European Journal of Pharmacology 691 (2012) 275–282

Contents lists available at SciVerse ScienceDirect

European Journal of Pharmacology

journal homepage: www.elsevier.com/locate/ejphar

Endocrine pharmacology Egonol gentiobioside and egonol gentiotrioside from Styrax perkinsiae promote the biosynthesis of by aromatase

Danfeng Lu, Lijuan Yang, Qilin Li, Xiaoping Gao, Fei Wang n, Guolin Zhang nn

Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, PR China article info abstract

Article history: Estrogen deficiency is associated with a variety of diseases, including , atherosclerosis, and Received 11 May 2012 Alzheimer’s disease. Aromatase cytochrome P450 is the only enzyme in vertebrates known to catalyze Received in revised form the biosynthesis of from . Inhibitors of aromatase have been developed for the 2 July 2012 treatment of estrogen-dependent breast cancer. However, small molecular agonists of aromatase, Accepted 2 July 2012 which would be useful to locally promote estrogen biosynthesis for the prevention of estrogen Available online 13 July 2012 deficiency-induced diseases, are rarely reported. In this study, we established a nonradioactive assay Keywords: for measuring aromatase activity by using human ovarian granulosa KGN cells and identified two Egonol estrogen biosynthesis-promoting compounds, egonol gentiobioside and egonol gentiotrioside from Estrogen Styrax perkinsiae. The compounds also promoted estrogen biosynthesis in 3T3-L1 preadipocyte cells. Aromatase Further study showed that neither compound affected the transcriptional and translational expression Styrax perkinsiae KGN cell line of aromatase in KGN cells, but that both significantly promoted the in vitro enzyme activity of recombinant expressed aromatase. Egonol gentiotrioside was also found to increase the serum estrogen level in ovariectomized rats. These results suggest that these two compounds may promote estrogen biosynthesis through the allosterical regulation of aromatase activity. Egonol gentiobioside and egonol gentiotrioside are, therefore, valuable targets for structural modification and warrant further investiga- tion for their potential as novel pharmaceutical tools for the prevention of estrogen deficiency-induced diseases. & 2012 Elsevier B.V. All rights reserved.

1. Introduction et al., 1995). In adipose cells and osteoblasts, aromatase expression is driven by promoter I.4, which is regulated by class I cytokines (Zhao Estrogens are hormones that regulate growth, differentia- et al., 1995; Shozu and Simpson, 1998). Thus, the regulation of tion, and function in a broad range of target tissues in the body. The estrogen biosynthesis in each tissue site of expression is unique, actions of estrogen are mediated by the estrogen receptors, which are and switches in promoter usage have been found to be associated expressed in a variety of cells and function through genomic or with the incidence of breast cancer (Simpson et al., 1997). nongenomic actions on target genes (Heldring et al., 2007). Estrogen Aromatase inhibitors, such as , , and exe- biosynthesis is catalyzed by aromatase (CYP19A1), which is respon- mestane, have been developed for the treatment of hormone- sible for binding of the C19 androgenic steroid substrate and dependent breast cancer in postmenopausal women since they catalyzing the necessary reactions to form the phenolic A ring show superiority over conventional anti- drugs characteristic of estrogens (Simpson et al., 2002). In humans, the such as (Johnston and Dowsett, 2003). In comparison expression of aromatase at the various sites is regulated by tissue- with the extensive number of studies dedicated to discovering specific promoters through the use of alternative splicing mechan- new aromatase inhibitors, the search for small molecular aroma- isms (Simpson, 2004). In the ovary and testes, aromatase expression tase agonists that promote estrogen biosynthesis has been small. is mediated by promoter II, which binds the transcription factors Although some herbicides, fungicides, and insecticides have been cAMP-response element binding protein (CREB) and steroidogenic found to promote aromatase expression in a cAMP-dependent or - factor-1. Aromatase expression inthegonadsisthusregulatedby independent manner (Sanderson et al., 2000; You et al., 2001; through the stimulation of cAMP generation (Michael Morinaga et al., 2004), more work is needed to find potent aromatase agonists that produce less side effects. Epidemiological studies have suggested the involvement of estrogen insufficiency in

n osteoporosis, neurodegenerative diseases, and cardiovascular dis- Corresponding author. Tel./fax: þ86 28 85256758. nn Corresponding author. Tel./fax: þ86 28 85229901. eases (Deroo and Korach, 2006). Estrogen therapy is an established E-mail addresses: [email protected] (F. Wang), [email protected] (G. Zhang). regimen for the prevention of these diseases, but recent evidence

0014-2999/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ejphar.2012.07.005 276 D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282 indicates that its long-term use is accompanied by side effects such 2.4. Cell-based estrogen biosynthesis assay as increased risks of breast, ovarian, and endometrial cancer (Davison and Davis, 2003). Thus, alternative methods that improve KGN cells or 3T3-L1 cells were seeded in 24-well plates the therapeutic efficacy and safety by locally promoting estrogen overnight. The following day, the cells were replaced with biosynthesis should be developed for the prevention and treatment serum-free DMEM/F-12 medium and pretreated with the test of these diseases caused by estrogen deficiency. chemicals for 24 h. (10 nM) was then added to each In this study, we established a nonradioactive assay for well, and the cells were incubated for a further 24 h. The culture measuring aromatase activity by using human ovarian granulosa medium and cell lysate were collected and stored at 20 1C. The KGN cells. Two natural products, egonol gentiobioside and egonol 17b- in the culture medium was quantified using a 17b- gentiotrioside from Styrax perkinsiae, were identified to signifi- estradiol magnetic particle-based ELISA according to the manu- cantly promote estrogen biosynthesis, and their effects on aro- facturer’s instructions. The OD was measured at 550 nm with the matase activity in vitro and in vivo were examined. Verioskan Flash Multimode Reader (Thermo Scientific, Waltham, MA, USA). The results, normalized to the total cellular protein content, were expressed as percentages of the control. Protein determination was carried out with the bicinchoninic acid (BCA) 2. Materials and methods protein assay kit (Bestbio, Shanghai, China). 2.1. Materials 2.5. Recombinant expressed aromatase activity assay Testosterone, , forskolin, dibenzylfluorescein, and An in vitro recombinant expressed aromatase activity assay nilestriol were purchased from Sigma (Shanghai, China), dissolved was conducted as described previously with minor modifications in DMSO to a concentration of 100 mM, and stored at 20 1C. (Stresser et al., 2000; Maiti et al., 2007). In brief, the test The NADPH regenerating system was purchased from Promega compounds (5 l) were preincubated with an NADPH regenerat- (Madison, WI, USA). The recombinant expressed human aroma- m ing system (45 l of 2.6 mM NADPþ , 7.6 mM glucose-6-phos- tase plus reductase was purchased from BD Biosciences (San Jose, m phate, 0.8 U/ml glucose-6-phosphate dehydrogenase, and 1 mg/ml CA, USA). The magnetic particle-based 17b-estradiol enzyme- albumin, in 50 mM potassium phosphate, pH 7.4) for 10 min at linked immunosorbent assay (ELISA) kit was purchased from 37 1Cbefore50 l of the enzyme and substrate mixture (40 pM Bio-Ekon Biotechnology (Beijing, China). Seeds of S. perkinsiae m recombinant aromatase and 0.4 M dibenzylfluorescein in 50 mM were collected from Yongde County within the Yunnan Province m potassium phosphate, pH 7.4) were added. The reaction mixture of China, in August 2003. The seeds were identified by Prof. Xin-fen was then incubated for 2 h at 37 1C to allow the aromatase to Gao of the Chengdu Institute of Biology, Chinese Academy of generate the product and quenched with 37.5 l of 2 N NaOH. The Sciences, based on the Flora Reipublicae Popularis Sinicae, in which m mixture was then shaken for 5 min and incubated for 2 h at 37 1Cto the Engeler system was used for the definition of the plant names. enhance the noise/background ratio. The fluorescence intensity was A voucher specimen (A-157) was deposited in the Herbarium of measured at 485 nm (excitation) and 530 nm (emission). Three the Chengdu Institute of Biology, Chinese Academy of Sciences. independent experiments were performed in duplicate.

2.2. Plant extraction and isolation 2.6. Quantitative real-time RT-PCR

The air-dried seeds of S. perkinsiae (1 kg) were extracted with Total cellular RNA was isolated using TRIzol reagent according 75% ethanol (3 8 l) for 1.5 h under reflux. The ethanol solution to the manufacturer’s instructions (Invitrogen). Total RNA (2 mg) was concentrated under reduced pressure. The resulting extract was reverse-transcribed using SuperScript III Reverse Transcrip-

(250 g) was suspended in warm H2O (2 l), and then the solution tase (Invitrogen) with oligo(dT)18 primers. Equal amounts (1 ml) was extracted with petroleum ether (5 3 l). The aqueous phase of cDNA were subjected to real-time quantitative PCR with the was added to D101 macropous resins, which were first eluted fluorescent dye SYBR Green I using a Chromo4 detection system with deionized water (15 l) and then desorbed with 75% ethanol. (Bio-Rad, Hercules, CA, USA). Real-time PCR reaction mixtures The 75% ethanol eluate was concentrated under reduced pressure contained 12.5 mlof2 TransStart Top Green qPCR SuperMix to give the residue (106 g). The isolation and identification of the (TransGen Biotech Co., Ltd., Beijing, China), 0.5 ml of each primer thirteen compounds from the seeds of S. perkinsiae, including (0.2 mM), and 1 ml of template cDNA. Sterile distilled water was egonol gentiobioside and egonol gentiotrioside, were as described added to a final volume of 25 ml. The primer pairs used in the before (Li et al., 2005). assays for aromatase and GAPDH were as follows: aromatase, 50-ACCCTTCTGCGTCGTGTC-30 (sense) and 50-TCTGTGGAAATCCT- GCGTCTT-30 (antisense), GAPDH, 50-CCACCCATGGCAAATTCCAT- 2.3. Cell culture GGCA-30 (sense) and 50-GGTGGACCTGACCTGCCGTCTAGA-30 (anti- sense). The thermal cycling conditions comprised an initial KGN cells (kindly supplied by Prof. Yiming Mu of the Chinese denaturation step at 95 1C for 10 s, followed by 40 cycles of PLA General Hospital, Beijing, China) were maintained in Dulbec- 95 1C for 5 s, 54 1C for 15 s, and 72 1C for 15 s. Standard curves co’s modified Eagle medium/Ham’s F-12 nutrient mix (DMEM/ were established for each primer set, and both reference and F-12) medium (Invitrogen, Carlsbad, CA, USA) supplemented with target gene reactions were carried out for each sample. The 5% (v/v) fetal bovine serum (Invitrogen), penicillin (100 units/ml), relative quantity (n-fold) of aromatase mRNA was calculated by 1 and streptomycin (0.1 g/l) in cell culture flasks at 37 Cina the D (DCt) method, using GAPDH as a reference amplified from humidified atmosphere containing 5% CO2. 3T3-L1 cells (obtained the same sample. from the Cell Resource Center, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences) were cultured at 37 1Cin 2.7. Western blotting DMEM supplemented with 5% (v/v) fetal bovine serum containing penicillin (100 units/ml) and streptomycin (0.1 g/l) at 37 1Cina KGN cells cultured in 60 mm dishes were lysed with RIPA Lysis humidified atmosphere containing 5% CO2. Buffer (Beyotime, Haimen, China) supplemented with protease D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282 277 inhibitor cocktail (Sigma). Protein lysate (30 mg) was loaded and As shown in Fig. 1A, the 17b-estradiol concentration was separated on a 10% sodium dodecyl sulfate-polyacrylamide gel significantly increased in KGN cells by using testosterone as a electrophoresis (SDS-PAGE) gel. After electrophoresis, the pro- substrate, indicating that KGN cells constitutively express high teins were blotted onto nitrocellulose membranes and probed amounts of aromatase, as previously reported (Nishi et al., 2001; with an anti-aromatase or anti-GAPDH antibody (Epitomics, Ohno et al., 2004). Forskolin significantly increased the produc- Burlingame, CA, USA) at 4 1C overnight, followed by a horseradish tion of 17b-estradiol. In contrast, formestane, an aromatase peroxidase-conjugated secondary antibody (Santa Cruz Biotech- inhibitor clinically used for the treatment of breast cancer nology, Santa Cruz, CA, USA), then enhanced chemiluminesence (Wiseman and Goa, 1996), was found to significantly inhibit detection (Amersham Bioscience, Piscataway, NJ, USA). 17b-estradiol biosynthesis. These results indicated that the estab- lished assay was suitable for identifying active compounds that could promote estrogen biosynthesis in KGN cells. 2.8. Experimental animals To isolate such compounds, we screened our chemical library consisting of 2367 natural product compounds and plant extracts All animal experiments were approved by the Bioethics for their effects on estrogen biosynthesis in KGN cells. As shown Committee of the Chengdu Institute of Biology, Chinese Academy in Fig. 1B, one plant ethanolic extract from the seeds of S. of Sciences, and the experimental procedures were conducted perkinsiae was found to significantly promote 17b-estradiol bio- strictly in accordance with generally accepted international rules synthesis at 1–100 mg/ml in a concentration-dependent manner. and regulations. Fifty female Sprague-Dawley rats aged 12 weeks Benzofurans are the major compounds in this genus (Akgul and were purchased (Institute of Laboratory Animals of Sichuan Anil, 2003; Li et al., 2005; Luo et al., 2007; Park et al., 2007). Thus, Academy of Medical Sciences & Sichuan Provincial People’s this result indicates that the benzofurans in S. perkinsiae may be Hospital) and acclimated to the laboratory conditions for 1 week before the experiment. The animals were housed in an air- conditioned room with a 12 h/12 h light–dark illumination cycle at a constant temperature (25 1C) and humidity (45–50%). Food and drinking water were supplied ad libitum. Rats with an average weight of 200 g were chosen for the experiments and weighed weekly. The rats were randomly divided into five groups (n¼10). Four groups were ovariectomized (OVX) and the other group was sham operated. Seven days after surgery, the sham-operated group was treated orally with the vehicle, whereas the OVX rats were divided into four groups and treated with vehicle (water), nilestriol (0.03 mg/kg/day), egonol gentiotrioside (30 mg/kg/day) by intragastric administration, and egonol gentiotrioside (15 mg/ kg/day) by intramuscular injection, for 30 days. The serum was then extracted from the rats and 17b-estradiol levels were determined using a 17b-estradiol magnetic particle-based ELISA.

2.9. Statistical analysis

All statistical analyses were carried out with GraphPad Prism 5.0. The results are expressed as the mean (standard deviation) of the individual values from 3 independent experiments. One-way ANOVAs, followed by Duncan’s multiple range tests, were used for the statistical comparisons. The P value less than 0.05 was considered to be significantly different from the control.

3. Results

3.1. Effect of S. perkinsiae extract on estrogen biosynthesis in human ovarian granulosa-like cells

To obviate the limitations caused by the use of isotope-labeled substrates and cancer cell lines expressing weak aro- matase activity, we first established a convenient and sensitive assay to examine estrogen biosynthesis in the human granulosa- like tumor cell line (KGN), using a magnetic particle-based ELISA. To minimize the effects of present in the fetal bovine Fig. 1. The effect of an extract of S. perkinsiae seeds on estrogen biosynthesis in KGN serum, experiments were performed using culture medium with- cells. (A) KGN cells were seeded in 24-well plates overnight, and pretreated with the test compounds for 24 h. Subsequently, the cells were supplemented with out serum, and there was no significant cytotoxicity when the 10 nM testosterone for a further 24 h. The 17b-estradiol concentration in the KGN cells were cultured in the serum-free medium for 48 h culture medium was quantified using a 17b-estradiol magnetic particle-based (Fig S1). Forskolin, an adenylate cyclase agonist, has been shown ELISA. (B) KGN cells seeded in 24-well plates were pretreated with S. perkinsiae to increase intracellular cAMP and increase the expression of seed extract at the indicated concentrations for 24 h. The cells were then supplemented with 10 nM testosterone for a further 24 h and the 17b-estradiol aromatase by activating the protein kinase A/CREB pathway concentration in the culture medium was quantified. T, 10 nM testosterone. FSK, (Watanabe and Nakajin, 2004), thus it was used here as a positive 10 mM forskolin. FOR, 10 mM formestane. Cont, DMSO. *Po0.05, **Po0.01, control. ***Po0.001 compared with the control (n¼3). 278 D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282 the active compounds responsible for promoting estrogen bio- 3.3. Effect of egonol gentiobioside and egonol gentiotrioside on synthesis in KGN cells. estrogen biosynthesis in 3T3-L1 adipocytes

After , adipose tissue becomes the major source of 3.2. Effect of isolated compounds on estrogen biosynthesis in KGN estrogens. To determine whether egonol gentiobioside and egonol cells gentiotrioside also affect estrogen biosynthesis in adipose tissue, 3T3-L1 preadipocyte cells were treated with various concentra- Thirteen compounds from the seeds of S. perkinsiae were tions of these compounds. As shown in Fig. 3, both compounds isolated and identified, four of which were new compounds (Li significantly promoted 17b-estradiol biosynthesis. At 10 mM, et al., 2005), and their effects on estrogen biosynthesis were approximately 1.53- and 1.71-fold increases in 17b-estradiol examined in KGN cells. Only egonol gentiobioside and egonol biosynthesis were observed with egonol gentiobioside and egonol gentiotrioside were found to significantly promote 17b-estradiol gentiotrioside, respectively, compared with the control. The biosynthesis, and others had no effect (data not shown). The formestane completely abolished the effect chemical structures of egonol gentiobioside and egonol gentio- of the two compounds on 17b-estradiol biosynthesis. The two trioside are shown in Fig. 2A. As shown in Fig. 2B, approximately compounds had no effect on 17b-estradiol biosynthesis in the 1.62- and 1.95-fold increases in 17b-estradiol biosynthesis were absence of testosterone, indicating that their effects were not observed with egonol gentiobioside and egonol gentiotrioside at caused by cross-reactivity with the ELISA assay (Fig S2). There 10 mM, respectively, compared with the control. The aromatase was no significant cytotoxic effect when the 3T3-L1 cells were inhibitor formestane completely abolished the effects of the two cultured in the serum-free medium for 48 h (Fig S1). compounds on 17b-estradiol biosynthesis. These results suggest that egonol gentiobioside and egonol gentiotrioside might be the 3.4. Egonol gentiobioside and egonol gentiotrioside promote active compounds responsible for the 17b-estradiol biosynthesis- estrogen biosynthesis through aromatase promoting activity of the seeds of S. perkinsiae. The two com- pounds had no effect on 17b-estradiol biosynthesis in the absence KGN cells cannot synthesize androgen or estrogen by them- of testosterone, indicating that their effects were not caused by selves due to the absence of 17a-hydroxylase (Nishi et al., 2001), cross-reactivity with the ELISA assay (Fig S2). and an aromatase inhibitor completely abolished the effect of egonol gentiobioside and egonol gentiotrioside on 17b-estradiol biosynthesis in KGN cells (Fig. 2B) and 3T3-L1 cells (Fig. 3). Therefore, the effects of egonol gentiobioside and egonol gentio- trioside on 17b-estradiol biosynthesis in the presence of testos- terone may be caused by aromatase, the only enzyme that converts testosterone into 17b-estradiol. To examine whether the estrogen biosynthesis-promoting effect of these two com- pounds was caused by an increased aromatase protein expression or stability, the aromatase protein levels were detected in KGN cells treated with these compounds. As shown in Fig. 4A, forskolin significantly increased aroma- tase protein expression, as reported previously (Watanabe and Nakajin, 2004). Compared with the negative control, egonol

Fig. 2. The effects of the isolated compounds on estrogen biosynthesis in KGN cells. (A) The chemical structures of egonol gentiobioside and egonol gentiotrioside. Fig. 3. The effects of egonol gentiobioside and egonol gentiotrioside on estrogen (B) KGN cells seeded in 24-well plates were pretreated with the compounds at the biosynthesis in 3T3-L1 cells. 3T3-L1 cells seeded in 24-well plates were pretreated indicated concentrations for 24 h. Subsequently, the cells were supplemented with the compounds at the indicated concentrations for 24 h. Subsequently, the with 10 nM testosterone for a further 24 h and the 17b-estradiol concentration in cells were supplemented with 10 nM testosterone for a further 24 h and the 17b- the culture medium was quantified. For the formestane treatment, 10 mM estradiol concentration in the culture medium was quantified. For the formestane formestane was added to the 10 mM EGB- or EGT-treated KGN cells for 24 h, and treatment, 10 mM formestane was added to the 10 mM EGB- or EGT-treated 3T3-L1 then supplemented with 10 nM testosterone for a further 24 h. Cont, DMSO. FSK, cells for 24 h, and then supplemented with 10 nM testosterone for a further 24 h. 10 mM forskolin. EGB, egonol gentiobioside. EGT, egonol gentiotrioside. þFOR, 10 mM Cont, DMSO. FSK, 10 mM forskolin. EGB, egonol gentiobioside. EGT, egonol formestane plus 10 mM EGB or EGT. *Po0.05, **Po0.01, ***Po0.001 compared with gentiotrioside. þFOR, 10 mM formestane plus 10 mM EGB or EGT. *Po0.05, the control (n¼3). **Po0.01 compared with the control (n¼3). D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282 279

(0.1–1 mM), both egonol gentiobioside and egonol gentiotrioside significantly promoted dibenzylfluorescein hydrolysis, indicating a promoting effect of the two compounds on the aromatase activity. However, at a high concentration (10 mM) both compounds inhibited dibenzylfluorescein hydrolysis, possibly due to nonspecific inhibition by the chemical compounds present at high concentrations, as observed by others (McGovern et al., 2002). The aromatase inhibitor formestane completely abolished the effects of both egonol gentio- bioside and egonol gentiotrioside (0.1–1 mM) on dibenzylfluorescein hydrolysis, indicating that the two compounds acted directly through the recombinant aromatase. The change in dibenzylfluorescein hydrolysis was directly mediated by aromatase, because both egonol gentiobioside and egonol gentiotrioside had no significant effect on dibenzylfluorescein hydrolysis in the absence of aromatase (Fig S3).

3.5. Egonol gentiotrioside promotes estrogen biosynthesis in vivo

To examine whether egonol gentiotrioside could also promote estrogen biosynthesis in vivo, we treated rats with egonol gentio- trioside and detected their serum estrogen levels. OVX rats were used to exclude the variability in the serum estrogen level due to the estrous cycle. Nilestriol was reported to increase the serum estrogen level in OVX rats (Xu et al., 2002), thus it was used here as a positive control. Egonol gentiotrioside was delivered to the rats by intramuscular injection or intragastric administration to examine the effect of intestinal on its estrogen biosynthesis-promoting function. Only one dose of a high con- centration of egonol gentiotrioside was examined owing to the limited availability of purified egonol gentiotrioside. The rats were continuously treated with egonol gentiotrioside for 30 days and their serum 17b-estradiol levels were examined. As shown in Fig. 5,theserum17b-estradiol level in OVX rats was significantly less than that in sham-operated rats, indicating that removal of the ovaries significantly reduced the circulating estrogen

Fig. 4. Egonol gentiobioside and egonol gentiotrioside promote estrogen biosynthesis level. Treatment with egonol gentiotrioside at 30 mg/kg/day via through aromatase. KGN cells were pretreated with EGB (10 mM) and EGT (10 mM) intragastric administration slightly increased the serum 17b-estra- for 24 h. Subsequently, the cells were supplemented with 10 nM testosterone for diol level compared with that in OVX rats, but this apparent increase another 24 h. The cell lysates were immunoblotted with an anti-aromatase was not statistically significant. However, treatment with egonol antibody (A). The total cellular RNA was extracted and used to detect the gentiotrioside at 15 mg/kg/day via intramuscular injection signifi- expression of aromatase mRNA by real-time qPCR. The results were presented as the induction (n-fold) relative to the basal levels in untreated cells. GADPH was cantly increased the serum 17b-estradiol level, indicating that used as an internal control (B). An in vitro aromatase assay using recombinant egonol gentiotrioside can indeed promote estrogen biosynthesis expressed aromatase and dibenzylfluorescein as a substrate was conducted as described in the Materials and methods. For the formestane treatment, 1 mM formestane was added to the recombinant aromatase for 30 min, and then mixed with 1 mM of an EGB- or EGT-treated NADPH regenerating system (C). Cont, DMSO. FSK, 10 mM forskolin. FOR, 10 mM formestane. EGB, egonol gentiobioside. EGT, egonol gentiotrioside. þFOR, 1 mM formestane plus 1 mM EGB or EGT. ***Po0.001 compared with the control (n¼3). gentiobioside and egonol gentiotrioside had no obvious effects on aromatase protein expression. Quantitative real-time RT-PCR was also used to evaluate the effects of egonol gentiobioside and egonol gentiotrioside on aromatase gene transcription, and the results of this showed that the two compounds did not increase the aromatase mRNA level (Fig. 4B). These results suggest that egonol gentiobioside and egonol gentiotrioside promote estrogen biosynthesis without modulating aromatase gene transcription or translation. Based on this finding and the observation that egonol gentiobioside and egonol gentiotrioside also promoted estrogen biosynthesis in 3T3-L1 preadipocyte cells, where aromatase expression is controlled by different promoters and cytokines, it appears possible that egonol gentiobioside and egonol gentiotrio- Fig. 5. Egonol gentiotrioside promotes estrogen biosynthesis in vivo. Ovariectomized side directly bind to aromatase and alter its activity. To examine rats were treated with EGT for 30 days and the 17b-estradiol concentrations in their serum were quantified using a magnetic particle-based 17b-estradiol ELISA. this, an in vitro recombinant aromatase assay was conducted. OVX, ovariectomized rats treated with vehicle. Sham, sham-operated rats treated As shown in Fig. 4C, the aromatase inhibitor formestane signifi- with vehicle. EGT, egonol gentiotrioside. *Po0.05, **Po0.01, ***Po0.001 com- cantly inhibited dibenzylfluorescein hydrolysis. At low concentrations pared with the control (n¼10). 280 D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282 in vivo. The dose of egonol gentiotrioside required for increasing the in vitro and in vivo through an increase in the mRNA expression of serum 17b-estradiol level when administered by intragastric admin- the aromatase gene in a cAMP-dependent or -independent man- istration was higher than that required when administered by ner, none of these compounds could promote aromatase activity intramuscular injection, indicating that possible hydrolysis of egonol by directly binding to and modulating the activity of the aroma- gentiotrioside may occur during intestinal absorption and this will tase enzyme as observed for the aromatase inhibitors (Sanderson significantly affect its pharmacological activity. et al., 2000; You et al., 2001; Morinaga et al., 2004). Thus, egonol gentiobioside and egonol gentiotrioside are the first compounds identified thus far that promote aromatase activity by directly 4. Discussion modulating aromatase enzyme structure, making them suitable as pharmaceutical tools for studying conformational changes that Currently, aromatase activity is determined by measuring the occur in aromatase during the androgen conversion process. amount of 3H-water released upon the enzymatic conversion of Although aromatase has been extensively studied for more than radiolabeled androgens in a cell-free assay using human placental 35 years, the mechanism by which it converts androgens to estro- microsomes or recombinant human aromatase protein (Njar et al., gens remains poorly understood. Recently, the crystal structure of 1995; Vinggaard et al., 2000) or in a cell-based assay using aromatase derived from human placenta was successfully elucidated mammalian cell lines such as the human JEG-3 and JAr cell lines andshedsomelightonthisprocess(Ghosh et al., 2009). However, (Drenth et al., 1998; Brueggemeier et al., 1997). However, these the proposed androgen-aromatase interaction mechanism was static, assays require the use of radioactive materials and the detection and it is still unknown whether any conformational changes occur procedures are time- and labor-consuming, which means they are during estrogen biosynthesis. The aromatase modulators found to not suitable for high-throughput screening. An alternative fluor- date are inhibitors of the conversion of androgens, but no small escent cell-free method has been developed using recombinant molecules have ever been reported that instead promote the con- human aromatase protein and dibenzylfluorescein, but it cannot version of androgens. In this study, we found that egonol gentiobio- detect aromatase induction because it utilizes a cell-free system side and egonol gentiotrioside at concentrations of 0.1–1 mMcould (Stresser et al., 2000). significantly promote the substrate conversion catalyzed by purified To obviate the limitations caused by the utilization of cell-free aromatase in vitro, indicating that the binding of these compounds systems and radiolabeled androgen substrates, a method for detect- may modulate the protein conformation of aromatase and enhance ing aromatase activity in the human ovarian granulosa-like tumor its substrate conversion efficiency. This positive allosteric effect has cell line KGN was developed using an ELISA (Ohno et al., been widely found in ligand-binding receptors and enzyme catalysis 2004). However, conventional polystyrene-based ELISA methods are (Changeux and Edelstein, 2005). It has been reported that aromatase often limited by problems with their sensitivity, reproducibility, and activity is enhanced by posttranslational modification through the stability. Magnetic particle-based ELISAs perform much better in mitogen-activated protein kinase or phosphoinositide 3-kinase path- these respects than conventional ELISAs (Lawruk et al., 1996). In this way (Miller et al., 2008; Su et al., 2011). Thus, whether egonol study, we established a method using a 17b-estradiol magnetic gentiobioside and egonol gentiotrioside may induce a conformational particle-based ELISA to detect estrogen biosynthesis in KGN cells, change of aromatase to enhance its activity as the posttranslational which showed a better sensitivity and precision than the conven- modification does needs to be further studied. Further elucidation of tional polystyrene-based ELISA (Fig S4), and a low cross-reactivity the specific sites at which egonol gentiobioside and egonol gentio- with the examined compounds, except 17b-estradiol (Table S1). trioside interact with aromatase will reveal a new catalytic mechan- A similar 17b-estradiol magnetic particle-based ELISA method ism of estrogen biosynthesis. correlated well with a commercially available radioimmunoassay We also found that egonol gentiotrioside could promote and was successfully applied to detect 17b-estradiol in environ- estrogen biosynthesis in OVX rats, an effect that may possibly mental water (Zhao and Lin, 2005; Xin et al., 2009). This method will act through intracrine tissues, such as adipose tissue (Luu-The greatly facilitate the discovery of chemical compounds that induce and Labrie, 2010). The effect of egonol gentiotrioside when or inhibit aromatase expression or activity in a tissue-specific administered by intramuscular injection was more prominent manner in a high-throughput format. than that when administered by intragastric administration, Using this estrogen detection method, we found that egonol indicating that the oligosaccharide chain in egonol gentiotrioside gentiobioside and egonol gentiotrioside isolated from S. perkinsiae could be partially hydrolyzed during gastrointestinal absorption. significantly promoted estrogen biosynthesis in both granulosa- This further indicates the important role of the oligosaccharide like KGN cells and preadipocyte 3T3-L1 cells. In humans, the chain of egonol gentiotrioside in promoting aromatase activity. biosynthesis of estrogens occurs at a number of different sites. We found the potency of the egonol compounds in terms of The major sites are the granulosa cells of the ovary in premeno- promoting estrogen biosynthesis to be as follows: egonol gentio- pausal women and the stromal cells of the adipose tissue in trioside4egonol gentiobioside4egonol glucoside4egonol (Fig S5), postmenopausal women (Simpson, 2000). Aromatase is the only thus, the oligosaccharide chain of egonol is indispensable for its enzyme that converts testosterone to 17b-estradiol in both types estrogen biosynthesis-promoting effect. of cells, but it is transcriptionally regulated by different cytokines Oligosaccharide-protein interactions play many important and signaling pathways in each cell type (Michael et al., 1995; biological roles in mediating cellular signaling transduction and Zhao et al., 1995; Shozu and Simpson, 1998). Egonol gentiobioside protein functions (Varki, 1993). It is still unknown whether and egonol gentiotrioside promoted estrogen biosynthesis in both aromatase is glycosylated and how oligosaccharides regulate its types of cell, indicating that the compounds may directly regulate function. Our study provides new clues to address these ques- the activity of the aromatase protein. This possibility was further tions, and further studies of the structure-activity relations of demonstrated by our observation that the compounds had no egonol compounds would help discover new mechanisms of effect on aromatase gene transcription and translation in KGN aromatase activity regulation. Such work may also enable us to cells, but directly promoted aromatase activity in vitro. Although prevent the oligosaccharide chain of egonol gentiotrioside from aromatase activity was previously found to be induced by being hydrolyzed during its gastrointestinal absorption, which 2-chloro-s-triazine herbicides (, simazine, and propazine) would be key for preclinical studies. The serum 17b-estradiol and benzimidazole fungicides (benomyl and its metalolitecarben- level in the OVX rats was still relatively high. This was probably dazim) in vitro,orp,p0-dichlorodiphenyldichloroethane (p,p0-DDE) the result of local estrogen biosynthesis in peripheral intracrine D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282 281 tissues, such as the liver, bone, and adipose tissue, using andro- Changeux, J.P., Edelstein, S.J., 2005. Allosteric mechanisms of signal transduction. gens secreted from the extraglandular tissues, such as kidney, Science 308, 1424–1428. Dalla Valle, L., Vianello, S., Belvedere, P., Colombo, L., 2002. Rat cytochrome stomach and duodenum, where CYP17A1 is expressed to mediate P450c17 gene transcription is initiated at different start sites in extraglandular glucocorticoids and androgen biosynthesis (Dalla Valle et al., and glandular tissues. J. Steroid. Biochem. Mol. Biol. 82, 377–384. 2002) Davison, S., Davis, S.R., 2003. Hormone replacement therapy: current controver- Estrogen therapy is an established regimen for the prevention sies. Clin. Endocrinol. 58, 249–261. Deroo, B.J., Korach, K.S., 2006. Estrogen receptors and human disease. J. Clin. and treatment of estrogen insufficiency-related diseases such as Invest. 116, 561–570. osteoporosis; however, its side effects limit its clinical use and Drenth, H.J., Bouwman, C.A., Seinen, W., Van den Berg, M., 1998. Effects of some efficacy. Thus, new estrogen biosynthesis modulators, which can act persistent halogenated environmental contaminants on aromatase (CYP19) activity in the human choriocarcinoma cell line JEG-3. Toxicol. Appl. Pharma- locally through specific target tissues, would be preferred. In this col. 148, 50–55. study, we found that egonol gentiobioside and egonol gentiotrioside Ghosh, D., Griswold, J., Erman, M., Pangborn, W., 2009. Structural basis for could promote estrogen biosynthesis directly through aromatase androgen specificity and oestrogen synthesis in human aromatase. Nature 457, 219–223. both in vitro and in vivo, providing a new way to obviate the side Halbreich, U., 1997. Role of estrogen in postmenopausal depression. Neurology 48, effects of estrogen by locally promoting estrogen biosynthesis S16–S19. through tissue-specific drug absorption or delivery, especially useful Heldring, N., Pike, A., Andersson, S., Matthews, J., Cheng, G., Hartman, J., Tujague, M., Strom,¨ A., Treuter, E., Warner, M., Gustafsson, J.A., 2007. Estrogen for postmenopausal women. Estrogen replacement therapy has receptors: how do they signal and what are their targets. Physiol. Rev. 87, shown efficacy to improve mood and cognition in menopausal and 905–931. postmenopausal women and decrease vulnerability to depression Johnston, S.R., Dowsett, M., 2003. Aromatase inhibitors for breast cancer: lessons (Halbreich, 1997; Schmidt et al., 2000). Interestingly, a resinous from the laboratory. Nat. Rev. Cancer 3, 821–831. Lawruk, T.S., Gueco, A.M., Milhaliak, C.A., Dolder, S.C., Dial, lG.E., Herzog, D.P., material (An-Xi-Xiang) secreted by the genus Styrax has been widely Rubio, F.M., 1996. Development of a magnetic particle-based enzyme immu- used in clinical practice to treat depression and other psychological noassay for the quantification of chlorpyrifos in water. J. Agric. Food Chem. 44, disorders in China for thousands of years. Our study here suggests 2913–2918. Li, Q.L., Li, B.G., Qi, H.Y., Gao, X.P., Zhang, G.L., 2005. Four new benzofurans from that the clinical benefit of this medicinal plant in improving seeds of Styrax perkinsiae. Planta Med. 71, 847–851. psychological disorders, such as depression, may be caused by its Luo, Y., He, Z., Li, H., 2007. 2-Aryl benzofurans and their derivatives from seeds of estrogen biosynthesis-promoting effect and that egonol gentiobio- Styrax macranthus. Fitoterapia 78, 211–214. Luu-The, V., Labrie, F., 2010. The intracrine sex steroid biosynthesis pathways. side and egonol gentiotrioside might be the chemical constituents Prog. Brain Res. 181, 177–192. responsible for this effect. Maiti, A., Cuendet, M., Croy, V.L., Endringer, D.C., Pezzuto, J.M., Cushman, M., 2007. In conclusion, we established a sensitive estrogen biosynthesis Synthesis and biological evaluation of (þ/-)-abyssinone II and its analogues as aromatase inhibitors for chemoprevention of breast cancer. J. Med. Chem. 50, evaluation assay based on human ovarian granulosa KGN cells 2799–2806. and identified two chemical compounds, egonol gentiobioside McGovern, S.L., Caselli, E., Grigorieff, N., Shoichet, B.K., 2002. A common mechan- and egonol gentiotrioside from S. perkinsiae, that potently pro- ism underlying promiscuous inhibitors from virtual and high-throughput moted estrogen biosynthesis in ovarian granulosa KGN cells and screening. J. Med. Chem. 45, 1712–1722. Michael, M.D., Kilgore, M.W., Morokashi, K., Simpson, E.R., 1995. Ad4BB/SF1 mouse 3T3-L1 preadipocyte cells. These compounds had no effect regulates cyclic AMP-induced transcription from the proximal promoter (PII) on the transcriptional or translational expression of aromatase in of the human aromatase P450 (CYP19) gene in the ovary. J. Biol. Chem. 270, KGN cells, but they significantly increased the catalytic activity of 13561–13566. Miller, T.W., Shin, I., Kagawa, N., Evans, D.B., Waterman, M.R., Arteaga, C.L., 2008. in vitro expressed recombinant aromatase, indicating that aroma- Aromatase is phosphorylated in situ at serine-118. J. Steroid Biochem. Mol. tase activity may be allosterically regulated by these compounds. Biol. 112, 95–101. Furthermore, egonol gentiobioside significantly promoted estro- Morinaga, H., Yanase, T., Nomura, M., Okabe, T., Goto, K., Harada, N., Nawata, H., 2004. A benzimidazole fungicide, benomyl, and its metabolite, carbendazim, gen biosynthesis in OVX rats in vivo. Egonol gentiobioside and induce aromatase activity in a human ovarian granulose-like tumor cell line egonol gentiotrioside are valuable for structural modification and (KGN). Endocrinology 145, 1860–1869. warrant further investigation for their potential as pharmaceu- Nishi, Y., Yanase, T., Mu, Y., Oba, K., Ichino, I., Saito, M., Nomura, M., Mukasa, C., Okabe, T., Goto, K., Takayanagi, R., Kashimura, Y., Haji, M., Nawata, H., 2001. tical tools for estrogen-deficiency induced diseases. Establishment and characterization of a steroidogenic human granulosa-like tumor cell line, KGN, that expresses functional follicle-stimulating hormone receptor. Endocrinology 142, 437–445. Acknowledgments Njar, V.C., Grun,¨ G., Hartmann, R.W., 1995. Evaluation of 6,7-aziridinyl steroids and related compounds as inhibitors of aromatase (P-450arom). J. Enzyme Inhib. 9, 195–202. This work was supported by the National Natural Science Ohno, K., Araki, N., Yanase, T., Nawata, H., Iida, M., 2004. A novel nonradioactive Foundation of China (Grants nos. 20932007, 30572221, and method for measuring aromatase activity using a human ovarian granulosa- like tumor cell line and an estrone ELISA. Toxicol. Sci. 82, 443–450. 30900769), the West Light Foundation of Chinese Academy of Park, S.Y., Lee, H.J., Lee, O.K., Kang, H.Y., Choi, D.H., Paik, K.H., Khan, M., 2007. Sciences, and the National New Drug Innovation Major Project of Benzofurans from the seeds of Styrax obassia. Bull. Korean Chem. Soc. 28, China (2011ZX09307-002-02). 1874–1876. Sanderson, J.T., Seinen, W., Giesy, J.P., van den Berg, M., 2000. 2-Chloro-s-triazine herbicides induce aromatase (CYP19) activity in H295R human adrenocortical carcinoma cells: a novel mechanism for estrogenicity? Toxicol. Sci. 54, Appendix A. Supporting information 121–127. Schmidt, P.J., Nieman, L., Danaceau, M.A., Tobin, M.B., Roca, C.A., Murphy, J.H., Rubinow, D.R., 2000. Estrogen replacement in perimenopause-related depres- Supplementary data associated with this article can be found sion: a preliminary report. Am J. Obstet. Gynecol. 183, 414–420. in the online version at http://dx.doi.org/10.1016/j.ejphar.2012. Shozu, M., Simpson, E.R., 1998. Aromatase expression of human osteoblast-like 07.005. cells. Mol. Cell. Endocrinol. 139, 117–129. Simpson, E.R., Zhao, Y., Agarwal, V.R., Michael, M.D., Bulun, S.E., Hinshelwood, M.M., Graham-Lorence, S., Sun, T., Fisher, C.R., Qin, K., Mendelson, C.R., 1997. Aromatase expression in health and disease. Recent Prog. Horm. Res. 52, References 185–213. Simpson, E.R., 2000. Role of aromatase in sex steroid action. J. Mol. Endocrinol. 25, 149–156. Akgul, Y.Y., Anil, H., 2003. Benzofurans and another constituent from seeds of Simpson, E.R., Clyne, C., Rubin, G., Boon, W.C., Robertson, K., Britt, K., Speed, C., Styrax officinalis. Phytochemistry 63, 939–943. Jones, M., 2002. Aromatase—a brief overview. Annu. Rev. Physiol. 64, Brueggemeier, R.W., Gilbert, N.E., Gu, X., O’Reilly, J.M., Lovely, C.J., 1997. Aromatase 93–127. inhibition in JAr choriocarcinoma cells by 7a-arylaliphatic androgens. Simpson, E.R., 2004. Aromatase: biologic relevance of tissue-specific expression. J. Steroid Biochem. Mol. Biol. 61, 73–77. Semin. Reprod. Med. 22, 11–23. 282 D. Lu et al. / European Journal of Pharmacology 691 (2012) 275–282

Stresser, D.M., Turner, S.D., McNamara, J., Stocker, P., Miller, V.P., Crespi, C.L., Xin, T.B., Wang, X., Jin, H., Liang, S.X., Lin, J.M., Li, Z.J., 2009. Development of Patten, C.J., 2000. A high-throughput screen to identify inhibitors of aromatase magnetic particle-based chemiluminescence enzyme immunoassay for the (CYP19). Anal. Biochem. 284, 427–430. detection of 17b-estradiol in environmental water. Appl. Biochem. Biotechnol. Su, B., Wong, C., Hong, Y., Chen, S., 2011. Growth factor signaling enhances 158, 582–594. aromatase activity of breast cancer cells via post-transcriptional mechanisms. Xu, X., Gong, Q.Y., Lu, Y.Q., Yao, M.H., 2002. Effects of soy isoflavones on bone J. Steroid Biochem. Mol. Biol. 123, 101–108. density and estrogen activity in ovariectomized rats. Chin. J. New Drugs Clin. Varki, A., 1993. Biological roles of oligosaccharides: all of the theories are correct. Rem. 21, 321–325. Glycobiology 3, 97–130. You, L., Sar, M., Bartolucci, E., Ploch, S., Whitt, M., 2001. Induction of hepatic Vinggaard, A.M., Hnida, C., Breinholt, V., Larsen, J.C., 2000. Screening of selected aromatase by p,p0-DDE in adult male rats. Mol. Cell. Endocrinol. 178, for inhibition of CYP19 aromatase activity in vitro. Toxicol. In Vitro 207–214. 14, 227–234. Zhao, Y., Nichols, J.E., Bulun, S.E., Mendelson, C.R., Simpson, E.R., 1995. Aromatase Watanabe, M., Nakajin, S., 2004. Forskolin up-regulates aromatase (CYP19) activity P450 gene expression in human adipose tissue: role of a Jak/STAT pathway and gene transcripts in the human adrenocortical carcinoma cell line H295R. in regulation of the adipose-specific promoter. J. Biol. Chem. 270, J. Endocrinol. 180, 125–133. 16449–16457. Wiseman, L.R., Goa, K.L., 1996. Formestane: a review of its pharmacological Zhao, L., Lin, J.M., 2005. Development of a micro-plate magnetic chemilumines- properties and clinical efficacy in the treatment of postmenopausal breast cence enzyme immunoassay (MMCLEIA) for rapid- and high-throughput cancer. Drugs Aging 9, 292–306. analysis of 17b-estradiol in water samples. J. Biotechnol. 118, 177–186.