Journal of Steroid Biochemistry & Molecular Biology 80 (2002) 125130

Antiestrogenic activities of Cimicifuga racemosa extracts

Oliver Zierau a , Claudia Bodinet b , Susanne Kolba a , Marina Wulf a , Gnter Vollmer a,
a

Molecular Cell Physiology and Endocrinology, Technical University Dresden, Mommsen street 13, 01062 Dresden, Germany b Schaper and Brmmer GmbH and Co. KG, Bahnhof street 35, 38259 Salzgitter, Germany Received 18 May 2001; accepted 17 October 2001

Abstract Despite the wide use of extracts from the rhizome of black cohosh (Cimicifuga racemosa) for the treatment of menopausal complaints, surprisingly little is known on their potential estrogenic properties, e.g. on estrogen dependent gene transcription. In addition, available informations on the effects on cell proliferation are contradictory. We therefore, tested for estrogenic and antiestrogenic effects of Cimicifuga racemosa extracts on proliferation of MCF-7 cells and on gene expression using ethanolic and iso-propanolic extracts of this medical plant. Estrogenic properties of plant extracts could neither be detected in proliferation assays, nor on gene expression using an estradiol-inducible yeast assay or the estrogen-inducible MVLN cells. In contrast, in all three experimental systems Cimicifuga racemosa antagonized estradiol induced activities. Estradiol induced stimulation of proliferation was inhibited by a dosage >1 g/ml of extract concentration, gene expression was suppressed by doses of 1001000 g/ml of Cimicifuga racemosa extracts. From these results we conclude, that extracts from the rhizome of Cimicifuga racemosa contain compounds with antiestrogenic properties. 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Antiestrogenic activities; Estradiol-inducible yeast; MCF-7 cells

1. Introduction Extracts from the rhizome of black cohosh Cimicifuga racemosa (CR) are used in phytomedicine for treatment of menopausal and post-menopausal disorders (for review see [1]), but their estrogenic potencies are controversly discussed. Evidence for estrogenic activity of individual compounds of CR was proposed as consequence of an increase in cell numbers of the estrogen dependent MCF-7 cell line in response to treatment with individual compounds isolated from the extract [2]. In contrast, the growth of the estrogen receptor positive cell line T47D was inhibited by extracts from black cohosh [3]. No indication of an estrogenic activity could be found by testing extracts from CR in rats and mice in vivo [4]. The rst aim of this study was to clarify the potential effects of extracts from CR on proliferation of MCF-7 cells. In addition, almost no information is available on potential effects of extracts from CR on gene transcription. In the second part of our study, we therefore, investigated the effects of CR on estrogen dependent gene expression. Two independent transactivation systems, an estrogen-inducible yeast assay [5] and the estrogen responsive MVLN cell line [6,7] were used. In a rst series of experiments we tested for

potential estrogenic effects and in a second series antiestrogenic properties of CR extracts were evaluated. Our results provide a clear indication for antiestrogenic properties of CR extracts in both transactivation assays, as well as in the proliferation assay. 2. Material and methods 2.1. Materials 17 -Estradiol and tamoxifen (citrate salt) were purchased from Sigma (Deissenhofen, Germany). Ethanolic and isopropanolic extracts from the rhizome of CR were prepared according to standard procedures and were provided by the manufacturer (Schaper and Brmmer GmbH and Co. KG, Salzgitter-Ringelheim, Germany). 2.2. Cells Estrogen receptor-positive (ER+ ) MCF-7 cells were purchased from American type culture collection (ATCC, HTB 22). MVLN cells were obtained from Dr. M. Pons (INSERM U439, Montpellier, France). MVLN cells are a derivative of the estrogen receptor- (ER ) positive MCF-7 cell line stably transfected with a vitellogenin-A2-promotor/luciferase reporter construct [6,7].

Corresponding author. Tel.: +49-351-4631922; fax: +49-351-4631923. E-mail address: guenter.vollmer@mailbox.tu-dresden.de (G. Vollmer).

0960-0760/02/$ see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 0 7 6 0 ( 0 1 ) 0 0 1 7 8 - 9

126

O. Zierau et al. / Journal of Steroid Biochemistry & Molecular Biology 80 (2002) 125130

The yeast estrogen receptor assay was supplied by Dr. J.P. Sumpter (Brunel University, Uxbridge, UK, [5]). The yeasts for the estrogen-inducible yeast assay carry the stably transfected human estrogen receptor- and an estrogen responsive element fused to the LacZ-reporter ( -galactosidase; [5]). In the case of MVLN cells luciferase activity and in the case of the estrogen-inducible yeast expression system -galactosidase activity was measured. Both represent a direct method to measure the estrogenic activity of test compounds. 2.3. Transactivation assays Basically, yeast and MVLN cells were cultured as described elsewhere (57). To test for estrogenicity in a rst series of experiments cultures of yeasts were treated with ethanolic or iso-propanolic extracts from the rhizome of CR in a concentration dependent manner (109 to 102 -fold nal dilutions of the primary extract) were used in the test. These dilutions correspond to an approximate concentration range of the extract of 0.0011000 g/ml in the case of the iso-propanolic extract and 0.00002200 g/ml in the case of the ethanolic extract from CR. As controls we used estradiol (5 1011 to 108 M) and vehicle controls. To test for antiestrogenicity each well of cells was incubated with either 5 1010 or 109 M of estradiol in combination of increasing concentrations from extracts of CR (range 109 to 102 -fold dilutions, absolute concentrations see above). For the reason of direct comparability of different experiments values for 108 M of estradiol were set to 100% and all other values were calculated accordingly. In a second series of experiments we used MVLN cells. Using this bioassay we tested for estrogenicity in a concentration range of 104 to 103 -fold dilutions of the primary ethanolic or iso-propanolic extracts, corresponding to concentrations of 220 (ethanolic extract) and 10100 g/ml (iso-propanolic extract). As references, we used vehicle treatment and 108 M estradiol as negative and positive controls. To test for antiestrogenicity concentrations of 108 , 5 107 and 107 M of estradiol were competed by 103 -fold dilutions of the ethanolic or iso-propanolic extracts from CR. Lower dilutions of the extracts could not be tested because MVLN cells do not tolerate ethanol concentrations above 0.1%. 2.4. Proliferation assay with MCF-7 cells Prior to use in the proliferation assays MCF-7 cells were cultured for at least one passage in Eagles MEM without phenol red, supplemented with non-essential amino-acids, 10 g/ml insulin, 1 mM sodium pyruvate and 5% charcoal stripped fetal calf serum (CSF). The 200 l of a MCF-7 cell suspension (5 104 cells/ml) were plated per well into 96 microtiter plates and incubated at 37 C and 5% CO2 for 24 h. Thereafter, the supernatants were carefully removed and 150 l fresh culture medium was added. The test sub-

stances were dissolved, diluted in the cell culture medium and pipetted in four parallels at 100 l per well. As controls the cell culture medium as well as the corresponding solvent dilutions were tested simultaneously. 17 -Estradiol and tamoxifen were dissolved in DMSO and diluted in cell culture medium, the CR-extract was diluted directly with the cell culture medium. The CR-extract was tested in a dilution series of 103 to 8 in the MCF-7 proliferation assay, corresponding to 10 1000.001 g/ml (nal concentrations 400.0004 g/ml) in relation to the dry residue. Before the CR-extract was tested in the MCF-7 proliferation assay it was checked for its cytotoxic activity. Dilutions 103 (100 g/ml) showed no toxic activities. After 2 days of incubation at 37 C and 5% CO2 , cells were pulsed with 25 l per well 6-3 H-thymidine (spectrum activity 2 Ci/mmol, 0.25 Ci per well; Amersham, Braunschweig, Germany) for 8 h. Thereafter, the cells were harvested according to standard procedures (Cell Harvester, Inotech) onto glass ber lters and counted in a liquid scintillation counter. 2.5. Toxicity assay with MCF-7 cells For the determination of cytotoxicity a uorescence assay was performed using 4-methylumbelliferylheptanoat (4-MeUH) as uorogenic substrate for cellular esterases. The substrate is non-uorescent until taken up and cleaved by esterases in living cells. MCF-7 cells were plated at an initial cell density of 3 104 cells per well in Eagles MEM without phenol red, supplemented with non-essential amino-acids, 1 mM sodium pyruvate, 10 g/ml insuline and 5% CSF. After an incubation at 37 C and 5% CO2 for 24 h, test substances in cell culture medium were added. The microtiter plates were incubated for another 48 h and then centrifuged at 800 rpm for 10 min. The supernatants were removed and 200 l 4-MeUH (0.1 mg/ml in PBS) per well was added. After 120 min the uorescence units per well were measured in a microtiter plate uorimeter (Fluoroskan II). 2.6. Statistical analysis The results were expressed as mean standard deviation (S.D.). The pharmacological data were analyzed by the Students t-test and signicance was assumed at P < 0.05. 3. Results First, we tested whether ethanolic or iso-propanolic extracts from CR activate the estrogen-inducible yeast expression system. We treated yeast cells in a concentration depending manner using 109 to 102 -fold nal dilutions (0.00002200 g/ml for ethanolic extracts and 0.00011000 g/ml for iso-propanolic extracts in the test) of the primary extracts. Neither the ethanolic extract of CR

O. Zierau et al. / Journal of Steroid Biochemistry & Molecular Biology 80 (2002) 125130

127

Fig. 1. Estrogen-inducible yeast receptor assay. In this assay we tested for estrogenicity (a) and antiestrogenicity (b) of ethanolic (open circles) and iso-propanolic (solid circles) extracts from the rhizome of CR in a concentration dependent manner. Conditions of hormonal treatment are given in Section 2. Open squares in Fig. 1a represent the estradiol curve, open circles ethanolic extracts and solid circles iso-propanolic extracts of CR.

(Fig. 1a) nor the iso-propanolic extract (Fig. 1a) induced -galactosidase reporter gene activity from the ERE of the reporter construct, whereas estradiol treatment induced reporter gene activity in a concentration dependent manner (Fig. 1a). In contrast, if yeasts were simultaneously treated

with 5 1010 M of estradiol (Fig. 1b) and extracts from CR, estradiol induced reporter gene activity was inhibited in a concentration dependent manner. The 103 and 102 -fold dilutions of the primary extracts, corresponding to either 20200 (ethanolic extract) or 1001000 g/ml

128

O. Zierau et al. / Journal of Steroid Biochemistry & Molecular Biology 80 (2002) 125130

(iso-propanolic extrac) concentrations, were most effective (Fig. 1b). Before the CR extract was tested in the MCF-7 cell based assays, MVLN luciferase or proliferation assay, it was checked for its cytotoxic activity. Up to a dilution of 102 the extract induced toxic effects on the MCF-7 cells. Dilutions 103 (100 g/ml in relation to the dry residue)

showed no toxic activities (data not shown). The CD50 was calculated 357 g/ml (dry residue, least square method). In MVLN cells neither the ethanolic extract of CR, nor the iso-propanolic extract signicantly stimulated reporter gene activity, in contrast basal reporter gene activity detectable in the vehicle control appeared to be reduced (Fig. 2a). To test for antiestrogenicity MVLN cells were

Fig. 2. Estrogen-inducible MVLN cells: with this assay we tested for the estrogenic and antiestrogenic properties of ethanolic and iso-propanolic extracts from the rhizome of CR, using concentrations selected from results of the estrogen-inducible yeast receptor assay. Ceth = ethanolic extract of CR; Ciso = iso-propanolic extract from CR; E2 = estradiol. Concentration of the CR extracts refer to g/ml.

O. Zierau et al. / Journal of Steroid Biochemistry & Molecular Biology 80 (2002) 125130

129

Fig. 3. 3 H-thymidine incorporation assay. In order to study potential effects of extracts from CR on proliferation a 3 H-thymidine incorporation assay was performed.

simultaneously incubated with estradiol (108 to 107 M) and ethanolic or iso-propanolic extracts of CR using a 103 M dilution (100 g/ml for the iso-propanolic extract and 20 g/ml of the ethanolic extract). With this concentration the half maximal suppression of reporter gene activity in the estrogen-inducible yeast assay could be achieved. Using this experimental model estradiol induced luciferase (reporter gene) activity was only inhibited by the iso-propanolic extract (Fig. 2b). Due to a limit of tolerance towards higher concentrations of the solvent (ethanol or iso-propanol) of MVLN cells, lower dilutions representing higher concentrations could not be used. Finally, due to the higher activity in estrogen-inducible gene expression assays the antiestrogenic activity of the iso-propanolic extract from CR was studied in a cell proliferation assay using MCF-7 cells. 3 H-thymidine incorporation into MCF-7 cells following a treatment period of 48 h with estradiol was almost doubled. By a simultaneous treatment of MCF-7 cells with 108 M estradiol and 105 M tamoxifen thymidine incorporation was reduced below unstimulated control levels (Fig. 3). Extracts from CR if given alone in a dosage range of 1100 g/ml also signicantly reduced basal thymidine incorporation (Fig. 3). If MCF-7 cells were treated simultaneously with 108 M estradiol and increasing concentrations of CR, a signicant dose dependent inhibition of estradiol induced stimulation of

proliferation of MCF-7 cells in a dosis range of 1100 g/ ml of CR became apparent (Fig. 3).

4. Discussion The overall amount of information on potential hormonal activities of extracts from the rhizome of CR is limited. Nonetheless potential estrogenic effects of these extracts are controversely discussed. Our results provide unambigous evidence for a potential antiestrogenic activity of extracts from CR on the estrogen receptor- . It is out of question that components of the extracts are capable to bind to the estrogen receptor [8]. However, controversial results are described for cell or organ based assays. Findings of stimulated proliferation by individual components of the extract [2] are challenged by studies on T47D breast cancer cells or by another study with MCF-7 cells with whole extracts when antiproliferative effects became apparent [3,9]. The latter ndings are further substantiated by our results in the cell proliferation assay. Antiestrogenicity of extracts from CR can also explain ndings of Einer-Jensen et al. [4], not being able to detect estrogenic effects of the extracts in mice and rats on uterine growth and vaginal cornication in vivo. Another activity of extracts from the rhizome of CR related to estrogen action

130

O. Zierau et al. / Journal of Steroid Biochemistry & Molecular Biology 80 (2002) 125130 [3] D. Dixon-Shanies, N. Shaikh, Growth inhibition of human breast cancer cells by herbs and phytoestrogens, Oncol. Rep. 6 (1999) 13831387. [4] N. Einer-Jensen, J. Zhao, K.P. Andersen, K. Kristoffersen, Cimicifuga and Melbrosia lack estrogenic effects in mice and rats, Maturitas 25 (1996) 149153. [5] E.J. Routledge, J.P. Sumpter, Oestrogenic activity of surfactants and some of their degradation products assessed using a recombinant yeast screen, Environ. Toxicol. Chem. 15 (1996) 241248. [6] M. Pons, D. Gagne, J.C. Nicolas, M. Mehtali, A new cellular model of response to estrogens: a bioluminescent test to characterize antiestrogen molecules, Biotechniques 9 (1990) 450459. [7] E. Demirpence, M.J. Duchesne, E. Badia, D. Gagne, M. Pons, MVLN cells: a bioluminescent MCF-7-derived cell line to study the modulation of estrogenic activity, J. Steroid Biochem. Mol. Biol. 46 (1993) 355364. [8] H. Jarry, G. Harnischfeger, E. Dker, The endocrine effects of constituents of Cimicifuga racemosa. II. In vitro binding of constituents to estrogen receptors, Planta Med. 4 (1985) 316319. [9] T. Ne elhut, C. Schellhase, R. Dietrich, W. Kuhn, Studies on mamma carcinoma cells regarding the proliferative potential of herbal medications with estrogen-like effects, Arch. Gynecol. Obstet. 254 (1993) 817818. [10] E. Dker, L. Kopanski, H. Jarry, W. Wuttke, Effects of extracts from Cimicifuga racemosa on gonadotropin release in menopausal women and ovariectomized rats, Planta Med. 57 (1991) 420427. [11] B.S. Katzenellenbogen, I. Choi, R. Delage-Mourroux, T.R. Ediger, P.G. Martini, M. Montano, J. Sun, K. Weis, J.A. Katzenellenbogen, Molecular mechanisms of estrogen action: selective ligands and receptor pharmacology, J. Steroid Biochem. Mol. Biol. 74 (2000) 279285. [12] D.B. Muchmore, Raloxifene: a selective estrogen receptor modulator (SERM) with multiple target system effects, Oncologist 5 (2000) 388392. [13] L.A. Fitzpatrick, Selective estrogen receptor modulators and phytoestrogens: new therapies for post-menopausal women, Mayo Clin. Proc. 74 (1999) 601607.

has been described previously. The discussion whether CR has a potential estrogenic or antiestrogenic activity remains open. This is not at least a consequence of conicting results concerning treatment associated, selective reduction of LH, but not of FSH levels in humans and rats [10]. Taking together the various activities attributed from extracts of the rhizome of CR the results suggest two different possibilities. Either the extracts contain more than one active constituent, which was suggested before [10] or they contain components with activities comparable to SERM-like (selective estrogen receptor modulator) activities [11,12] or both. Either hypothesis would explain most of the existing data. In conclusion, using two independent cell based estrogeninducible assays and combining results, thereof, with those from in an in vitro cell proliferation assay we provide evidence for potential antiestrogenic activities of extracts from the rhizome of CR. Therefore, it will be challenging to identify and test individual components of the extracts thereby identifying new substances with phyto-antiestrogen or phyto-SERM activity, which may have benecial health effects [13]. References
[1] E. Liske, Therapeutic efcacy and safety of Cimicifuga racemosa for gynecological disorders, Adv. Ther. 15 (1998) 4553. [2] S.O. Kruse, A. Lhning, G.F. Pauli, H. Winterhoff, A. Nahrstedt, Fukiic and piscidic acid esters from the rhizome of Cimicifuga racemosa and the in vitro estrogenic activity of fukinolic acid, Planta Med. 65 (1999) 763764.