Interaction of Phytoestrogens with Estrogen Receptors Α and Β (II)

Home , Biochanin A, Daidzin, Tectoridin, Tectorigenin

48 Biol. Pharm. Bull. 25(1) 48—52 (2002) Vol. 25, No. 1

Interaction of Phytoestrogens with Estrogen Receptors a and b (II)

a a a b c Keiko MORITO, Tohru AOMORI, Toshiharu HIROSE, Junei KINJO, Junichi HASEGAWA, d d e ,a Sumito OGAWA, Satoshi INOUE, Masami MURAMATSU, and Yukito MASAMUNE* Department of Molecular and Cellular Biology, Faculty of Pharmaceutical Sciences, Kanazawa University,a 13–1 Takara- machi, Kanazawa, Ishikawa 920–0934, Japan, Laboratory of Pharmacognosy, Faculty of Pharmaceutical Sciences, Fukuoka University,b 8–19–1 Nanakuma, Fukuoka 814–0180, Japan, Ichimaru Pharcos Co., Ltd.,c 318–1 Shinsei, Motosu- gun, Gifu 501–0475, Japan, Department of Geriatric Medicine, Graduate School of Medicine, The University of Tokyo,d 7–3–1 Hongo, Bunkyo-ku, Tokyo 113–8655, Japan, and Department of Biochemistry, Saitama Medical School,e 38 Morohongo, Moroyama-machi, Iruma-gun, Saitama 350–0451, Japan. Received June 29, 2001; accepted September 11, 2001

We investigated the estrogenic activities of isoflavone derivatives in competition binding assays with human estrogen receptor (hER) a or hER b protein, and in a gene expression assay using a yeast system. Coumestrol binds as strongly as 17b-estradiol to both hERs. Biochanin A, 5-OMe-genistein, formononetin, and tectorigenin bind well to hER b, but significant binding to hER a is only observed with 5-OMe-genistein, formononetin and tectorigenin. The binding of 7-OMe-genistein and irisolidone is poor to both receptors. Among the glucosides, sissotorin binds both receptors and the binding is stronger than genistin. Coumestrol induces transcription as strongly as genistein. Tectorigenin also induces transcription with both hERs. Though biochanin A, 5-OMe- genistein, 7-OMe-genistein, irisolidone and formononetin slightly induce transcription with hER b, they act as antagonists in the induction of transcription by 17b-estradiol. The results show that methylation or glucosidation of isoflavones generally inhibits their phytoestrogenic activities. Key words isoflavone; human estrogen receptor (hER) a and b; hER isoflavone binding; hER-dependent gene expression

Estrogens play important hormonal roles in all vertebrates. isoflavone derivatives by (A) binding to human estrogen re- Animal estrogens are exclusively steroidal compounds, and ceptor (hER) a and b, and (B) effect on estrogen receptor- the principal physiological estrogen in most species is 17b- dependent transcriptional expression.4) estradiol. Many plants produce isoflavones that possess estrogenic activity in animals and are, thus, called phytoestrogens. MATERIALS AND METHODS Two estrogen receptors (ERs) have been identified to date1,2) and the physiological responses to estrogen are known to be Chemicals [2,4,6,7-3H(N)]-17b-Estradiol (72 Ci/mmol) mediated within specific tissues by at least these two recep- was purchased from Dai-Ichi Pure Chemicals Co., Ltd. 4- tors. The ERs are a 3A member of the nuclear hormone re- OH-Tamoxifen was purchased from Sigma Chemicals Co., ceptor family and act as ligand-activated nuclear transcrip- Ltd. tion factors.3) Isoflavones Isoflavones examined in this paper are Among the foods consumed by humans, soybeans contain shown in Fig. 1. 5-OMe-Genistein and 7-OMe-genistein the highest concentration of isoflavones. We have examined were purchased from Nakahara Science Co., Ltd. Biochanin the estrogenic activity of these soy isoflavones (e.g., daidzin, A, irisolidone, tectorigenin, irisolidone-7-O-b-D-glucoside, genistin and glycitin) and their metabolites (e.g., daidzein, tectoridin and sissotorin were extracted from Pueraria thom- genistein, glycitein, equol, dihydrogenistein and dihydro- sonii and lobata.5) Coumestrol and formononetin were ex- glycitein) by enteric bacteria in the previous paper.4) In this tracted from P.lobata.6) paper, we examined the estrogenic activities of several other Preparation of the Extract of hER a and b hER a and

Fig. 1. Structures of Isoﬂavones Examined in This Paper

∗ To whom correspondence should be addressed. e-mail: [email protected] © 2002 Pharmaceutical Society of Japan January 2002 49 b was prepared following the method described in the previ- sured by adding various concentrations of the derivatives. ous paper.4) The concentrations of hER a and b were 0.6 and 0.3% of the total protein, respectively. These receptors were RESULTS stable at Ϫ80 °C for several months. Competition Assay of Isoflavone The binding of iso- Binding of Isoflavones to hERs Genistein and genistin, flavone derivatives to hER a or b was examined by the which were examined in the previous paper,4) were used as method described in the previous paper.4) Estrogen binding to controls. Genistein binds and induces transcription with both hER a or b was determined by incubation of 250 ml of TKE hERs (Figs. 2a, 3a). Genistin bound poorly and did not in- reaction mixture (20 mM Tris–HCl, pH 7.4, 1 mM EDTA, duce transcription (Figs. 2c, 3c). The binding of biochanin 3 20 mM KCl) containing 5 ml hER a or b with 1.25 pmol [ H]- A, 5-OMe-genistein, 7-OMe-genistein and 4-OH-tamoxifen, 17b-estradiol at 0 °C in the presence of various concentra- which is a strong competitive inhibitor of 17b-estradiol,7) to tions of isoflavone derivatives. hER a and b are shown in Fig. 2a. 4-OH-Tamoxifen binds as Yeast Strain Carrying Full-Length hER a or b The strongly as 17b-estradiol to both receptors. Data for tectori- preparation of Saccharomyces cerevisiae Y190 (MATa, ura3- genin, irisolidone, coumestrol and formononetin are shown 52, his3-D200, ade2-101, trp1-901, leu2-3, 112, gal4Dgal80D, in Fig. 2b and data for irisolidone-7-O-b-D-glucoside, sisso- URA3:::GAL-lacZ, cyhr2,LYS2:::GAL-HIS3) carrying torin and tectoridin are shown in Fig. 2c. Coumestrol binds pGBT9-hER a or pGBT9-hER and pGAD424-hTIF2 were as strongly as 17b-estradiol to both hERs. 5-OMe-Genistein, described in the previous paper.4) tectorigenin and formononetin bound well to both hERs, al- ER-Dependent Transcriptional Expression Induced by though slightly less than genistein. Biochanin A bound to Isoflavone The effect of isoflavones on the ER-dependent hER b, though not so strong as genistein. 7-OMe-genistein transcription of b-galactosidase in yeast was examined fol- and irisolidone bound poorly to both hERs. Among the glu- lowing the methods described by Morito et al.4) cosides, only sissotorin bound to both receptors. Anti-Estrogen Assay To examine the antagonistic activ- Figure 3a shows the results of b-galactosidase activities ity of the derivatives of isoflavones, the inhibition of b-galac- induced by biochanin A, 5-OMe-genistein and 7-OMe-genis- tosidase activity induced by 1 nM 17b-estradiol was mea- tein and 4-OH-tamoxifen. Figure 3b shows those of tectori-

Fig. 2. Assays of Binding of Estrogenic Compounds to hERs (a) ᭹, 17b-estradiol; ᭺, genistein; ᭿, biochanin A; ᮀ, 5-OMe-genistein; ᭡, 7-OMe-genistein; ᭝, 4-OH-tamoxifen, (b) ᭹, 17b-estradiol; ᭺, tectorigenin; ᭿, irisolidone; ᮀ, coumestrol; ᭡, formononetin, (c) ᭹, 17b-estradiol; ᭺, genistin; ᭿, irisolidone-7-O-b-D-glucoside; ᮀ, sissotorin; ᭡, tectoridin. Binding to ERs was examined by competition as described in Materials and Methods. The bar at each point is the standard deviation of at least three independent experiments. 50 Vol. 25, No. 1

Fig. 3. Assays of Estrogen Receptor-Dependent b-Galactosidase Induction (a) ᭹, 17b-estradiol; ᭺, genistein; ᭿, biochanin A; ᮀ, 5-OMe-genistein; ᭡, 7-OMe-genistein; ᭝, 4-OH-tamoxifen, (b) ᭹, 17b-estradiol; ᭺, tectorigenin; ᭿, irisolidone; ᮀ, coumestrol; ᭡, formononetin, (c) ᭹, 17b-estradiol; ᭺, genistin; ᭿, irisolidone-7-O-b-D-glucoside; ᮀ, sissotorin; ᭡, tectoridin. ER-dependent b-galactosidase induction was measured by the increase of OD415 nm by o-nitrophenol produced by the digestion of o-nitrophenyl-b-D-galactoside. The bar at each point is the standard deviation of at least three independent experiments. genin, irisolidone, coumestrol and formononetin. Figure 3c DISCUSSION shows those of irisolidone-7-O-b-D-glucoside, sissotorin and tectoridin. Coumestrol induced b-galactosidase almost as Estrogens are critical to the functioning and maintenance strongly as genistein with both hERs. Tectorigenin induced of a diverse array of tissues and physiological systems in the enzyme significantly with both hERs. Biochanin A in- mammals. The physiological responses to estrogen are duced the enzyme with hER b. The induction by 5-OMe- known to be mediated within specific tissues by at least two genistein, 7-OMe-genistein, irisolidone and formononetin ERs, ER a and b.8) Isoflavones are known to have estrogenic was poor. The glucosides and 4-OH-tamoxifen did not in- activity.9) We are interested in their activities on hER a and duce transcription. b. In the previous paper,4) we examined isoflavones isolated Figure 4a shows the results of antagonistic activities of after the digestion of soybeans by enteric bacteria. The estro- genistein, biochanin A, 5-OMe-genistein and 7-OMe-genis- genic activities were examined with respect to their binding tein on the induction of b-galactosidase by 17b-estradiol. to hER a and b, ER-dependent transcriptional expression 4-OH-Tamoxifen is included as a control. Figure 4b shows and growth of MCF-7 cells. The binding and gene expression those of tectorigenin, irisolidone, coumestrol and for- of genistein was strongest among the isoflavones examined.4) mononetin. Figure 4c shows those of irisolidone-7-O-b-D- In this study, we examined other isoflavones isolated from glucoside, sissotorin and tectoridin. Strong inhibition of the Pueraria thomsonii and lobata and methylated derivatives of induction of b-galactosidase was observed with 4-OH-ta- genistein. The most active compound was coumestrol, which moxifen, biochanin A, 7-OMe-genistein, formononetin and is known to induce infertility in sheep10,11) and binds almost irisolidone with both hERs. Tectorigenin slightly inhibited as strongly as 17b-estradiol to both hERs. Though the induc- the induction with hER b. Genistein, 5-OMe-genistein and tion of transcription by coumestrol is almost the same as glycosides did not inhibit. Coumestrol slightly stimulated the genistein, the binding to hER a is much stronger than that of induction by 17b-estradiol. genistein. This strong binding to hER a might effect the in- It was confirmed that these derivatives did not affect the fertility. A distinctive character of coumestrol is the presence growth of yeast cells (data not shown). of a carbonyl group at position 2 and an ether bridge between January 2002 51

Fig. 4. Assays of Anti-Estrogenic Activities of Isoflavones (a) ᭺, genistein; ᭿, biochanin A; ᮀ, 5-OMe-genistein; ᭡, 7-OMe-genistein; ᭝, 4-OH-tamoxifen, (b) ᭺, tectorigenin; ᭿, irisolidone; ᮀ, coumestrol; ᭡, formononetin, (c) ᭺, genistin; ᭿, irisolidone-7-O-b-D-glucoside; ᮀ, sissotorin; ᭡, tectoridin. The inhibition of b-galactosidase activity induced by 1 nM 17b-estradiol was measured by adding various concentrations of isoflavone derivatives. Arrow indicates the concentration of 17b-estradiol (1 nM) added to the reaction mixture. The bar at each point is the standard deviation of at least three independent experiments. positions 2Ј and 4. This bridge induces a flat structure and in- both receptors, however, the inhibition by tectorigenin of the creases the similarity to 17b-estradiol, which may contribute induction of b-galactosidase by 17b-estradiol is weaker than to the strong binding of coumestrol to both hERs. The estro- that by formononetin. The reason is presumed to be that the genic activity of coumestrol may also be enhanced by the former induces transcription by itself, but the latter does not. carbonyl group at position 2. X-ray analysis of hER b bound Though formononetin is known to induce infertility in horses with genistein suggested that 4Ј-OH corresponds to the 3-OH and cows,13) the phytoestrogenic activity seems not so strong of 17b-estradiol and 7-OH correspond to the 17-OH of the as coumestrol. Activity is assumed to be induced after diges- hormone.12) Methylation or glucosylation of these sites gen- tion by enteric bacteria. 5-OMe-genistein and sissotorin com- erally weakens the binding to hERs and the induction of gene peted with 17b-estradiol at the hERs binding sites. However, expression. The exception is sissotorin, which is methylated they did not show any inhibition of gene expression. The per- at 4Ј and glucosylated at 7. Though it did not induce tran- meation of these compounds into yeast cells might be inhib- scription of b-galactosidase, it bound to both hERs. It was ited. confirmed that sissotorin is not degraded during the assay. It has been reported that intake of isoflavone reduced As shown in Fig. 4, biochanin A and 7-OMe-genistein in- the serum concentration of estradiol by feedback regula- hibited the induction of b-galactosidase by 17b-estradiol. tion,14—16) and that tectorigenin inhibited the synthesis of 17) Since both compounds did not compete with 17b-estradiol at prostaglandin E2 and the induction of cyclooxygenase-2. the binding site of hER a, it is very hard to explain the inhi- Recently, Simoncini et al. reported that they found a physio- bition. These compounds might compete with 17b-estradiol logically important non-nuclear estrogen-signaling pathway in terms of permeation into the cell. Since these compounds involving the direct interaction of hER a with PI(3)K.18) Tec- did not inhibit the growth of cells, the inhibition is not due to torigenin seems also to inhibit this pathway. cytotoxic effects. On the contrary, coumestrol slightly stimu- Isoflavones seem not only to antagonize 17b-estradiol but lated induction by 17b-estradiol. This stimulation is due to also affect various biochemical reactions. These functions of the strong inducing activity of this compound. Tectorigenin isoflavone will also help to reduce the risk of cancer. and formononetin bind with almost the same efficiency to 52 Vol. 25, No. 1

Acknowledgement This work was partially supported 7) Andrew K. S., Danielle B., Paula M. L., Lin C., Peter J. K., David A. by a grant from the Ministry of Health and Welfare of Japan. A., Geoffrey L. G., Cell, 95, 927—937 (1998). 8) Gustafsson J. Å., J. Endocrinol., 163, 379—383 (1999). 9) Davis S. R., Dalais F. S., Simpson E. R., Murkies A. L., Recent. Prog. REFERENCES Horm. Res., 54, 185—210, discussion 210—211 (1999). 10) Bennetts H. W., Underwood E. J., Shier F. L., Aust. Ver. J., 22, 2 1) Green S., Walter P., Greene G., Krust A., Gofﬁn C., Jensen E., Scrace (1946). G., Waterﬁeld M., Chambon P., J. Steroid Biochem., 24, 77—83 11) Mathieson R. A., Kitts W. D., J. Endocrinol., 85, 317—325 (1980). (1986). 12) Pike A. C., Brzozowski A. M., Hubbard R. E., Bonn T., Thorsell A. 2) Mosselman S., Polman J., Dijkema R., FEBS Lett., 392, 49—53 G., Engstrom O., Ljunggren J., Gustafsson J. Å., Carlquist M., EMBO (1996). J., 18, 4608—4618 (1999). 3) Nuclear Receptors Nomenclature Committee, Cell, 97, 161—163 13) Lundh T. J., Petterssosn H. I., Martinsson K. A., J. Agric. Food Chem., (1999). 38, 1530—1534 (1990). 4) Morito K., Hirose T., Kinjo J., Hirakawa T., Okawa M., Nohara T., 14) Mousavi Y., Adlercreutz H., Steroids, 58, 301—304 (1993). Ogawa S., Inoue S., Muramatsu M., Masamune Y., Biol. Pharm. Bull., 15) Cassidy A., Bingham S., Setchell K. D., Am. J. Clin. Nutr., 60, 333— 24, 351—356 (2001). 340 (1994). 5) Kinjo J., Aoki K., Okawa M., Nohara T., Nakajima Y., Yamasaki T., 16) Lu L. J., Anderson K. E., Grady J. J., Nagamani M., Cancer Epi- Hosono T., Someya M., Niiho Y., Kurashige T., Abstracts of Papers, demiol. Biomarkers Prev., 5, 63—70 (1996). The 45th Annual Meeting of the Japanese Society of Pharmacognosy, 17) Kim Y. P., Yamada M., Lim S. S., Lee S. H., Ryu N., Shin K. H., Sendai, Oct. 1998, p. 121. Ohuchi K., Biochim. Biophys. Acta, 1438, 399—407 (1999). 6) Kinjo J., Furusawa J., Baba J., Yamasaki M., Nohara T., Chem. Pharm. 18) Simoncini T., Hafezi-Moghadam A., Brazil D. P., Ley K., Chin W. W., Bull., 35, 4846—4850 (1987). Liao J. K., Nature (London), 407, 538—541 (2000).