Human Reproduction Vol.17, No.3 pp. 589–594, 2002
Acute and chronic effects of genistein, tyrphostin and lavendustin A on steroid synthesis in luteinized human granulosa cells Saffron A.Whitehead1, Julia E.Cross, Clare Burden and Michael Lacey Department of Physiology, St George’s Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK 1To
whom correspondence should be addressed. E-mail:
[email protected]
BACKGROUND: Phytoestrogens, including genistein and other inhibitors of tyrosine kinases (TKs), inhibit specific steroidogenic enzymes. This study was designed to compare the effects of genistein, with two other TK inhibitors, on steroid synthesis in human granulosa luteal (GL) cells and to identify which steroidogenic enzymes they may affect. METHODS: GL cells, obtained from women undergoing IVF procedures, were cultured for various periods of time with and without substrates for progesterone and estradiol synthesis, in the presence or absence of the TK inhibitors. RESULTS: The TK inhibitors significantly suppressed progesterone and estradiol synthesis in a dosedependent manner over a 48 h culture period. Progesterone production in the presence of 10–7 mol/l pregnenolone during a 4 h period was inhibited by both acute (4 h) and chronic (24 h) exposure of GL cells to 50 µmol/l genistein (P < 0.05) whilst no significant effects of 50 µmol/l tyrphostin A23 were observed. Genistein (4 and 24 h exposure) inhibited the production of estradiol using 10–7 mol/l estrone as a substrate, but inhibition of estradiol synthesis using androstenedione or testosterone as substrates was only observed after a 24 h exposure. In contrast, tyrphostin acutely stimulated estradiol synthesis when androstenedione and testosterone were used as substrates (P < 0.05) but not estrone. CONCLUSIONS: Genistein directly inhibits 3 and 17β-hydroxysteroid dehydrogenase activity, whilst tyrphostin has an acute stimulatory effect on aromatase activity. Over a longer time (24 and/or 48 h period), both TK inhibitors suppress steroid synthesis. Key words: estradiol/genistein/human granulosa luteal cells/progesterone/tyrphostin
Introduction Epidemiological studies have suggested that the phytoestrogen genistein, which is found in substantial amounts in soy products, may have a protective effect against breast and prostate cancer (Knight and Eden, 1996; Kurzer and Xu, 1997). Since genistein can bind with estrogen receptors, albeit at a much lower affinity than estradiol (Miksicek, 1995; Kuiper et al., 1998), it has been assumed that its protective effects on endocrine-sensitive tumours may be derived from an antiestrogenic effect. Genistein, however, stimulates the growth of breast cancer cells lines and only blocks the 17β-estradiolstimulated proliferation of breast cancer cell lines at high doses (Wang and Kurzer, l998). Similar high doses of the phytoestrogen (⬎10 µmol/l) are also required to inhibit serumstimulated growth in breast cancer cell lines (Dixon-Shanies and Shaikh, 1999). In contrast, biological estrogenic effects of genistein have been observed in vivo (Awoniyi et al., 1998) and studies in vitro have demonstrated that it can stimulate transcriptional activity of both α and β estrogen receptors (ERs) in transfected cells (Kuiper et al., 1998; Morito et al., 2001). Such evidence has stimulated the promotion of soy as a natural alternative to hormone replacement therapy, though © European Society of Human Reproduction and Embryology
experimental studies do not demonstrate any marked estrogenic effects of dietary soy in menopausal women (Duncan et al., 1999). An alternative target for the action of genistein is on steroidogenic enzymes, and in this respect genistein has been shown to inhibit the reduction of estrone to the more active estradiol by inhibiting 17β-hydroxysteroid dehydrogenase (17β-HSD) type 1 (Ma¨kela¨ et al., 1995). It could therefore exert anti-estrogenic effects by inhibiting the production of the most potent estrogen, estradiol. More recently, it has been reported that genistein can inhibit the reductive/oxidative activity of 17β-HSD type 5, thereby inhibiting the conversion of androstenedione to testosterone and androstenediol to androsterone respectively (Krazeisen et al., 2001). Since both testosterone and androstenedione are substrates for the action of aromatase, which converts testosterone to estradiol and androstenedione to estrone, genistein may also affect the local conversion of these androgen precursors to the active estradiol or estrone. Indeed, it has been shown that genistein has a weak inhibitory activity on aromatase (Pelissero et al., 1996). Genistein is, however, also a potent inhibitor of tyrosine kinase activity (Akiyama et al., 1987). Whilst the classical 589
S.A.Whitehead et al.
stimulatory pathway for steroidogenesis is via a cyclic AMPstimulated signalling system, there is evidence of cross talk between adenylate cyclase- and tyrosine kinase (TK)-dependent signalling pathways in the control of steroidogenesis (Andreis et al., 2000). Recent experiments in our laboratory, concerning the effects of a variety of endocrine disrupting chemicals on steroidogenesis, have shown that genistein and the non-estrogenic TK inhibitor lavendustin A can potently and dose-dependently inhibit progesterone production in cultured rat granulosa luteal (GL) cells (Whitehead and Lacey, 2000). In light of these studies, and the evidence that local conversion of relatively inactive steroid metabolites to the highly active estradiol, may be important in the aetiology of breast cancer (Labrie et al., 2000), we hypothesized that the protective effect of genistein on breast cancer could be exerted by inhibiting the activity of the enzymes that convert steroid precursors to estradiol. We further hypothesized that such effects were induced by the inhibitory effect of genistein on protein TKs rather than its ability to bind weakly to estrogen receptors. Thus the present studies were undertaken to compare the effects of three TK inhibitors—genistein, lavendustin A and tyrphostin A23—on progesterone and estradiol production in human GL cells. Subsequently, we aimed to identify which enzymes might be targeted by genistein and tyrphostin A23 by using pregnenolone and androstenedione, testosterone or estrone as substrates for progesterone and estradiol synthesis respectively.
Materials and methods Cell cultures Granulosa luteal (GL) cells were obtained from the Assisted Conception Units at St George’s Hospital and King’s College Hospital, London after signed consent from the patients. Within 2–5 h of oocyte retrieval, GL cells were washed and purified on a 60% Ficoll gradient and cultured at a concentration of 2.5 ⫻ 104 cells/250 µl medium in 96-well culture plates. The culture medium was Dulbeco’s modified Eagle medium (DMEM) supplemented with 2 mmol/l glutamate, 100 IU/ml penicillin and 100 µg/ml streptomycin. The cells were cultured for 48 h in medium containing 5% fetal calf serum (FCS) and, after washing, the cells were cultured in serum-free medium, but in the presence of tyrosine kinase inhibitors, with or without recombinant (r)HCG or human (h)FSH for a further 48 h. Aromatase activity was investigated by adding 10–7 mol/l testosterone as a substrate for estradiol synthesis. The second series of experiments were carried out to investigate which steroidogenic enzymes might be affected by TK inhibitors by exposing the cultures to steroid substrates—pregnenolone, androstenedione, estrone or testerone—for 4 h at the end of two culture periods with subsequent measurement of progesterone or estradiol. GL cells were cultured for 48 h with fetal calf serum, for 24 h in serum-free medium but in the presence of hFSH or rHCG and, after changing the medium, for a further 4 h in the presence of drugs and the appropriate steroid substrate. After removal of the medium for the first steroid measurements, cells were washed and cultured for a further 24 h in the presence of the same TK inhibitors. Subsequently, in fresh medium, cells were exposed to the same steroid substrates for a final 4 h to determine enzyme activity. In this way the same samples could be used to investigate the possible acute action of TK
590
inhibitors on steroidogenic enzymes as well as their ‘chronic’ effects over a 24 h period. At the end of all experiments, media samples were taken and stored at –20°C and viability of cells was tested either with the Trypan Blue exclusion test or by measuring cellular dehydrogenases with methyl thiazolyl tetrazolium (MTT) (Shakil and Whitehead, 1994). Drugs and chemicals The following drugs were used in these experiments: androstenedione, genistein (4⬘,5,7-trihydroxyisoflavone), lavendustin A (5-amino-[N2,5-dihydroxybenzyl-N⬘-2-hydroxybenzyl] salicylic acid), estrone, pregnenolone, testosterone, tyrphostin A23 ([3,4-dihydroxybenzylidene] malono-nitrile, α-cyano-[3,4-dihydroxy] cinnamonitrile), hFSH and rHCG. With the exception of hFSH and rHCG, kindly supplied by The National Hormone and Pituitary Agency, Torrance, CA, USA, and lavendustin A and tyrphostin A 23 supplied by the Alexis Corporation, Nottingham, UK, all other drugs, chemicals and culture media were supplied by Sigma, Dorset, UK. Steroids, genistein and tyrphostin A23 were initially dissolved in ethanol (methanol for lavendustin A) and diluted appropriately with culture medium before they were stored as stock solutions. Stock solutions were then diluted appropriately with medium and added to cultures in 10 µl volumes to give the desired final concentration. Controls were performed to ensure that the maximum volume (1 µl) of ethanol/methanol diluent did not affect cellular responses. Steroid assays Progesterone and estradiol concentrations in the culture medium were measured in duplicate by direct radioimmunoassay kits (ICN Pharmaceuticals Ltd, Basingstoke, UK) according to the manufacturer’s instructions. All drugs used in these experiments were tested for their possible cross reactivity with the anti-sera used, but with the exception of estrone, none was detected. The cross reactivity of the progesterone anti-serum with 20α-dihydroprogesterone, 17αhydroxyprogesterone and pregnenolone was 5.4, 0.6 and 0.4% respectively and the estradiol anti-serum with estrone, estriol, progesterone and testosterone was 20, 1.5, ⬍0.01 and ⬍0.01%. Inter- and intraassay coefficients of variation were 6.2 and 3.0% for the progesterone assay and 9.1 and 5.2% for the estradiol assay respectively. In view of the high cross reactivity of the estradiol antiserum with estrone, in each experiment 10–7 mol/l estrone was incubated in culture medium without cells for 4 h. The mean concentration of estradiol in the ‘blank’ triplicate wells was subsequently subtracted from the estradiol concentrations measured in the cell cultures that had been exposed to 10–7 mol/l estrone. Statistical analysis Data shown represent mean ⫾ SEM of triplicate cultures obtained from at least three independent experiments and n ⫽ the total number of observations. Comparisons were only made with paired controls within each experiment and dose–responses were obtained in the same experiments. This was to ensure that inter-experimental variability did not bias the analysis. Statistical differences in the dose–responses to the different drugs were compared with an analysis of variance followed by Gabriel’s test, which is suitable for groups of unequal size. A Student’s t-test was used to test significance between two groups of data.
Results Basal and stimulated steroid production in human GL cells was highly variable between different pooled samples, as was
Tyrosine kinase inhibitors and steroid synthesis
their responsiveness to gonadotrophins. Less variability was seen when steroid production was only measured over a 4 h period using appropriate steroid substrates, as was the case in the second series of experiments. Genistein, lavendustin A and tyrphostin A23 all significantly inhibited basal and rHCG-induced progesterone synthesis when GL cells were exposed to these drugs over a 48 h period (Figure 1A–C). Whilst significant stimulation of progesterone production by rHCG alone was seen in those experiments investigating the effects of genistein and, to a lesser extent, lavendustin A, no effect of this gonadotrophin was seen in the experiments using tyrphostin A23. There is no obvious explanation for this discrepancy, except that the initial experiments comparing the action of genistein with lavendustin A were carried out on GL cells obtained exclusively from St George’s Hospital. However, the inhibitory effects of all these compounds were dose-dependent, though genistein induced a significant inhibition at 1 µmol/l, lavendustin A at 10–50 µmol/l and tyrphostin A23 at 50–100 µmol/l. Since both lavendustin A and tyrphostin A23 had similar effects on progesterone production and tyrphostin has been more widely used to investigate the effects of TK inhibition on follicular steroidogenesis, our subsequent experiments only compared the effects of genistein and tyrphostin A23 on steroidogenesis. These compounds also induced a dose-dependent inhibition of estradiol production when GL cells were cultured in the presence of these inhibitors for a period of 48 h (Figure 2A,B). With genistein, significant inhibition was only observed with a dose of 50 µmol/l; thus, in subsequent experiments a concentration of 50 µmol/l genistein or tyrphostin A23 was chosen. Different response profiles were observed when progesterone and estradiol production were measured over a 4 h period using appropriate steroid substrates. Simultaneous addition of TK inhibitors and steroid substrates after a 72 h pre-incubation period (see Materials and methods) showed that genistein had an acute and significant inhibitory effect on the production of progesterone in the presence of pregnenolone. There was a trend for Tyrphostin A23 to reduce progesterone synthesis, but the results were not statistically different from controls (Figure 3a). When the same cultures were subsequently exposed to the TK inhibitors for 24 h (termed ‘chronic’ treatment) and then challenged with pregnenolone for 4 h, a similar pattern of responses was observed (Figure 3b). Acute exposure of GL cells to genistein had no effect on the production of estradiol when either androstenedione or testosterone were used as substrates, but when estrone was
Figure 1. (A–C) Dose-dependent effects of genistein, lavendustin A and tyrphostin A23 on progesterone production. Cells were cultured for 48 h in the presence of 5% fetal calf serum and for a further 48 h in serum-free medium, but in the presence of increasing doses of TK inhibitors with or without rHCG. Data represent mean ⫾ SEM of three to four independent experiments of triplicate observations. Different letters denote statistical significance, P ⬍ 0.05–0.01, Gabriel’s test, *P ⬍ 0.05 compared with the control value, Student’s t-test.
591
S.A.Whitehead et al.
Figure 3. Acute (4 h) and chronic (24 h) effects of genistein (G) and tyrphostin A23 (T) on the conversion of pregnenolone to progesterone. Granulosa luteal cells were cultured for 48 h in medium containing 5% fetal calf serum and for a further 24 h in serum-free medium but in the presence of hFSH. The TK inhibitors (50 µmol/l) and pregnenolone (10–7 mol/l) were then added to fresh medium for 4 h (a). The medium was then changed and the same TK inhibitors were added for a further 24 h. Pregnenlone was then added for the final 4 h of the experiment after another change of medium (b). Data represent mean ⫾ SEM of three independent experiments (triplicate observations) and * P ⬍ 0.01 compared with control values, Student’s t-test.
term (48 h) exposure of GL cells to this drug could not be attributed to loss of cell viability. Figure 2. (A) and (B) Dose-dependent effects of genistein and tyrphostin A23 on estradiol (E2) production in granulosa luteal cells. Cells were cultured for 48 h in the presence of 5% fetal calf serum and for a further 48 h in serum-free medium in the presence of the tyrosine kinase inhibitors and 10–7 mol/l testosterone as a substrate for estradiol synthesis. Data represent mean ⫾ SEM of three independent experiments of triplicate observations. Different letters denote statistical significance of P ⬍ 0.05, Gabriel’s test, *P ⬍ 0.05 compared with the control value, Student’s t-test.
used as the substrate, estradiol production was significantly reduced compared with controls (Figure 4a). Chronic 24 h exposure of the same GL cells to genistein reduced estradiol production from all three steroid substrates, although in the presence of testosterone the reduction failed to reach statistical significance (Figure 4b). In contrast to the inhibitory effects of tyrphostin A23 on the production of estradiol from testosterone during a 48 h culture period, significant stimulatory effects of the TK inhibitor on estradiol synthesis were observed when GL cells were simultaneously exposed to tyrphostin A23 and either androstenedione or testosterone for a 4 h period. Estradiol production in the presence of estrone was unaffected (Figure 4a). When the same cultures of GL cells had been exposed to tyrphostin A23 for 24 h, the conversion of all three substrates to estradiol over a 4 h period was not statistically different from controls, irrespective of the substrate (Figure 4b). Routine monitoring of cell viability showed that neither lavendustin A, genistein or tyrphostin A23 had any effects on cell viability compared with controls (Table I) and thus even the change in effect of tyrphostin A23 from stimulatory to inhibitory that occurred between the short-term (4 h) and long592
Discussion The present experiments define certain actions of TK inhibitors on steroidogenesis and show that the effects of these drugs are both dose- and time-dependent. Over a 48 h period, genistein and tyrphostin A23 induced a dose-dependent inhibition of progesterone and estradiol production (Figures 1 and 2). Genistein, but not tyrphostin A23, also induced an acute (4 h) inhibition of progesterone and estradiol production in the presence of their respective precursors, pregnenolone and estrone (Figures 3a and 4a). There was no acute effect of genistein on estradiol synthesis from either androstenedione or testosterone. These data suggest that genistein may directly inhibit the activity of both 3β-hydroxysteroid dehydrogenase (3β-HSD) and 17β-HSD type 1, but not aromatase. The metabolism of testosterone and androstenedione to estradiol requires aromatase and, for androstenedione, additionally 17βHSD type 5, if it is shunted via testosterone, or 17β-HSD type 1 if the pathway is via estrone (Labrie et al., 2000). After a 24 h exposure of GL cells to genistein there was a generalized inhibition of enzyme activity (Figure 4b), as was seen after 48 h. These results suggest that genistein may have a direct, acute effect on the activity of specific enzymes in the steroidogenic pathway, but that inhibition of other enzymes requires suppression of enzyme expression or other genomic effects. The time course of the effects of tyrphostin A23 were very different. An acute 4 h exposure of this drug tended to reduce progesterone synthesis, but stimulated estradiol production when androstenedione or testosterone were used as substrates.
Tyrosine kinase inhibitors and steroid synthesis
Figure 4. Acute (4 h) and chronic (24 h) effects of genistein (G) and tyrphostin A23 (T) on the conversion of androstenedione, estrone and testosterone to estradiol (E2). Granulosa luteal cells were cultured for 48 h in medium containing 5% fetal calf serum and for a further 24 h in serum-free medium, but in the presence of 2 ng/ml hFSH. The TK inhibitors (50 µmol/l) and substrates for estradiol synthesis (10–7 mol/l) were then added to fresh medium for 4 h (a). The medium was then changed and the same TK inhibitors were added for a further 24 h. Steroid substrates were then added for the final 4 h of the experiment after another change of medium (b). Data represent mean ⫾ SEM of three independent experiments (triplicate observations) and *P ⬍ 0.01 compared with control (C) values, Student’s t-test.
Table I. Effects of genistein, tyrphostin A23 and lavendustin A on cell viability as assessed by the ability of cellular dehydrogenases to reduce MTT Control
50 µmol/l genistein
50 µmol/l tyrphostin
50 µmol/l lavendustin A
0.61 ⫾ 0.11
0.80 ⫾ 0.14
0.88 ⫾ 0.13
0.72 ⫾ 0.15
Results are presented as mean ⫾ SEM of optical density measurements (570–630 nmol/l) obtained from granulosa luteal cell cultures after they had been exposed to the drugs for 48 h (n ⫽ 12–15)
No effects were observed in the presence of estrone (Figure 4a,b). These data suggest that tyrphostin might stimulate the activity of aromatase. After GL cells had been exposed to tyrphostin A23 for 24 h, all steroidogenic responses were similar to controls and it was not until cells has been exposed to this inhibitor for 48 h that both progesterone and estradiol were inhibited. The present results agree with our recent studies on the effects of genistein and lavendustin A on progesterone production in rat granulosa cells (Whitehead and Lacey, 2000), although in these studies acute effects on enzyme activity were not investigated. Similarly Haynes-Johnson et al. found that the TK inhibitor RG 50810 and genistein inhibited FSHstimulated estradiol and progesterone production in rat granulosa cells (Haynes-Johnson et al., 1999). Gregoraszczuk et al. also reported that genistein, tyrphostin and herbimycin inhibited prolactin-stimulated progesterone production by porcine thecal and luteal cells, tyrphostin being least potent in this respect (Gregoraszczuk et al., 1999). In fetal and post-natal human adrenal cortical cells, genistein and another phytoestrogen, diadzein, suppressed cortisol synthesis but not androgen synthesis (Mesiano et al., 1999). There was no reduction in the expression of any steroidogenic enzyme, only a reduced activity of 21-hydroxylase. In line with the current experiments, Ma¨ kela¨ et al. reported that genistein, as well as several other phytoestrogens, including apigenin and coumesterol, inhibited 17β-HSD type 1 (Ma¨ kela¨ et al., 1995). This may have important clinical implications. Expression of 17β-HSD type 1 has not only been found in the human ovary, but also in breast tissue, and its expression is sometimes high in the stromal cells of malignant tissue (Poutanen et al., l995). The growth-promoting effects of estradiol on breast cancer cells can only be mimicked by estrone when cells are cultured in the presence of 17β-HSD type 1, but not in the absence of this enzyme (Miettinen et al., 1996). It is possible that the inhibition of this enzyme by genistein could explain the association between dietary soy intake and the lower incidence of breast cancer. Another enzyme which has been shown to be targeted by genistein is 17-βHSD type 5, which reduces androstenedione to testosterone. Krazeisen et al. demonstrated that, in a cellfree system, genistein inhibited the activity of recombinant human 17-βHSD type 5 with an IC50 ⬎20 µmol/l (Krazeisen et al., 2001). However, they also found that a wide range of other phytoestrogens inhibited the activity of this enzyme, many of them at lower IC50 values. For example, biochanin A and coumesterol were inhibitory at ⬍15 µmol/l and zearalenone at ⬍5 µmol/l. Thus, genistein had relatively weak inhibitory activity in this model. The lack of any acute effects of genistein on the conversion of androstenedione to estradiol observed in the present experiments could be explained by the relatively low dose of genistein (50 µmol/l) used in a cell culture that could convert genistein to inactive metabolites. There is increasing evidence that TKs can interact with other cell signalling pathways and in this way alter the activity/ expression of steroidogenic enzymes. For example tyrphostins, as well as phytoestrogens such as genistein, can inhibit the degradation of cAMP by inhibiting certain phosphodiesterase 593
S.A.Whitehead et al.
isozymes (Nichols and Morimoto, 2000) and Andreis et al. demonstrated that tyrphostin A23 enhanced basal and ACTHinduced steroid secretion from dispersed human and rat adrenocortical cells (Andreis et al., 2000). Their results suggested that this effect was mediated by TK inhibition of phosphdiesterase and they concluded that the observed increase in intracellular cAMP concentrations increased PKA activation and hence steroid hormone production. Along a similar line of reasoning, the acute stimulatory effects of tyrphostin A23 seen in these experiments could be due to increased FSH-induced cAMP accumulation in the GL cells. It would appear that residual stimulatory effects are also observed after a 24 h exposure of GL cells to tyrphostin A23 (Figure 4b) but by 48 h, only inhibition was observed (Figure 2B). Overall, the present study shows that TK inhibitors suppress steroidogenesis in human GL cells, although it has not addressed the mechanism of such suppression. There is clearly a divergence between the acute effects of genistein and tyrphostin A23, but when cells are exposed to these drugs for a period of 48 h, both progesterone and estradiol production are dose-dependently inhibited. Such parallel effects of three different TK inhibitors on steroidogenesis suggest that TKs are important in the expression of steroidogenic enzymes. The observed acute effects of these drugs suggest a direct inhibition of 3β-HSD and 17β-HSD type 1 activity by genistein, but an increase of aromatase activity by tyrphostin A23, perhaps mediated by sustaining increased intracellular concentrations of cAMP induced by FSH. Whether or not any of the effects of genistein are mediated by its ability to bind to estrogen receptors remains to be determined. In this respect, there is emerging evidence that genistein may act on transcriptional processes and can even alter proliferation of cells which do not express estrogen receptors (Barnes et al., 2000). Acknowledgements The authors thank the National Hormone and Pituitary Program, Torrance, CA, USA for their kind gift of hFSH and rHCG. Julia Cross was supported by the Woolfson Foundation.
References Akiyama, T., Ishida, J., Nakagawat, S., Ogawara, H., Watanabe, S., Itoh, N., Shibuya, M. and Fukame, Y. (1987) Genistein, a specific inhibitor of tyrosine-specific protein kinases. J. Biol. Chem., 262, 5592–5595. Andreis, P.G., Neri, G., Tortorella, C., Gottardo, L. and Nussdorfer, GG. (2000) Tyrphostin-23 enhances steroid hormone secretion from dispersed human and rat adrenocortical cells. Endocrinol. Res., 26, 319–332. Awoniyi, C.A., Roberts, D., Veeramachaneni, D.N. and Hurst, B.S. (1998) Reproductive sequelae in female rats after in utero and neonatal exposure to the phytoestrogen genistein. Fertil. Steril., 70, 440–447. Barnes, S., Kim, H., Darley-Usmar, V., Patel, R., Xu, J., Lavalleur, J., Phipps, W.R. and Kurzer, M.S. (2000) Beyond ERα and ERβ: estrogen receptor binding is only part of the isoflavone story. J. Nutr., 130, 656S–657S.
594
Dixon-Shanies, D. and Shaikh, N. (1999) Growth inhibition of human breast cancer cells by herbs and phytoestrogens. Oncol. Rep., 6, 1383–1387. Duncan, A.M., Underhill, K.E., Xu, X. Lavalleur, J., Phipps, W.R. and Kurzer, M.S. (1999) Modest hormonal effects of soy isoflavones in postmenopausal women. J. Clin. Endocrinol. Metab., 84, 3479–3484. Gregoraszczuk, E., Slomczynska, M. and Stoklosowa, S. (1999) Effect of genistein, tyrphostin and herbimycin on prolactin-stimulated progesterone reproduction by procine thecal and luteal cells. J. Physiol. Pharmacol., 50, 477–484. Haynes-Johnson, D., Lai, M.T., Campen, C. and Palmer, S. (1999) Diverse effects of tyrosine kinase inhibitors on follicle-stimulating hormonestimulated estradiol and progesterone production from rat granulosa cells in serum-containing epidermal growth factor. Biol. Reprod., 61, 147–153. Knight, D.C. and Eden, J.A. (1996) A review of the clinical effects of phytoestrogens. Obstet. Gynecol., 87, 897–904. Krazeisen, A., Breitling, R., Mo¨ ller, G. and Adamski, J. (2001) Phytoestrogens inhibit human 17β-hydroxysteroid dehydrogenase type 5. Mol. Cell. Endocrinol., 171, 151–162. Kuiper, G.G.J.M., Lemmen, J.G., Carlson, B., Corton J.C., Safe, S.H., van der Saag, P.T., van der Burg, B. and Gustafsson, J.A. (1998) Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor β. Endocrinology, 139, 4252–4563. Kurzer, M.S. and Xu, X. (1997) Dietary phytoestrogens. Ann. Rev. Nutr., 17, 353–381. Labrie, F., Luu-The, V., Lin, S.X., Simard, J., Labrie, C., El-Alfy, M., Pelletier, G. and Be´ langer, A. (2000) Intracrinology: role of the family of 17β-hydroxysteroid dehydrogenases in human physiology and disease. J. Mol. Endocrinol., 25, 1–16. Ma¨ kela¨ , S., Poutanen, M., Lehtima¨ ki, J., Kostain, M.L., Santti, R. and Vihko, R. (1995) Estrogen specific 17β-hydroxysteroid oxidoreductase type 1 (E.C.1.1.1.62) as a possible target for the action of phytoestrogens. Proc. Soc. Exp. Biol. Med., 208, 51–59. Mesiano, S., Katz, S.L., Lee, J.Y. and Jaffe, R.B. (1999) Phytoestrogens alter adrenocortical functions: genistein and daidzein suppress glucocorticoid and stimulate androgen production by cultured adrenal cortical cells. J. Clin. Endocrinol. Metab., 84, 2443–2448. Miettinen, M., Poutanen, M.H. and Vihko, R.K. (1996) Characterization of estrogen-dependent growth of cultured MCF-7 human breast cancer cells expressing 17β-hydroxysteroid dehydrogenase type 1. Int. J. Cancer, 68, 600–604. Miksicek, R.J. (1995) Estrogenic flavonoids: structural requirements for biological activity. Proc. Soc. Exp. Biol. Med., 208, 44–50. Morito, K., Hirose, T., Kinjo, J., Hirakawa, T., Okawa, S., Inoue, S., Mura, M. and Masamune, Y. (2001) Interaction of phytoestrogens with estrogen receptors α and β. Biol. Pharm. Bull., 24, 351–356. Nichols, M.R. and Morimoto, B.H. (2000) Differential inhibition of multiple cAMP phosphodiesterase isozymes by isoflavones and tyrphostins. Mol. Pharmacol., 57, 738–745. Pelissero, C., Lenczowski, M.J. Chinzi, D., Davail-Cuisset, B., Sumpter, J.P. and Fostier, A. (1996) Effects of flavonoids on aromatase activity, an in vitro study. J. Steroid. Biochem. Mol. Biol., 57, 215–223. Poutanen, M., Isomaa, V., Peltoketo, H. and Vihko, R. (1995) Role of 17βhydroxysteroid dehydrogenase type 1 in endocrine and intracrine estradiol biosynthesis. J. Steroid Biochem. Mol. Biol., 55, 525–532. Shakil, T. and Whitehead, S.A. (1994). Inhibitory action of peritoneal macrophages on progesterone secretion from co-cultured rat granulosa cells. Biol. Reprod., 50, 1183–1189. Wang, C. and Kurzer, M.S. (1998) Effects of phytoestrogens on DNA synthesis in MCF- 7 cells in the presence of estradiol or growth factors. Nutr. Cancer, 31, 90–100. Whitehead, S.A. and Lacey, M. (2000) Protein tyrosine kinase activity of the phytoestrogen genistein and lavendustin A on progesterone synthesis in cultured ovarian cells of the rat. Fertil. Steril., 73, 613–619. Submitted on July 20, 2001; accepted on October 31, 2001
