Effect of flavonoids on androgen and glucocorticoid receptors based on in vitro reporter gene assay Yosuke Nishizaki a , Yoichi Ishimoto a , Yudai Hotta a , Akifumi Hosoda a , Hiromichi Yoshikawa b , Miki Akamatsu c , Hiroto Tamura a, * a Department of Environmental Bioscience, Meijo University, Nagoya 468-8502, Japan b Department of Functional Materials Engineering, Fukuoka Institute of Technology, Fukuoka 811-0295, Japan c Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan article info Article history: Received 18 May 2009 Revised 15 June 2009 Accepted 16 June 2009 Available online 23 June 2009 Keywords: Androgen Antiandrogen Glucocorticoid Flavonoid Structure–activity relationship abstract The effect of 32 flavonoids on androgen (AR) and glucocorticoid receptors (GR) was investigated using an MDA-kb2 human breast cancer cell line to predict potential AR and GR activities. Among them, 5-hydroxyflavone (7) had the highest AR antagonistic activity with an IC 50 value of 0.3 lM, whereas 6-methoxyflavone (11) had the highest induced luciferase activity with an EC 150 value of 0.7 lM. Genistein (2) and daizein (1) showed a sufficient increase of luciferase activities as their concentrations increased with EC 150 values of 4.4 and 10.1 lM, respectively. These findings provide evidence of a funda- mental property of their structure–activity relationship with AR and/or GR. Ó 2009 Elsevier Ltd. All rights reserved. Phytoestrogens with functional similarity to estrogens have been found to influence a variety of biological processes, including the reduction of cancer risk by modulating estrogen receptors (ER). 1,2 To date, the effect of isoflavones such as genistein and daidzein on the mammalian reproductive system have been well understood; however, few studies of dietary flavonoids derived from plants on their hormonal effects have been reported. Our pre- vious studies revealed that some ER agonists, such as estrone (E1), ethinyl estradiol (EE), estriol, diethylstilbestrol (DES) and alkyl phenols, act as androgen receptor (AR) antagonists. 3 Well-known estrogens agonists, such as isoflavones and related flavonoids, have heightened our interest in the structural requirements necessary to interact with AR and glucocorticoid receptors (GR) using estab- lished MDA-kb2 human breast cancer cells with endogenous AR and GR, and stably expressing AR and the GR-responsive luciferase reporter gene, MMTV-neo-luc. 3–6 Thus, this reporter gene assay system elicits an increase in luciferase activity (RLU) in the pres- ence of androgens or glucocorticoids. The assay procedures were described in detail previously. 3–6 Briefly, cells were treated with various concentrations of test chemicals from 10 À8 to 10 À4 M to determine the AR or GR agonist and/or antagonist activity in the absence and presence of AR agonist, dihydrotestosterone (DHT) or GR agonist, dexamethasone (Dex) at concentrations of 0.2 and 7 nM, respectively. Data were expressed as induced luciferase activity (%) compared to that of the control, that is, induced lucif- erase activity (%) = [RLU (test chemical)/RLU (control)] Â 100. When a test chemical has no activity, the obtained induced lucifer- ase activity (%) is 100%. When a test chemical has AR and/or GR agonistic activity, its induced luciferase activity (%) increases as its concentration increases in the absence of both DHT and Dex. When a test chemical has AR antagonist activity, its induced lucif- erase activity (%) decreases as its concentration increases in the presence of 0.2 nM DHT, by which induced luciferase activity was used as a control value. When a test chemical has AR and/or GR agonistic activity, its induced luciferase activity (%) also in- creases synergistically as its concentration increases in the pres- ence of 0.2 nM DHT or 7 nM Dex. In particular, the value of IC 50 for AR antagonistic activity is the concentration of the test chemi- cal producing 50% inhibition of 0.2 nM DHT-induced luciferase activity 3 and the value of EC 150 for AR and/or GR agonistic activity is the concentration of test chemical producing 150% stimulation of luciferase activity in the absence of both 0.2 nM DHT and 7 nM Dex. Since microscope observation of the damage to treated cells was correlated with decreased induced luciferase activity (%) in the absence of both DHT and Dex, induced luciferase activity less than 85% in the absence of both DHT and Dex was assigned for cell toxicity and the corresponding concentrations were not used to calculate IC 50 in the presence of DHT. The data were analyzed by Student’s t-distribution with Excel (Microsoft, USA) and p values 0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.bmcl.2009.06.073 * Corresponding author. Tel.: +81 52 838 2446; fax: +81 52 833 5524. E-mail address: hiroto@ccmfs.meijo-u.ac.jp (H. Tamura). Bioorganic & Medicinal Chemistry Letters 19 (2009) 4706–4710 Contents lists available at ScienceDirect Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl