estro- Norethynodrel Norgestrel gen for 6-7 days before transfection, and medium was changed every other day. On the day of transfection, cells were fed with 9 ml complete medium 1 to 3 h prior to the addition of the precipitate. Transfections OH were carried out by the phosphate method, with 10 #g of CAT reporter plasmid together with 5 #g of pCMV~ plasmid/plate (36). O"~CBCH ERE-CAT (pvitEREl 5-TK-CAT) (37), which consists of a palindromic ERE derived from the vitellogenin 2A gene and the thymidine kinase promoter derived from herpes simplex virus; pvitERE15 mt-TK-CAT, Norethindrone which consists of an mutant ERE and thymidine kinase promoter; pGRE15T (pGRE15T-TK-CAT) (37), which consists of consensus GRE and thymidine kinase promoter; and the parental plasmid pBL. O,'C/ "C CAT.2, which consists of the thymidine kinase promoter linked to the ~- OCOCH3 gene for CAT, were used as reporter plasmids, pCMV/~ plasmid, which 0~cH3 consists of the Escherichia coli ~-galactosidase gene driven by the hu- man cytomegalovirus immediate early gene promoter and enhancer, was used as a reference plasmid for internal control. Six h after trans- 0 fection, medium was removed, and cells were incubated for 3 min with 10% glycerol in complete medium and then refed with medium con- MPA R5020 taining compounds for an additional 48 h. Cells were rinsed twice with Fig. l. Structures of synthetic progestins. Norethynodrel,norgestrel, and nore- phosphate-buffered saline (50 mM sodium phosphate, 100 mM NaCI, thindrone are commonly used as progestins in OC. pH 7.35) followed by the addition of 1 ml 40 mM Tris-HCl (pH 7.5), 1 mM EDTA, 150 mM NaCI and incubated on ice for 5 min. Cells were T47DC4 cells were cultured in dextran-coated charcoal-stripped fetal then harvested by scraping with a policeman, and cells were pelleted and bovine serum and phenol red-free RPMI 1640. suspended in 100 ~tl ice-cold 0.25 M Tris-HCI (pH 7.5). Cell extracts In addition, MDA-MB-231 cells and BT-20 cells (hormone-indepen- were prepared by 3 freeze-thaw cycles. Five to ten ~tl of cell extracts were dent ER- breast cancer cell lines), obtained from the American Type used to determine the content of/3-galactosidase activity (38) after Culture Collection, were used in parallel experiments with the MCF-7 transfection, and cell extracts were heated to 65~ for 10 min before the cells. For MDA-MB-231 cells, the culture condition was the same as CAT assay. CAT reaction (36) was performed with cell extracts con- previously described for MCF-7 cells. For BT-20 cells, the culture taining equal amounts (2-4 units) of/3-galactosidase activity (usually condition was the same as previously described for MCF-7 cells except 2-5 #1 of cell extracts) in the presence of 0.05 ~Ci [14C]chloramphenicol for the replacement of 5% calf serum with 10% fetal bovine serum (heat (55 mCi/mmol; Amersham), 0.53 mM acetyl-CoA, and 0.22 M Tris-HCl inactivated). MDA-MB-231 cells were between passages 30 and 40. (pH 7.5) at 37~ for 25-30 min. Thin-layer chromatography was per- BT-20 cells were between passages 180 and 190. formed to separate the acetylated chloramphenicol from the unacety- Medium components were obtained from Gibeo (Grand Island, NY) lated chloramphenicol. Autoradiography was accomplished by expos- unless otherwise stated. Cells were routinely passed by an initial wash ing the thin-layer chromatography plates to Kodak X-Omat AR film with calcium- and -free Hanks' balanced salt solution fol- at -70~ lowed by a 5-min incubation with I mM EDTA, also in Hanks' solution. Enzyme Immunoassay. Cells were plated in Cell stocks were kept in 162-cm 2 flasks (Costar Corp., Cambridge, MA) 15-cm-diameter dishes and deprived of estrogen. Compounds were in a humidified atmosphere of 95% air/5% CO2 at 37~ The cells were added for 6 days, and cells were collected by scraping. Cytosols were negative for the presence of Mycoplasma. prepared in buffer containing 1 mM monothioglycerol, 10 mM Tris, 1.5 Hormone Treatment. E2, norethynodrel, norgestrel, norethindrone, mM EDTA, 5 mM Na2MoO4, and 0.4 M KCI and collected following and MPA were purchased from Sigma Chemical Co. R5020 was pur- 100,000 x g centrifugation for 45 rain. The level of PR in cytosol was chased from New England Nuclear Products (Boston, MA). 4-OHT measured by using progesterone receptor enzyme immunoassay kits and ICI 164,384 were obtained from ICI Pharmaceuticals (Maccles- obtained from Abbott Laboratories (N. Chicago, IL). field, England). RU486 was a gift from Roussel Uclaf (Romdinville, France). All compounds were prepared in a concentrated form in 100% ethanol and diluted 1:1000 (v/v) into cell culture medium. Final ethanol RESULTS concentrations in the media were 0.1 to 0.2%. Growth-Response Assay. Cells (except T47DC4) were plated in me- Growth Stimulation of ER + Human Breast Cancer Cells by dia containing phenol red and whole serum at a density of 14,000 E2, Norethynodrel, Norgestrel, and Norethindrone and Block- cells/well into 24-well plates (Costar Corp.) in triplicate. The next day, ade by Antiestrogens but not Antiprogestin. Previously, we media were changed to phenol red-free media containing dextran- showed that norethindrone, a 19-nortestosterone derivative coated charcoal-stripped serum, and media were changed again the next which is a commonly used progestin in OC, stimulated the day. Cells were deprived of estrogen for a total of 4 days before the growth of MCF-7 cells (26). We have now investigated whether addition of compounds. Media that contained compounds were then other 19-nortestosterone derivatives, which are also used in added for 5 days and were changed every other day. Cells were then OC, can stimulate human breast cancer cells to grow. We chose harvested for DNA assays. Briefly, cell monolayers were treated with 1 norethynodrel and norgestrel (levonorgestrel) because these ml/well hypotonic 0.1 x calcium- and magnesium-free Hanks' balanced two progestins are also widely included in OC (39). We found salt solution and sonicated in the well for 20 s with a Kontes ultrasonic cell disrupter (Kontes, Vineland, N J). Samples (50-100 ~tl) were taken that, at a concentration of 10 -6 M, not only norethindrone for DNA determination. DNA was measured fluorometrically by incu- (5.6-fold over control) but also norethynodrel (7.2-fold over bating samples with Hoechst dye 33258 (Calbiochem-Behring Corp., control) and norgestrel (4.8-fold over control) could stimulate La Jolla, CA) according to the method described by LaBarca and the growth of MCF-7 cells in culture (Fig. 2,4). Fig. 2,4 is Paigen (35) and analyzed on an SLM-Aminco Fluoro-Colorimeter III representative of data from 7 independent experiments. The 6540 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGENICITY OF PROGESTINS

B Fig. 2. A, effects of various progestins and 25. 14 E2 on the growth of human breast cancer ER+ MCF-7 cells. B, effects of various progestins 20- 12 and E 2 on the growth of ER § T47DA18 cells. ~ ~o C, effects of 4-OHT on the E2- and norethyno- tn drel-induced cell proliferation in MCF-7 cells. ~ 15- ~ a D, effects of 4-OHT and RU486 on the E2- i and progestin-induced cell proliferation in ~ 10 MCF-7 cells. Cells were seeded at a density of 14,000 cells/well into 24-well plates in tripli-

potency of norethynodrel (ECso = 4 x 10 -s M) or norethin- drone (ECso = 3 x 10 -8 M) was greater than norgestrel (ECso = 2 x 10 -7 M). However, E2 (EC5o = 8 x 10 -]3 M) was even more Control potent as a growth stimulator than the progestins. E 2 at a concentration of 10 -1~ M stimulated cell growth (9.4-fold over Norethynodrel control). R5020, a progestin lacking the 19-C group like the 19-nortestosterone derivatives, stimulated minimal cell growth Norethindrone at 10 -6 M as reported by other investigators (40). The progestin MPA, which is not a 19-nortestosterone derivative, did not stimulate the growth of MCF-7 cells. Similar results were ob- Norgestrel tained using other ER + human breast cancer T47DA18 (Fig. 2B) and ZR-75-1 cells (25). R5020 Antiestrogens (17, 27-29) block E2-induced growth in MCF-7 cells. The addition of the antiestrogens 4-OHT (Fig. 2, 0 10 2'0 30 C and D) or ICI 164,384 (data not shown) also blocked the DNA (pg/well, +/- SEM) growth induced by the three progestins. Competitive reversal of Fig. 3. Effects of the combination of E2 and progestins on the growth of the blockade by the antiestrogens was observed. Interestingly, MCF-7 cells. Cells were plated and treated with E2 in combination with various the antiprogestin RU486 did not influence the growth stimula- progestins as mentioned in Fig. 2. All points for each DNA measurement repre- tion induced by these progestins at a concentration of 10 -7 M sent a mean of three sampled wells. Q, compounds used were control (0.1% ethanol), norethynodrel (10 -6 M), norethindrone (10 -6 M), norgestrel (10 -6 M), (Fig. 2D) or 10 -6 M (data not shown). Furthermore, it is im- and R5020 (10 -6 M). II, E2 (10 -10 M) was used in addition to the above com- portant to note that the addition of progestins did not com- pounds. pletely block the E2-induced cell proliferation to the control level (Fig. 3). The ability of the 19-nortestosterone derivatives and MPA to drel, norgestrel, and norethindrone, we believed that the growth regulate the growth of ER- human breast cancer MDA-MB- stimulation induced by these progestins was mediated by ER. 231, BT-20, and T47DC4 cells was also examined. Neither To test this hypothesis, the pvitERE-TK-CAT (ERE-CAT) (37) these progestins nor E 2 altered the growth of ER- MDA-MB- was used as the reporter gene to examine whether these 231 (Fig. 4), BT-20, or T47DC4 (data not shown) human breast progestins were estrogenic in inducing the expression of the cancer cells. CAT gene downstream from an ERE. ERE-CAT consists of Induction of pvitERE-TK-CAT Activities by E 2 and Proges- an ERE derived from the vitellogenin 2A gene and the thymi- tin Components in OC and Blockade by Antiestrogens but not dine kinase promoter derived from herpes simplex virus linked Antiprogestin. Since the antiestrogens 4-OHT and ICI to the gene for CAT. We observed that norethynodrel induced 164,384 blocked the growth stimulation induced by norethyno- the ERE-CAT activity in a concentration-dependent fashion 6541 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGENICITY OF PROGESTINS

12 5B). When the concentration of E2 reached l0 -~ ~, a slight induction of ERE-CAT activity was observed, and this induc- 10- tion reached a maximum level at a concentration of 10 -~ (7.4-fold over control). Antiestrogens 4-OHT and ICI 164,384 O')"=' 8- blocked both norethynodrel- and E2-induced ERE-CAT activ- ities (Fig. 5, ,4 and B). However, when the concentrations of 6" norethynodrel and E 2 were increased to 3 x 10 -6 and 10 -7 M O) 4 respectively, norethynodrel and E2 reversed the blockade by < antiestrogens on ERE-CAT activity. z r~ 2- Progestins that stimulated the growth of human breast can- cer cells were able to induce the activity of ERE-CAT (Fig. 6). Fig. 6 is representative of data from 5 independent experiments. The extent of induction of ERE-CAT activity by these pro- - Log conc. M gestins (10 -6 M) was identical to that observed for E2 (10 -~~ M) (Fig. 6). We found that antiestrogens 4-OHT and ICI 164,384 Fig. 4. Effects of various progestins and E2 on the growth of ER(-) MDA- MB-231 cells. Cells were seeded at a density of 14,000 cells/well into 24-weU could block the progestin-induced ERE-CAT activities (Figs. 6 plates in triplicate, deprived of estrogens for 4 days, and then treated with and 7). Furthermore, the antiprogestin RU486, which blocks compounds for 5 days. Media were changed every 2 days. D, control (0.1% ethanol); O, E~; II, norethynodrel; ~, norethindrone; A, norgestrel; &, R5020; e, progestin binding to the PR and forms nonactive complexes MPA. with the hormone response element (31, 32), did not block the progestin-induced growth stimulation (Fig. 2D) and did not prevent the expression of ERE-CAT (Fig. 7). Therefore, these (ECho = 3.2 x l0 -9 M) (Fig. 5,4). Fig. 5,4 is representative of progestins behave more like estrogens with respect to cellular data from 5 independent experiments. Norethynodrel at con- proliferation. centrations of 10 -~ a to 10- ~ M induced no activity. When the In contrast to the ER + MCF-7 cells, in which norethynodrel, concentration reached 10 -9 M some induction was observed, norgestrel, norethindrone, and E2 induced the expression of which demonstrated a maximum stimulation at 10 -6 M (7.4- ERE-CAT, in ER- MDA-MB-231, BT-20, and T47DC4 cells fold over control). E2 also regulated the ERE-CAT activity in a we observed no induction of ERE-CAT activity by any of these concentration-dependent fashion (ECho = 3.2 x 10 -~ M) (Fig. compounds (Fig. 8).

A

ICI164,384106M -!- 4- 4- 4- --I--I-" "-I-- -I.-

4-OHT 10-7M 4- --I- -I- -'1-- -'!- -I'- --t- I- --

NORETHYNODREL-- ~ o ='o ~o o ~~ " =. " ~o ~o ~ ~ = ,,-

9= 9 ,~ 9 9 ~ i~ 9149 ~84~!~ ~ ~ ~:i~ !i~::~/%~ Fig. 5. Dose response of norethynodrel (,4) and E2 (B) on the induction of pvitERE-TK- CAT activity in the presence and absence of antiestrogens 4-OHT and ICI 164,384 in MCF-7 cells. Cells were plated into 10-cm- diameter dishes at a density of 8 • 105 cells in phenol red-free medium containing stripped calf serum. Cells were then deprived of estro- gen for 6-7 days before transfection, and me- dia were changed every other day. Transfec- tions were done by the calcium phosphate method with 10 ~g of pvitERE-TK-CAT plas- B mid together with 5 ug of pCMV/~ plasmid/ plate. After transfection, cells were treated ICI164.384 10 .6 M + 4-+ 4-4-++ -- + with compounds for an additional 48 h, and /~-galactosidase activity and CAT activity were 4-OHT 10 .7 M 4- 4- ++ 4- 4- + +-- then determined. CAT reactions were per- formed with cell extracts containing equal E2 amounts of/~-galactosidase activity. ~ 0 0 ~ ~

~i~ ~84149~...... ~ 6542 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGENICITY OF PROGESTINS

ICI164,384 106M + ++++-+- + that the progestins in OC are estrogenic and that antiestrogens are able to block their actions. 4-OHT 10"7 M +++4-+ +-+--

9 o DISCUSSION _ 7 ~ ~ We have investigated the effects of progestins on the prolif-

7" eration of human breast cancer cells and determined the possi- o~ ble mechanism underlying their action. We conclude that the 19-nortestosterone derivatives norethynodrel, norgestrel, and norethindrone are estrogenic in stimulating the growth of ER + human breast cancer cells and that this action is mediated through ER. Plasma progestin concentrations in women taking OC is in the range of 10 -7 to 10 -$ M (39, 41, 42), consistent with the Fig. 6. Effects of various progestins on the induction of pvitERE-TK-CAT activity in the presence and absence of antiestrogens 4-OHT and ICI 164,384 in levels we used in vitro. Therefore, our finding that the growth of MCF-7 cells. Cells were seeded, transfected, and assayed for CAT activity as mammary epithelia in culture can be stimulated by the described in Fig. 5. Norethynodrel (10-6 M), norgestrel (10-6 M), norethindrone progestins in OC supports the hypothesis that these compounds (10 -6 M), R5020 (10 -6 M), MPA (10 -6 M), and E2 (10-1~ M) were used. Similar results were observed in an additional 4 independent experiments. might do the same in vivo (16). In addition, our study provides further support for the possibility that women taking OC for an

4-OHT 10-7 M 4-4- 4- "P 4- extended period of time will maintain this risk of developing breast cancer. Our results (Fig. 3) demonstrate that a modest RU486 107M 4-- 4- + Jr- -'1- protective effect at best could be expected from these progestins

NORETHYNODREL o =. ~ e o e ~. ~. o e ~, in the prevention of breast cancer. Unlike antiestrogens, the o o o o o o o o ~ o o o o -- progestins do not have a profound effect on estrogen-stimulated growth. The dominant action of 19-nortestosterone derivatives is as a growth promoter. This position is supported by the epidemiological data, which show that women are not protected from breast cancer by OC use. Recently, Battersby et al. (43) demonstrated that the ER was decreased and PR was increased by the progestins in OC in the normal breast. In addition, Williams et al. (44) showed that OC use increased cell proliferation in normal breast and suggested Fig. 7. Comparison of the effects of antiestrogen 4-OHT and antiprogestin RU486 on the norethynodrel-inducedERE-CAT activity in MCF-7 cells. Cells that OC use is a risk factor for breast cancer. This is consistant were seeded, transfected, and assayed for CAT activities as described in Fig. 5. with the observation we obtained in cultured human breast cancer cells. The ability of the progestin components in OC to Effects of E2 and Progestins on the TK-CAT (pBL.CAT.2), stimulate the growth of human breast cancer cells and to induce pvitEREI5-TK-CAT, pvitEREl5 mt-TK-CAT, and pGRE15T- PR synthesis in a manner that was blocked by antiestrogens but TK-CAT (pGRE15T) Activities in MCF-7 Cells. In order to not antiprogestin led us to speculate that these progestins were confirm that norethynodrel, norgestrel, and norethindrone estrogenic in stimulating the growth of human breast cancer could bind to ER and induce ERE-CAT activity, we also used a cells and that this action was mediated through ER. Further- mutant ERE construct (pvitERE15 mt-TK-CAT) (Fig. 9,4), more, it is possible that 19-nortestosterone derivatives could act which cannot be induced by the ER- complex (37). We found that these progestins could not induce the CAT activities MCF-7 MDA-MB-231 BT-20 T47DC4 of parental plasmid pBL.CAT.2 and the pvitERE15 mt-TK- I I[ 11 I[ I CAT containing the mutant ERE (Fig. 9B). Klock et al. (37) have shown that ERE and PRE/GRE are closely related but distinct. We showed that these progestins .u ...J also induced the CAT activity of pGRE15T (Fig. 9,4), and the tr fl: rr addition of RU486 blocked this progestin-induced CAT activity 8,,.,o 8~o ~o o,,.=~176 ~o z~O o~ z r (Fig. 9C). ~_ ~- e ~ ~- Regulation of PR Level by Progestins. To determine whether the progestins that were able to stimulate cellular proliferation o ~z z z~ ~z z z z ~ and induce ERE-CAT activity could regulate the synthesis of an endogenous estrogen target gene, the change in PR level was examined after treatment of MCF-7 cells with the progestins and E2. The progestins that were estrogenic (i.e., stimulate cellular proliferation and induce ERE-CAT activity) were also able to induce endogenous PR synthesis (Table 1). Therefore, these progestins can regulate gene expression in a manner sim- ilar to that of E2. The progestins MPA or R5020, which were Fig. 8. Effects of various progestins on the induction of pvitERE-TK-CAT not distinctly estrogenic, did not induce PR synthesis. Interest- activity in ER- MDA-MB-231, BT-20, and T47DC4 cells. Cells were seeded, transfected, and assayed for CAT activity as described in Fig. 5. E2 (10 -1~ M), ingly, the antiestrogen 4-OHT was able to block the norethin- norethynodrel (10 -6 M), norgestrel (10 -6 M), norethindrone (10-6 U), and MPA drone-induced PR synthesis. This agrees with our hypothesis (l 0-6 u) were used. 6543

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGEN|CITY OF PROGESTINS

A pvit.ERE15-TK-CAT 5, ACA,G +CACA T AC ' "IFCCAGTGTCACT(;GAEG ATC

pvil-ERE15mt-'rK-cKf S' CTAGtI,C,~.TCAC...~C~TG~T_([t 3' 1[TC~_AGTGTCACTAGA~GATC p GRE15T-TK-CAT 5' CTAGA~,GAACACAGTGT'FCTIG3' TC]~GTGTCAC.~G~GATC

Xba I !

pBL.CAT.2 pvitE R E 15-TK-CAT pvitE R E 15mt-T K-CAT II II I _.I _J I..U ._1 B UJ 111 W I..IJ W C n- z fr z n-" z r7 .j 0 r7 j 0 o ~0 Ow n- O w 0 wn- Z ~-- sZ Z Z3 Z rf" D .--I -- -.I Z ..j Z RU486 10-6 M -I- "-I- 4- -{- -4- -I-- o o nr I-- 13: I-- NORETHYNODREL-- o o 'o 6 4= o ,~ 'o o &,-- z n- n- n- ,~ ~ z rc n" fr ,,~ z fr rr fr ~ ,~ o ooo , oo oo o~ ooo - o u? z z z n" u.P'z z z rr w~ z z z rr ~

coo e 9 oooooooQ. + o o OOO OOO

Fig. 9. Effects of various progestins and E2 on the induction of pBL.CAT.2, pvitERE 15-TK-CAT, pvitERE15 mt-TK-CAT, and pGREI 5T-TK-CAT activities..4, plasmid constructs of various reporter genes. B, effects of various progestins and E2 on the induction of pBL.CAT.2, pvitERE 15-TK-CAT, and pvitERE 15 mt-TK-CAT activities. Cells were seeded, transfected, and assayed for CAT activity as described in Fig. 5. The concentrations of compounds used were: E2, 10-1~M; norethynodrel, 10-6 M; norgestrel, 10-6 M; norethindrone, 10 -6 M; R5020, 10-6 M; and MPA, 10-6 M. C, effects of norethynodrel on the induction of pGRE15T-TK-CAT activity. MCF-7 cells were seeded, and 10-~~m E2 was added for 4 days before the transfection. Cells were then transfected, and norethynodrel at various concentrations was added for an additional 48 h, CAT was assayed as described in "Materials and Methods." through ER (45), since it was suggested that the estrogenic ceptors, which interact with the ERE sequence located in target action of is mediated through ER in human breast genes (52, 53). Meyer et al. (54) showed that ER cannot activate cancer MCF-7 cells (46). Several reports have described the transcription controlled by a PRE/GRE in vivo or bind to a estrogenic actions of 19-nortestosterone derivatives (25, 46- PRE/GRE in vitro, nor can PR bind to a synthetic palindromic 48). However, they did not have sufficient evidence to demon- ERE. Therefore, only ER, and not PR, can bind to a synthetic strate that ER, but not PR, is involved. Our approach using palindromic ERE. We favor the hypothesis that 19-nortest- hormone response element-CAT constructs proves the mecha- osterone derivatives that are progestins bind directly to ER nism is via ER. and then interact with the ERE in the target gene because (a) It is well known that steroid regulation of gene transcription only ER and not PR interacts with the vitERE (54) used in our in target cells is mediated by the steroid's cognate specific high- study; (b) these progestins induce the activities of only ERE- affinity intracellular receptor proteins (49-51). These receptors CAT but not the mutant ERE-CAT nor the parental plasmid; act as transcription factors that can regulate gene expression (c) other progestins do not induce ERE-CAT activities; (d) we positively or negatively by interacting with specific DNA se- do not observe any induction of the ERE-CAT activity in ER- quences (hormone response elements) (49-51). Estrogen regu- cells; (e) antiestrogens but not antiprogestin block the proges- lation of target gene transcription is mediated by estrogen re- tin-induced cell proliferation and ERE-CAT activities; and (f) 6544 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGENICITY OF PROGESTINS

Table 1 Effects of progestins and the antiestrogen 4-OHT on PR synthesis the future produce a decrease in breast cancer risk by reducing in breast cancer cells in culture MCF-7 cells were deprived of estrogen for 6 days and various compounds the "total" estrogenicity of the product. were added for 6 days. Media were changed every other day. PR were extracted and assayed with Abbott PgR-EIA monoclonal kit according to the manufactur- er's protocol. The experiments were carried out in duplicate wells. Data are the ACKNOWLEDGMENTS average of two determinants. Norethynodrel, norgestrel, and norethindrone in- creased the PR level. Antiestrogen 4-OHT blocked the level increase in PR. We would like to thank Dr. J. Gorski, Dr. F. E. Murdoch, and C. MPA and R5020 did not cause increases in the PR level. 4-OHT was estrogenic Kao for their helpful discussions. We are especially grateful to Dr. K. B. and caused an increase in PR. Horwitz for her suggestions with the PRE/GRE study, Dr. M. Fritsch PR levels for his valuable suggestions during the preparation of the manuscript, Treatments (fmol/mg cytosol protein) and S. M. Langan-Fahey for performing high-performance liquid chro- Control 4 __. 1 matography studies on the purity of the progestins. We also thank Dr. E2 (10 -11 M) 128 _+ 11 G. Schiitz for the generous gift of ERE and PRE/GRE CAT expression E2 (10 -l~ M) 323 _+ 55 Norethynodrei (10 -6 M) 92 + 5 plasmids, Dr. G. MacGregor for the plasmid pCMV~, Dr. A. Wakeling Norgestrel (10 -6 M) 26 _+ 3 for ICI 164,384, Roussel Uclaf for RU486, and Lyndell Rubin for Norethindrone (10 -6 M) 83 + 7 typing the manuscript. Norethindrone (10 -6 M) + 4-OHT (10 *-7 M) 14 + 1 4-OHT (10 -7 M) 18 + 1 MPA (10 -6 M) 4 _+ 0 REFERENCES R5020 (10 -6 M) 4 + 0 1. Piper, J. M., and Kennedy, D. L. Oral contraceptives in the United States: trends in content and potency. Int. J. Epidemiol., 1C 215-221, 1987. antiprogestin is able to block the progestin-induced GRE-CAT 2. Van Leeuwen, F. E., and Rookus, M. A. The role of exogenous hormones in the epidemiology of breast, ovarian and endometrial cancer. Eur. J. Cancer activity but not the progestin-induced growth and ERE-CAT Clin. Oncol., 25: 1961-1972, 1989. activity. 3. Prentice, R. L., and Thomas, D, B. On the epidemiology of oral contracep- tives and disease. Adv. Cancer Res., 49: 285-401, 1987. We also ruled out the possibility that these progestins were 4. Pike, M. C., Henderson, B. E., Krailo, M. D., Duke, A., and Roy, S. Breast converted into aromatized estrogens for their actions, since the cancer in young women and use of oral contraceptives: possible modifying aromatase inhibitors 4-hydroxyandrostenedione and aminoglu- effect of formulation and age at use. Lancet, 2: 926-929, 1983. 5. Bergkvist, L., Adami, H-O., Persson, I., Hoover, R., and Schairer, C. The risk tethimide did not block the growth stimulation induced by these of breast cancer after estrogen and estrogen-progestin replacement. N. Engl. progestins (data not shown). Nevertheless, it is still possible J. Meal., 321: 293-297, 1989. that these progestins were converted into estrogenic metabo- 6. Gay, J. W., and Cardwell, M. S. Oral contraceptives and breast cancer. Missouri Med., 87: 763-767, 1990. lites, as suggested by other investigators (55-58). This does not 7. Buehring, G. C. Oral contraceptives and breast cancer: What has 20 years of invalidate the conclusion, however, that the 19-nortestosterone research shown? Biomed. Pharmacother., 42: 525-530, 1988. 8. Olsson, H., M611er, T. R., and Ranstam, J. Early oral contraceptive use and derivatives are indeed estrogenic. Further studies will be neces- breast cancer among premenopausal women: final report from a study in sary to confirm this possibility. In addition, we were able to southern Sweden. J. Natl. Cancer Inst., 81: 1000-1004, 1989. show that the action of norethindrone was not due to the pres- 9. Meirik, O., Farley, T. M. M., Lurid, E., Adami, H-O., Christoffersen, T., and Bergsj6, P. Breast cancer and oral contraceptives: patterns of risk among ence of an estrogenic contaminant such as . When the parous and Nulliparous women-further analysis of the Swedish-Norwegian high-performance liquid chromatography-purified norethin- material. Contraception, 39: 471-475, 1989. drone peak was collected and used for both growth and ERE- 10. UK National Case-Control Study Group. Oral contraceptive use and breast cancer risk in young women. Lancet, 1: 973-982, 1989. CAT assays, we observed the same results as reported here (data 11. Romieu, I., Willett, W. C., Colditz, G. A., Stampfer, M. J., Rosner, B., not shown). Therefore, this study suggests that 19-nortestoster- Hennekens, C. H., and Speizer, F. E. Prospective study of oral contraceptive use and risk of breast cancer in women. J. Natl. Cancer Inst., 81:1313-1321, one derivatives in OC can activate ER to stimulate the growth 1989. of ER + human breast cancer cells and provides further support 12. Olson, H., Borg, A., Ferno, M., Ranstam, J., and Sigurdsson, H. Her-2/neu for the dual effects of the 19-nortestosterone derivatives used in and INT2 proto-oncogene amplification in malignant breast tumors in rela- tion to reproductive factors and exposure to exogenous hormones. J. Natl. OC. Cancer Inst., 83: 1483-1487, 1991. The observation that R5020 was able to induce ERE-CAT 13. Chalbos, D., and Rochefort, H. Dual effects of the progestin R5020 on activity only at high concentrations is nevertheless intriguing. proteins released by the T47D human breast cancer cells. J. Biol. Chem., 259: 1231-1238, 1984. R5020 may exhibit some weak estrogenic activity at high con- 14. Horwitz, K. B., and Freidenberg, G. R. Growth inhibition and increase of centrations. This is consistent with the observation by Hissom insulin receptors in antiestrogen-resistant T47Dco human breast cancer cells by progestins: implications for endocrine therapies. Cancer Res., 45: 167- and Moore (40) that R5020 stimulated proliferation of T47D 173, 1985. human breast cancer cells. It is important to note that R5020 is 15. Braunsberg, H., Coldham, N. G., and Wong, W. Hormonal therapies for also a steroid lacking the 19-C group like 19-nortestosterone breast cancer: Can stimulate growth? Cancer Lett., 30: 213- 218, 1986. derivatives. Therefore, it is reasonable to believe that R5020 16. Longman, S. M., and Buehring, G. C. Oral contraceptives and breast cancer: may have weak estrogenic actions at high concentration. in vitro effect of contraceptive on human mammary cell growth. Our results may help to explain the conflicting evidence link- Cancer (Phila.), 59: 281-287, 1987. 17. Jordan, V. C. (ed). Estrogen/Antiestrogen Action and Breast Cancer Ther- ing OC and breast cancer risk. A rigorous evaluation of the apy. Madison: University of Wisconsin Press, 1986. "total" estrogenicity of OC on breast cancer cells might produce 18. Jabara, A. G. Effects of progesterone on 9,10-dimethyl-l,2-benzanthracene- induced mammary tumours in Sprague Dawley rats. Br. J. Cancer, 21: 418- a better correlation with breast cancer risk. This study provides 428, 1967. direct evidence that 19-nortestosterone derivatives have intrin- 19. Key, T. J., and Pike, M. C. The role of oestrogens and progestogens in the sic estrogenic properties. In addition, these results should also epidemiology and prevention of breast cancer. Eur. J. Cancer. Clin. Oncol., 24: 29-43, 1988. prove to be helpful in evaluating the long-term safety of OC and 20. Pincus, G., Chang, M. C., Zarrow, M. X., Hafez, E. S. E., and Merrill, A. in understanding the underlying mechanism of the proposed Studies of the biological activity of certain 19-nor steroids in female animals. toxicities. Since estrogen is implicated in the development of Endocrinology, 59: 695-707, 1956. 21. Drill, V. A., and Riegel, B. Structural and hormonal activity of some new breast cancer, we believe that less estrogenic progestins should steroids. Recent Prog. Horm. Res., 14: 29-67, 1958. be selected for inclusion in OC. There is a decrease in ovarian 22. Edgren, R. A. Memoir--the beginning of oral contraceptives. Endocrinol- ogy, 129: 1144-1145, 1991. and endometrial carcinoma in OC users but no decrease in 23. Kloosterboer, H. J., Vonk-Noordegraaf, C. A., and Turpijn, E. W. Selectivity breast cancer incidence. Alteration in the formulation could in in progesterone and receptor binding of progestagens used in oral 6545

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. ESTROGENICITY OF PROGESTINS

contraceptives. Contraception, 38: 325-332, 1988. 41. Orme, M. L'E., Black, D. J., and Breckenridge, A. M. Clinical pharmacok- 24. Hammerstein, J. Prodrugs: Advantage or disadvantage? Am. J. Obstet. Gy- inetics of oral contraceptive steroids. Clin. Pharmacokinet., 8." 95-136, 1983. necol., 163: 2198-2203, 1990. 42. Goldzieher, J. W. Pharmacology of contraceptive steroids: a brief review. 25. Poulin, R., Baker, D., Poirier, D., and Labrie, F. Multiple actions of synthetic Am. J. Obstet. Gynecol., 160: 1260-1264, 1989. "progestins" on the growth of ZR-75-1 human breast cancer cells: an in vitro 43. Battersby, S., Robertson, B. J., Anderson, T. J., King, R. J. B., and McPher- model for the simultaneous assay of androgen, progestin, estrogen, and glu- son, K. Influence of menstrual cycle, parity and oral contraceptive use on cocorticoid agonistic and antagonistic activities of steroids. Breast Cancer steroid hormone receptors in normal breast. Br. J. Cancer, 65: 601-607, Res. Treat., 17: 197-210, 1990. 1992. 26. Jeng, M-H., and Jordan, V. C. Growth stimulation and differential regula- 44. Williams, G., Anderson, E., Howell, A., Watson, R., Coyne, J., Roberts, S. tion of transforming growth factor-~l (TGF/31), TGF~2, and TGF~3 mes- A., and Potten, C. S. Oral contraceptive (OCP) use increases proliferation senger RNA levels by norethindrone in MCF-7 human breast cancer cells. and decreases oestrogen receptor content of epithelial cells in the normal Mol. Endo., 5: 1120-1128, 1991. human breast. Int. J. Cancer, 48." 206-210, 1991. 27. Pasqualini, J. R., Sumida, C., and Giambiagi, N. Pharmacodynamic and 45. Black, L. J., and Kraay, R. J. Evaluation of two types of estrogen inhibition biological effects of anti-estrogens in different models. J. Steroid Biochem., with regard to effects on uptake and binding of [aH]estradiol in the uterus. J. 31: 613-643, 1988. Steroid Biochem., 4: 467-475, 1973. 28. Jordan, V. C. Biochemical pharmacology of antiestrogen action. Pharmacol. 46. Zava, D. T., and McGuire, W. L. Androgen action through estrogen receptor Rev., 36: 245-276, 1984. in a human breast cancer cell line. Endocrinology, 103: 624-631, 1978. 29. Jordan, V. C., and Murphy, C. S. Endocrine pharmacology of antiestrogens 47. Markiewicz, L., Hochberg, R. B., and Gurpide, E. Intrinsic estrogenicity of as antitumor agents. Endocr. Rev., I1: 578-610, 1990. 30. Baulieu, E-E. Contragestion and other clinical applications of RU 486, an some progestagenic . J. Steroid Biochem. Mol. Biol., 41: 53-58, 1992. antiprogesterone at the receptor. Science (Washington DC), 245:1351-1357, 48. Van Der Berg, B., Kalkhoven, E., Isbriicker, L., and De Laat, S. W. Effects 1989. of progestins on the proliferation of estrogen-dependent human breast cancer 31. Guiochon-Mantel, A., Loosfelt, H., Ragot, T., Bailly, A., Atger, M., Misrahi, cells under growth factor-defined conditions. J. Steroid Biochem. Mol. Biol., M., Perricaudet, M., and Milgrom, E. Receptor bound to antiprogestin form 42: 457-465, 1992. abortive complexes with hormone responsive elements. Nature (Lond.), 336: 49. Beato, M. Gene regulation by steroid hormones. Cell, 56: 335-344, 1989. 695-698, 1988. 50. Beato, M. Transcriptional control by nuclear receptors. FASEB J., 5: 2044- 32. Meyer, M-E., Pornon, A., Ji, J., Bocquel, M-T., Chambon, P., and Grone- 2051, 1991. meyer, H. Agonistic and antagonistic activities of RU486, on the functions of 51. Fuller, P. J. The steroid receptor superfamily: mechanisms of diversity. the human progesterone receptor. EMBO J., 9: 3923-3932, 1990. FASEB J., 5: 3092-3099, 1991. 33. Soule, H. D., Vazquez, J., Long, A., Albert, S., and Brennan, M. A human 52. Murdoch, F. E., and Gorski, J. The role of ligand in estrogen receptor cell line from a pleural effusion derived from a breast carcinoma. J. Natl. regulation ofgene expression. Moi. Cell. Endocrinol., 78: C103-C108, 1991. Cancer Inst., 51: 1409-1416, 1973. 53. Kumar, V., Green, S., Stack, G., Berry, M., Jin, J-R., and Chambon, P. 34. Murphy, C. S., Pink, J. J., and Jordan, V. C. Characterization of a receptor- Functional domains of the human estrogen receptor. Cell, 51: 941-951, negative, hormone-nonresponsive clone derived from T47D human breast 1987. cancer cell line kept under estrogen-free conditions. Cancer Res., 50: 7285- 54. Meyer, M-E., Gronemeyer, H., Turcotte, B., Bocquel, M-T., Tasset, D., and 7292, 1990. Chambon, P. Steroid hormone receptors compete for factors that mediate 35. LaBarca, C., and Paigen, K. A simple, rapid, and sensitive DNA assay pro- their enhancer function. Cell, 57: 433-442, 1989. cedure. Anal. Biochem., 102: 344-352, 1980. 55. Larrea, F., Vilchis, F., Chfivez, B., P6rez, A. E., Garza-Flores, J., and Perez- 36. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Smith, J. A., Palacios, G. The metabolism of 19-nor contraceptive progestins modulates Seidman, J. G., and Struhl, K. Current Protocols in Molecular Biology. New their biological activity at the neuroendocrine level. J. Steroid Biochem., 27: York: John Wiley and Sons, 1987. 657-663, 1987. 37. Klock, G., Str~hle, U., and Schiitz, G. Oestrogen and glucocorticoid respon- 56. Vilchis, F., Chfivez, B., P6rez, A. E., Garcia, G. A., Angeles, A., and P6rez- sive elements are closely related but distinct. Nature (Lond.), 329: 734-736, Palacios, G. Evidence that a non-aromatizable metabolite or 1987. induces estrogen-dependent pituitary progestin receptors. J. Steroid Bio- 38. Sambrook, J., Fritsch, E. F., and Maniatis, T. Molecular Cloning: A Labo- chem., 24: 525-531, 1986. ratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 57. Sahlberg, B-L., Landgren, B-M., and Axelson, M. Metabolic profiles of 1989. endogenous and ethynyl steroids in plasma and urine from women during 39. Kuhnz, W. of the contraceptive steroids levonorgestrel and administration of oral contraceptives. J. Steroid Biochem., 26: 609-617, after single and multiple oral administration to women. Am. J. 1987. Obstet. Gynecol., 163: 2120-2126, 1990. 58. Perel, E., Daniilescu, D., Kharlip, L., Blackstein, M. E., and Killinger, D. W. 40. Hissom, J. R., and Moore, M. R. Progestin effects on growth in the human The relationship between growth and metabolism in four breast cancer cell line T47D-possible therapeutic implications. Biochem. cell lines of human breast carcinoma cells in culture. Mol. Cell. Endocrinol., Biophys. Res. Commun., 145:706-711, 1987. 41: 197-203, 1985.

6546

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research. Estrogenic Potential of Progestins in Oral Contraceptives to Stimulate Human Breast Cancer Cell Proliferation

Meei-Huey Jeng, Christopher J. Parker and V. Craig Jordan

Cancer Res 1992;52:6539-6546.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/23/6539

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/52/23/6539. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1992 American Association for Cancer Research.