Inhibition of CYP17A1Activity by Resveratrol,Piceatannol, and Synthetic Resveratrol Analogs
Agneta Oskarsson,1* Carmela Spatafora,2 Corrado Tringali,2 and Åsa Ohlsson Andersson1
“Department of Biomedical Sciences and Veterinary Public Health, Swedish University of Agricultural Sciences, Uppsala, Sweden
2 Dipartimento di Scienze Chimiche, University of Catania, Catania, Italy
BACKGROUND. Resveratrol (RSV) and resveratrol analogs have a potential use in prostate cancer chemoprevention due to effects on for example, cell growth, apoptosis, angiogenesis, and metastasis. However, inhibition of CYP17A1, a key enzyme in the androgen biosynthesis and a target for prostate cancer therapy, has not been explored as a possible mechanism behind the effects on prostate cancer.
METHODS. Human adrenocortical carcinoma cells, H295R, were treated with RSV, piceatan- nol (PIC), 3,5,4’-triacetylresveratrol (RSVTA), 3,5-diacetylresveratrol (RSVDA), and 3,5,4’- trimethylresveratrol (RSVTM) for 24 hr at concentrations of 1, 5, 10, 25, and 50 p.M. Steroid secretion, enzyme activities, and gene expression of key steps in steroidogenesis were investigated.
RESULTS. Secretion of dihydroepiandrosterone (DHEA), testosterone, and cortisol were drastically decreased by all test compounds at concentrations that did not affect cell viability. Progesterone and aldosterone secretion were increased. This steroid secretion pattern can be explained by the demonstrated inhibition of CYP17A1 enzyme activity. The most efficient CYP17A1 inhibitors were the synthetic analogs RSVTA, RSVDA, and RSVTM. Inhibition by RSVTM was more selective on the 17,20-lyase activity than hydroxylase activity of CYP17A1. Treatment of cells with all compounds, except RSVTM, caused increased estradiol levels, which could be explained by the demonstrated inhibition of estrogen sulfate conjugation, catalyzed by SULT1E1.
CONCLUSIONS. Our results on CYP17A1 inhibition of RSV and RSV analogs suggest a novel mechanism for chemoprevention of prostate cancer by resveratrol and the analogs. Especially RSVTM, which has a preferential inhibition on the 17,20-lyase activity of CYP17A1, may be a promising candidate for prostate cancer chemoprevention. Prostate 74:839-851, 2014. C 2014 Wiley Periodicals, Inc.
KEY WORDS: 17,20-lyase; H295R; prostate cancer; chemoprevention; steroidogenesis
INTRODUCTION
Resveratrol (RSV, trans-3,5,4’-trihydroxystilbene) is a phytoalexin, produced in plants as a response to
fungal and other microbial attacks [1]. It is produced by various plants species, especially grapevines, pines, and legumes, and also present in nuts and berries [2].
*Correspondence to: Prof. Agneta Oskarsson, PhD, Department of Biomedical Sciences and Veterinary Public Health, Swedish Univer- sity of Agricultural Sciences, Box 7028, SE-750 07 Uppsala, Sweden. E-mail: agneta.oskarsson@slu.se
Received 17 January 2014; Accepted 13 February 2014 DOI 10.1002/pros.22801
Published online 9 March 2014 in Wiley Online Library (wileyonlinelibrary.com).
Abbreviations: RSV, resveratrol; PIC, piceatannol; RSVTA, 3,5,4’- triacetylresveratrol; RSVDA, 3,5-diacetylresveratrol; RSVTM, 3,5,4’- trimethylresveratrol; DHEA, dehydroepiandrosterone; DMSO, di- methyl sulfoxid.
Grant sponsor: The Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning; Grant number: 210/ 2004-0601; Grant sponsor: Swedish Fund for Research without Animal Experiments; Grant number: 38/10; Grant sponsor: MIUR, Ministero dell’Università e della Ricerca (PRIN 2009, Rome, Italy).
Conflict of interest: none.
Since RSV is present in the grape skin and white wines are not fermented with the skin, only red wines contain appreciable amounts of resveratrol [2]. RSV has antioxidant, anti-inflammatory, and anticancer properties. Numerous scientific publications have sug- gested health-promoting effects of resveratrol on the cardiovascular system, increased lifespan and dimin- ished neurodegeneration [2-4]. Piceatannol (PIC, trans- 3,5,3’,4’-tetrahydroxystilbene) is a hydroxylated res- veratrol analog, present in plants [5]. PIC has been identified as a metabolite of RSV through the activity of phase I enzymes, such as CYP1B1 [6,7], indicating that RSV may act as a pro-drug to PIC in vivo. PIC has a similar broad spectrum of biological activities as RSV but is much less studied than RSV.
RSV has gained considerable attention because its potential use in chemoprevention of various cancer forms, mediated via effects on for example, cell growth, apoptosis, angiogenesis, and metastasis [8]. Anti-proliferative effects resulting in suppression and inhibition of tumor formation by RSV and RSV analogs were recently demonstrated in a model of prostate cancer in mice [9]. In addition, RSV has been sug- gested to modulate the function of the androgen receptor in prostate cancer cells [10] and to display an inhibitory effect on synthesis of androgen steroids in
rat Leydig cells [11] and in rat ovarian theca interstitial cells [12].
Cytochrome P450 17a-hydroxylase/17,20-lyase (CYP17A1) is a key enzyme in the synthesis of steroid hormones in the adrenals, testes, placenta, and ovaries and also a key target in the treatment of castrate- resistant prostate cancer [13,14] (Fig. 1). CYP17A1 catalyses the two central reactions involved in the production of sex steroids: the conversion of pregnen- olone to 17a-hydroxypregnenolone, catalyzed by the hydroxylase activity of CYP17A1, and the subsequent synthesis of dehydroepiandrosterone (DHEA), cata- lyzed by the 17,20-lyase activity of CYP17A1. DHEA is converted by 3ß-hydroxysteroid dehydrogenase type II (HSD3B2) to androstenedione and further by 17ß- hydroxysteroid dehydrogenase (HSD17B1) to testos- terone, which in turn is converted to the more potent dihydrotestosterone.
Prostate cancer growth and progression is depen- dent on androgens. Deprivation of androgen produc- tion is the primary treatment of prostate cancer, usually accomplished by chemical or surgical castra- tion. However, castration will not affect adrenal and intratumor androgen production. Thus, adrenal secre- tion of androgens and conversion of adrenal andro- gens into testosterone in the prostate seems to play a
Cholesterol
StAR
DHEA-S
CYP11A1
SULT2A1
Pregnenolone
CYP17A1ª
17a-OH-
CYP17A1b
Pregnenolone
DHEA
HSD3B2
HSD3B2
HSD3B2
CYP17A1ª
17a-OH-
CYP17A1b
HSD17B
Progesterone
Progesterone
Androstenedione
Testosterone
CYP21A2
CYP21A2
CYP19A1
CYP19A1
11-Deoxycorticosterone
11-Deoxycortisol
Estrone
HSD17B
Estradiol
CYP11B1/ CYP11B2
CYP11B1
SULT1E1
SULT1E1
Corticosterone
Cortisol
Estrone-S
Estradiol-S
CYP11B2
Aldosterone
The Prostate
major role in the development of castrate-resistant prostate cancer [8,14]. Inhibition of CYP17A1 activity blocks androgen production from all sources and therefore CYP17A1 inhibitors are validated as effective therapeutic targets in the treatment of prostate cancer. CYP17A1 inhibitors in clinical use and clinical trials are ketoconazole, abiraterone, and orteronel with chemical structures similar to the CYP17A1 substrates pregnenolone and progesterone [13,14]. However, CYP17A1 is also important in the glucocorticoid production and thus inhibition of the hydroxylase activity of CYP17A1 will result in suppressed cortisol secretion, release of ACTH via negative feedback and increased levels of aldosterone, leading to side effects, such as hypertension, hypokalemia and fluid reten- tion [14].
RSV has a low bioavailability and the beneficial effects in humans are described as mostly tiny and remain controversial [4]. Despite a rapid and high absorption (at least 70%) the oral bioavailability is only less than 1%, due to its rapid phase II conjugation to sulfates and glucuronides [15,16]. Like RSV, PIC is rapidly glucuronidated in the liver and has a low bioavailability [17]. To improve the pharmacokinetic characteristics and possibly enhance the beneficial effects of RSV, chemical analogs have been synthe- sized [11,18-20] (Fig. 2). The acetylated analog of resveratrol, RSVTA, has been identified as a prodrug of resveratrol, and has improved pharmacokinetic properties compared to RSV with longer half-life, increased AUC and volume of distribution [21]. The substitution of hydroxy groups with methoxy groups in the RSV structure at the 3,5,4’ positions in RSVTM has been shown to increase the oral bioavailabili- ty [19,22] and the serum levels after repeated oral gavage [9] compared to RSV. RSVTM has been demonstrated to reduce cell growth in human cancer cell lines [23] and was the most active in inhibiting cell proliferation in DU-145 prostate cancer cells [24] and in LNCaP prostate cancer cells [9] compared to RSV and other resveratrol analogs. RSVTM and PIC are suggested to be potential chemopreventive agents for prostate cancer due to their potent anti-proliferative, anti-inflammatory and anti-clonogenic properties [9].
We have investigated the effects of RSV and RSV analogs on steroid hormone production with special focus on the efficiency to inhibit CYP17A1 enzyme activity and suppress androgen secretion. In order to investigate the effects on androgen biosynthesis as well as the glucocorticoid and mineralocorticoid ste- roidogenesis pathways we used the human adrenocor- tical cell line H295R. The cell line is derived from a primary human adrenocortical carcinoma and the cells express all the enzymes for the adrenocortical hor- mone production including cortisol, aldosterone, and
Resveratrol (RSV)
OH
HO
OH
3,5,4’-triacetylresveratrol (RSVTA)
OCOCH3
H3COCO
OCOCH3
3,5-diacetylresveratrol (RSVDA)
OH
H3 COCO
OCOCH3
3,5,4’-trimethylresveratrol (RSVTM)
OCH3
H3 CO
OCH3
Piceatannol (PIC)
OH
HO
OH
OH
the androgens dehydroepiandrosterone (DHEA) and testosterone [25,26] (Fig. 1). In addition, the cells have CYP19A1 activity and synthesize estradiol. The cell line is a useful in vitro model for studies both on basal
regulation of steroidogenesis and on disruption of steroidogenesis by chemicals [27,28]. The H295R steroidogenesis assay is a validated OECD test for chemical effects on the production of testosterone and estradiol [29]. In the current study we compare the concentration-dependent effects of RSV, PIC, RSVTA, RSVDA, and RSVTM on the steroidogenic pathways in the adrenocortical cells and demonstrate for the first time specific CYP17A1 inhibiting effects of these compounds, which may be of relevance for chemo- prevention of prostate cancer.
MATERIALS AND METHODS
Cell Culture
NCI-H295R cells (ATCC, Manassas, VA) were cultured as a monolayer in DMEM:F-12 medium with GlutaMAX™M I (Invitrogen, Carlsbad, CA), sup- plemented with 1% ITS+ Premix (BD Biosciences, Bedford, MA) and 1% penicillin-streptomycin (Lonza, Basel, Switzerland). As a serum substitute, 2.5% Nu- Serum I (BD Biosciences) was added to the cell culture medium. Cells were cultured in a humidified environ- ment at 37℃ containing 95% air and 5% CO2. Medium was changed two to three times a week and Tryple Express (Invitrogen) was used for sub-culturing of the cells.
Test Substances
Resveratrol (RSV) was purchased from Sigma- Aldrich (R5010, St Louis, MO) and piceatannol (PIC) was purchased from Tocris biosciences (1554, Park Ellisville, MO). Resveratrol analogs were synthesized according to previously reported methods; namely 3,5,4’-triacetylresveratrol (RSVTA) was obtained by acetylation of resveratrol, 3,5-diacetylresveratrol (RSVDA) was obtained by lipase-catalyzed regioselec- tive deacetylation of RSVTA [30], whereas 3,5,4’- trimethylresveratrol (RSVTM) was synthesized, as previously reported [23], based on an Arbuzov rear- rangement followed by the Horner-Emmons-Wads- worth reaction. Pregnenolone and forskolin was purchased from Sigma (P9129 and F3917), testosterone and genistein from Fluka (86500 and 91955, Deisenho- fen, Germany), formestane from Sigma-Aldrich (F2552), and trilostane from Sanofi (Newcastle, UK). All test compounds were dissolved in dimethyl sulfox- ide (DMSO) (D8418, Sigma).
Cell Viability Test
As a cell viability test, the capacity of the cells to reduce a tetrazolium compound (MTS) to the colored
product formazan was analyzed. The reaction is dependent on mitochondrial function and the amount of formazan formed is directly proportional to the number of living cells in culture. A total of 1.7 x 104 cells/well were seeded in 96-well plates in a volume of 100 ul serum substitute-supplemented medium. After 72 hr, medium was removed and replaced with serum-free medium containing test chemicals dis- solved in DMSO. The final concentration of DMSO in culture medium was 0.1%. Cells were treated with 0, 1, 5, 10, 25, and 50 u.M of RSV, the resveratrol analogs RSVTA, RSVDA, RSVTM and PIC. Staurosporin (Sigma) was used as a negative control of cell proliferation. Cell proliferation was analyzed after 24 hr of treatment by addition of 20 ul CellTiter 96® AQueos One Solution Reagent (Promega Corporation, Madison, WI) to each well. The plate was incubated at 37℃ for 1 hr, followed by measurement of absor- bance at 490 nm with Wallac Victor2 1420 microplate reader (PerkinElmer, Waltham, MA). Relative effects on cell viability compared to vehicle control were determined from the mean absorbance value based on five replicates.
Treatment of H295R Cells
Cells of passage 12-18 were seeded in serum substitute-supplemented medium and incubated for 72 hr. Cells in 12-well plates were seeded at a density of 2 × 105 cells/well, in a volume of 1.0ml medium. Cells seeded in 6-well plates for gene expression analysis were seeded at a density of 5 x 105 cells/well, in a volume of 2.5 ml medium. Medium was with- drawn after 72 hr and replaced with serum substitute- free medium supplemented with test chemicals or vehicle control for 24 hr. Cells were treated with 1, 5, 10, 25 and 50 MM (n=3) of RSV and RSV analogs. Following treatment, cell culture medium was removed and stored at -20℃ for subsequent hormone analysis. The experiments were repeated once. Cells from the 6-well plates, with 25 uM as the highest concentration, were used for RNA isolation and gene expression analysis.
Determination of Hormone Levels in Cell Culture Medium
Cortisol and testosterone levels in culture medium were measured by ELISA (#402710 and #402510 Neo- Gen Corporation, Lansing, MI, USA) according to manufacturer’s instructions. The samples were diluted in the provided kit buffer. Absorbance was measured at 620 nm and hormone concentrations were calculated based on standard curves. The ELISA assays covered a concentration range of 0.04-10 ng cortisol/ml and
The Prostate
0.02-0.2 ng testosterone/ml. Aldosterone, progester- one, DHEA and estradiol levels were measured by ELISA (DE4128, DE1561, DE3415 and DE4399, Deme- ditec Diagnostics GmbH, Kiel-Wellsee, Germany). Samples for aldosterone analysis were treated as urine samples and diluted in urine dilution buffer, according to manufacturer’s recommendations. Samples for pro- gesterone, DHEA and estradiol analysis were diluted in steroid free serum. Absorbance was measured with a 450 nm filter and concentrations were calculated based on standard curves. The assay range for the kits were 0.015-1 ng aldosterone/ml, 0.3-40 ng progester- one/ml, 0.37-30 ng DHEA/ml and 0.003-0.2 ng estra- diol/ml.
Enzyme Activity of CYP17A1and CYP19A1
Cells were cultured as above and seeded in 12-well plates at a density of 2 x 105 cells/well. Cells were incubated for 72 hr before change to serum-free medi- um and addition of the test substances at concentra- tions between 1 and 25 u.M (n =3), steroid substrate and, for CYP17A1 assay, HSD3B2 enzyme inhibitor. DMSO concentration did not exceed 0.3%. The effect of the test substances on CYP17A1 (hydroxylase and lyase) activity was determined in cells co-treated with 1 µM pregnenolone as a steroid substrate. Trilostane at a concentration of 0.1 µM was used to inhibit HSD3B2 activity [31]. After 24hr treatment of the cells the production of DHEA was measured in the medium by ELISA. The effect of the test substances on CYP19A1 activity was investigated by co-treating the cells with 1 µM testosterone. The secreted level of estradiol was determined by ELISA. As positive controls of CYP19A1 activation, 10 uM genistein and 10 uM for- skoline were used. Formestane, at a concentration of 1 µM, was used as a positive control of CYP19A1 inhibition.
Enzyme Activity of SULT2A1and SULT1E1
To investigate if RSV and PIC had an effect on SULT2A1, the secretion of DHEA sulfate (DHEA-S) was determined by ELISA (DE1562, Demeditec Diag- nostics GmbH, Kiel-Wellsee, Germany) in the samples tested for CYP17A1 activity, described above. Hence, the effect of the combined activity of CYP17A1 and SULT2A1 was investigated. Samples were diluted in steroid free serum and absorbance was measured with a 450 nm filter. The kit had an assay range of 0.1-10 µg DHEA-S/ml.
The effects of test compounds on estradiol secretion may be due to effects on CYP19A1 enzyme activity or effects on the subsequent metabolism of estradiol to the sulfate conjugate. Thus, inhibition of SULT1E1
may cause accumulation of estradiol as well as estrone (Fig. 1). As there was no indication of an effect on HSD17B after treatment with test compounds (see Results and Effects on Steroid Secretion Section) and no suitable assay for analysis of estradiol sulfate was available, estrone sulfate (E1S) was instead deter- mined. The secretion of E1S was measured by a direct RIA (DSL5400, Immunotech, Prague, Czech Republic) in medium from cells treated with RSV and PIC from the CYP19A1 activity assay, described above. The RIA was performed according to the manufacturer’s recommendation. The samples were diluted with steroid free serum and radioactivity determined in a Wallac Wizard2 2470 automatic gamma counter (PerkinElmer). The assay had a detection range of 0.01-80 ng E1S/ml.
RNA Extraction and Real-Time RT-PCR Analysis
Total RNA was isolated from the cells using Nucleo- Spin® RNA II Kit (Clontech, Mountain View, CA) following manufacturer’s instructions. Quant-iT™M RiboGreen® RNA Reagent (Invitrogen) was used to determine the concentration of the RNA samples according to the high-range protocol. Verso™ cDNA Kit (Thermo Fisher Scientific, Surrey, UK) was used to synthesize cDNA based on 1 µg of RNA, and quantita- tive real-time RT-PCR reactions were set up using dual labeled probes 5’-FAM and 3’-TAMRA (TAG Copenha- gen, Copenhagen, Denmark) and reagents in a final volume of 12.5 ul, as previously described [31]. Primers and probes have previously been described [27], except for SULT1E1 and HSD17B1. The SULT1E1 primer/ probe set was CAAATCCTGGATCCTTTCCA (for- ward), CCCTTTTCCCACCAAGATTT (reverse), and TGCAAGGACAGGTTCCTTATGGTTCC (dual labeled probe). The sequences of primers for HSD17B1 were TTCATGGAGAAGGTGTTGG (forward), AAGACTT- GCTTGCTGTGG (reverse), and ACATCCACACCTT- CCACCGCT (dual labeled probe). cDNA was added to the reaction mixture at a volume of 2.5 ul cDNA, corresponding to 25 ng input RNA/reaction. The real- time RT-PCR was performed using a Rotor-Gene™ 3000 (Corbett Life Science, Sidney, Australia). Results were calculated based on a real-time RT-PCR relative quantification strategy and presented as mean relative gene expression levels, compared to the vehicle control.
Statistical Analysis of Results
Statistical analysis was performed by Kruskal- Wallis, followed by Mann-Whitney U for individual comparisons against the control. Statistical analysis was done in StatView 5.0.1 (SAS Institute, Cary, NC).
RESULTS Cell Viability
The test compounds did not affect cell viability at 1 and 5 MM and at higher concentrations there were minor but significant effects (Fig. 3). Inhibition of cell viability was observed in cells treated with RSVTM, with a maximum inhibition to 81% of control levels at 50 p.M. PIC significantly stimulated cell viability to a maximum of 129% of control levels at 25 p.M.
Effects on Steroid Secretion
All test compounds caused reduced secretion of DHEA, testosterone and cortisol, while increased secretion of progesterone and aldosterone was ob- served (Fig. 4). Increased estradiol secretion was also observed following treatment with all compounds except RSVTM, which in contrast inhibited estradiol secretion (Fig. 4). A drastic increase in progesterone secretion up to 25-fold of the control levels was observed following treatment with RSVTA, RSVDA, PIC, and RSV (Fig. 4). Least stimulatory effect on progesterone secretion, but still statistically significant, was found following treatment with RSVTM, which stimulated secretion to a maximum 1.6-fold at a concentration of 25 p.M. Control levels of progesterone in the vehicle control samples were 0.57-0.84 ng/ml.
The suppression of DHEA and testosterone secre- tion was most pronounced and almost identical, indicating that HSD3B2 and HSD17B enzyme activi- ties were not affected (Fig. 2). Suppression of andro- gens was observed following treatment with all tested compounds. The most efficient inhibitors were RSVTA and RSVDA, inhibiting secretion of DHEA and testos- terone to 16-20% of control levels at 5 u.M. Higher concentrations were needed of RSV and PIC, reaching significant inhibition at 25 and 10 MM, respectively. Control levels of DHEA in the vehicle control samples
RSV
RSVTM
ZZZZZZ
RSVTA
PIC
Relative viability (% of control)
RSVDA
140
*
*
120
*
*
*
*
*
I
100
*
1
T
T
T
*
I
T
T
Z
I
T
* T
*
80
T
60
40
20
0
1
5
10
25
50
Concentration (UM)
were 4.44-5.72 ng/ml and of testosterone 2.09- 2.83 ng/ml.
In contrast to DHEA and testosterone, estradiol secretion was significantly increased by RSVTA and RSVDA to 1.4-fold, compared to controls, at a concen- tration of 5 p.M. PIC treatment caused a significant increase of 1.4-fold at a concentration of 10 uM and a similar level was reached by RSV at 25 MM (Fig. 4). RSVTM was the only test compound inhibiting estra- diol secretion, which was statistically significant al- ready at 1 p.M. At 25 p.M RSVTM inhibited estradiol secretion to 69% of control levels. Control levels of estradiol in the vehicle control samples were 0.28- 0.36 ng/ml.
The most potent inhibitors of cortisol secretion were RSVTA and RSVDA, causing inhibition at 5 u.M to 30% and 36%, respectively, of control levels. Treat- ments with PIC and RSVTM inhibited cortisol secre- tion significantly at 10 p.M. RSV caused significant inhibition from 25 p.M. Control levels of cortisol in the vehicle control samples were 10.1-17.9 ng/ml.
Increased secretion of aldosterone, although much less pronounced than for progesterone, was observed following treatment with all test compounds, except RSVTM. PIC was the most efficient compound to stimulate aldosterone secretion, with a maximum increase of 1.7-fold at a concentration of 10 p.M. RSVDA and RSVTA reached a maximum secretion of 1.4- and 1.3-fold, respectively, at a concentration of 10 µM. Following treatment of cells with RSV aldoste- rone secretion was significantly increased only at a concentration of 25 u.M (1.5-fold). Aldosterone secre- tion returned to control levels at 50 u.M for all compounds. Control levels of aldosterone in the vehi- cle control samples were 0.32-0.39 ng/ml.
Effects on Enzyme Activities
Enzyme activity assays were performed to investi- gate if CYP17A1 and SULT2A1 were involved in the observed decreases in DHEA and testosterone secre- tion. The observed decrease in testosterone concurrent- ly with an increase in estradiol was investigated by activity assays of CYP19A1 and SULT1E1.
The activity of CYP17A1 (hydroxylase and 17,20- lyase activities) was measured by determining the secretion of DHEA. Pregnenolone was used as steroid substrate and HSD3B2 was inhibited by trilostane to block the subsequent steroidogenic pathways (Fig. 1). Treatment with RSV, RSVTA, RSVDA, and RSVTM caused a concentration-dependent decrease in DHEA secretion, indicating inhibition of CYP17A1 activity (Fig. 5A). Strongest effect was observed for the analogs RSVTA and RSVDA, which reduced DHEA secretion to 9% and 11% of control levels, respectively, at a
The Prostate
Progesterone
DHEA
3000
2500
200
RSV
RSVTM
2000
IZZZZZZ
RSVTA
PIC
1500
150
RSVDA
1000
500
100
150
100
50
50
0
0
1
5
10
25
50
1
5
10
25
50
Hormone secretion (% of control)
Aldosterone
Testosterone
200
200
150
150
100
100
50
50
0
0
1
5
10
25
50
1
5
10
25
50
Cortisol
Estradiol
200
200
150
150
100
100
50
50
0
0
1
5
10
25
50
1
5
10
25
50
Concentration (uM)
Concentration (UM)
concentration of 5 uM. No inhibitory effect on CYP17A1 activity was observed following treatment with PIC.
The lack of inhibition for PIC in the CYP17A1 assay warranted further investigation. The level of DHEA-S was determined in the CYP17A1 assay samples from the PIC treatment, to find out if DHEA accumulated due to inhibition of SULT2A1 (Fig. 1). The samples from RSV treatment were used for comparison. The secretion patterns of DHEA-S (Fig. 6A) and DHEA (Fig. 5A) were similar, indicating that there was no effect on SULT2A1 activity. Control levels of DHEA-S in the vehicle control samples ranged between 0.17 and 0.21 ng/ml.
The activity of CYP19A1 was measured by deter- mining the secretion of estradiol after addition of testosterone as a substrate. To validate the assay, two known CYP19A1 stimulators (genistein and forskolin) and one CYP19A1 inhibitor (formestane) were includ- ed in the test. The estradiol secretion was increased by 10 uM genistein and 10 p.M forskoline to 3.4- and 45-fold, respectively, compared to the control (data not
shown). Formestane (1 µM) decreased estradiol secre- tion to 13% of the control (data not shown). Increased estradiol secretion was induced following treatment with all test compounds, except RSVTM (Fig. 5B). Strongest effect was observed following treatment with RSVDA, which reached a 4.3-fold stimulation at 5 u.M. RSV was the weakest stimulator, with increased activity only at a concentration of 25 u.M, reaching a 1.9-fold stimulation compared to the control.
The increased levels of estradiol may be due to stimulated CYP19A1 activity and/or by inhibition of estradiol conjugation to sulfate, catalyzed by SULT1E1 (Fig. 1). To investigate the effect on SULT1E1 activity the secretion of E1S was determined in the CYP19A1 assay samples from the RSV and PIC treatments. Both RSV and PIC appeared to inhibit SULT1E1 activity (Fig. 6B), based on comparison with the secretion patterns of estradiol following treatment with these substances, which were increased (Fig. 5B). The signifi- cant increase in estradiol secretion connected with the decrease in secretion of E1S indicates inhibition of SULT1E1 for both RSV and PIC. Thus, it is concluded
A
RSV
RSVTM
DHEA secretion (% of control)
RSVTA
PIC
160
RSVDA
140
120
100
80
*
T
60
*
*
40
20
*
*
* *
*
0
1
5
10
25
Estradiol secretion (% of control)
B
700
*
600
500
*
*
400
*
300
*
*
*
200
*
100
T
T
0
1
5
10
25
Concentration (uM)
that the suggested stimulatory effect of CYP19A1 (Fig. 5B) is at least partly due to inhibition of SULT1E1 for RSV and PIC. Control levels of E1S in vehicle control samples ranged between 0.6 and 1.37 ng/ml.
Effects on Steroidogenic Gene Expression
The effects on gene expression of nine steroidogenic genes were assessed following treatment with RSV, RSVTA, RSVTM, and PIC. As RSVTA and RSVDA had similar effects on steroid secretion only one of them, RSVTA, was tested by gene expression analysis. The two sulfotransferases were tested only with RSV and PIC, as in the enzyme activity tests.
RSV and RSVTM caused a significant down-regula- tion of CYP17A1 starting at 5 MM, to 68 and 53% of the controls, respectively (Fig. 7D). PIC caused a signifi- cant down-regulation of CYP11A1, CYP21A2, and HSD3B2 at 10 and/or 25 uM (Fig. 7B, C, and E). The expression of HSD3B2 was down-regulated by PIC to
DHEA-S secretion (% of control)
A
RSV
160
PIC
140
120
100
T
80
*
60
40
20
*
0
1
5
10
25
Concentration (µM)
Estrone-S secretion (% of control)
B
120
I
100
T
T
T
T
*
*
80
T
T
T
60
40
20
0
1
5
10
25
Concentration (uM)
44% compared to controls at 10 p.M. A dose-dependent up-regulation was demonstrated of CYP19A1 follow- ing treatment with RSVTA (Fig. 7G), reaching a maximum level of 1.9-fold at the highest tested concentration. The gene expression of SULT2A1 was significantly down-regulated by PIC (Fig. 8A), while SULT2E1 expression was significantly up-regulated both by PIC and RSV, causing a 1.7- and 1.9-fold increase at 25 p.M, respectively, compared to control (Fig. 8B). No statistically significant effects on gene expression were observed for StAR (Fig. 7A) and HSD17B1 (Fig. 7F).
DISCUSSION
Resveratrol (RSV) has gained considerable attention due to its potential cancer chemopreventive proper-
A
D
150
StAR
250
CYP17A1
RSV
RSVTA
200
RSVTM
PIC
100
150
50
100
50
0
0
B
150
CYP11A1
E
200
HSD3B2
150
100
100
Gene expression (% of control)
50
Gene expression (% of control)
50
0
0
C
F
150
CYP21A2
150
HSD17B1
100
100
50
50
0
0
1
5
10
25
Concentration (uM)
G
200
CYP19A1
*
*
150
100
50
0
1
5
10
25
Concentration (UM)
ties [8,32]. Prostate cancer is the most commonly diagnosed cancer in men, and chemopreventive agents effective in prevention and treatment of prostate cancer would be of utmost importance for public health. Several molecular mechanisms by which RSV may be active against prostate cancer have been demonstrated in vitro, including cell-cycle arrest, induction of apoptosis and attenuation in androgen receptor expression and function [33,34]. However, effects on androgen secretion have not been explored as a possible mechanism for RSV effects in prostate cancer. The adrenal cortex, as a producer of androgens, is a target in chemoprevention and treatment of prostate cancer. In the present study we have demon- strated that RSV and RSV analogs disrupt steroid secretion in a compound specific pattern in the
adrenocortical cell line H295R, causing a potent suppression of secretion of androgen steroids. Studies in the H295R cells also give information on effects on synthesis of glucocorticoids and mineralocorticoids, which are relevant for adverse effects of drugs inhibit- ing CYP17 activity.
We found a strong inhibition by all the test compounds on the secretion of androgen steroids. All compounds inhibited secretion of DHEA, testosterone, and cortisol, while secretion of progesterone was increased. Furthermore, aldosterone, which is down- stream of progesterone in the steroidogenic pathway, was also increased, but to a lower extent. Decreased secretion of androgens and cortisol in connection with increased secretion of progesterone and aldosterone is indicative of CYP17A1 enzyme inhibition. Indeed,
A
150
SULT2A1
RSV
PIC
100
*
Gene expression (% of control)
50
0
B
250
SULT1E1
*
200
*
150
100
50
0
1
5
10
25
Concentration (uM)
results from a CYP17A1 activity assay demonstrated inhibiting effects by all the compounds, except PIC. The reason for lack of effect by PIC in our CYP17A1 assay is not known, but a chemical interaction between PIC and trilostane or pregnenolone, added in the CYP17A1 assay, could not be ruled out.
CYP17A1 inhibitors are in focus for prevention and treatment of castrate-resistant prostate cancer, to inhib- it androgen production in the adrenal gland and the tumor, as steroidogenesis in these tissues are not affected by the chemical or surgical castration [14,35]. The antifungal drug and pesticide ketoconazole has been used in treatment of prostate cancer, as a potent but nonspecific CYP17A1 inhibitor with significant side effects, such as hepatotoxicity and adrenal insuffi- ciency, due to general inhibition of CYP enzymes. Abiraterone, a highly selective CYP17A1 inhibitor reducing serum androgen levels, was approved by FDA in 2012 for treatment of castrate-resistant prostate cancer [36]. Abiraterone has also been demonstrated to act by reducing androgen receptor activity in prostate cancer cells [37]. Orteronel is an imidazole CYP17A1 inhibitor, more selective on the 17,20-lyase activity, inhibiting androgen synthesis in rats and presently undergoing clinical evaluation in phase 3 [38,39]. In a
phase 2 trial in castrate-resistant prostate cancer patients orteronel treatment lowered the levels of circulating androgens, reduced prostate-specific anti- gen (PSA) and decreased the levels of circulating tumor cells [13].
The impact of orteronel, abiraterone, and ketocona- zole on the steroid secretion pattern has been investi- gated in the adrenocortical H295R cells [39], that is, the same model used by us in the present investigation and the effects could thus be compared with our results from resveratrol compounds. Interestingly, Yamaoka et al. [39] reported similar steroid secretion patterns, after treatment of cells with the three prostate cancer drugs, as we found in the present study with resveratrols, namely decreases in androgen and corti- sol secretion and increases in progesterone and aldo- sterone. The androgens were inhibited at lower concentrations compared to cortisol by orteronel and abiterone, indicating a higher inhibition of the 17,20- lyase than the hydroxylase activities of CYP17A1, while ketoconazole seemed to inhibit both CYP17A1 activities to a similar extent [39]. In contrast to our results on RSV and analogs, treatment of cells with orteronel and abiraterone resulted in higher secretion of aldosterone than of progesterone.
In our investigation cortisol secretion was less sup- pressed by the test compounds than was DHEA and testosterone, indicating a higher inhibition on the 17,20-lyase than on the hydroxylase activities of CYP17A1. Especially this was the case for RSVTM. RSVTM was the only compound to inhibit viability of the H295R cells. The adrenocortical H295R cells are not dependent on androgens for proliferation and a decrease in viability is interpreted as a consequence of cytotoxicity. However, in our studies inhibition of androgen synthesis was observed at concentrations below those causing decrease in cell viability, which implies that it is due to a specific mechanism and not general toxicity. Our results on CYP17A1 inhibition suggest a novel mechanism for RSV and analogs in chemoprevention of prostate cancer. It should be emphasized that the physiological significance of this effect from in vitro studies remains to be confirmed in vivo.
All the test compounds caused an almost identical effect on DHEA secretion as on testosterone secretion, which indicates that there were no treatment-related effects on enzyme activities of HSD3B2 or HSD17B. However, the potency of the effects differed between the compounds. The most efficient inhibitors of the hormone secretion and on the CYP17A1 activity were RSVTA and RSVDA. A weaker inhibition was dis- played by RSVTM. RSVTM down regulated the gene expression of CYP17A1 and showed a less steep concentration response relationship than the other test
The Prostate
compounds. PIC was a more efficient inhibitor of DHEA, testosterone, and cortisol secretion, while the stimulating effect on progesterone secretion was less prominent, than RSV. Gene expression of HSD3B2 was significantly down regulated by PIC, which apparent- ly did not affect the enzyme activity.
All the test compounds, except RSVTM, stimulated secretion of estradiol. This was demonstrated both in the basal cell system and in cells supplemented with testosterone in the CYP19A1 and SULT1E1 assay. Similarly, Basini et al. [18] reported increased secretion of estradiol after treatment with a hydroxylated resveratrol analog (2-hydroxy-3,5,4’-trimethoxystil- bene), but not after RSVTM, in swine granulosa cells supplemented with androstenedione. Increased estra- diol secretion could be due to induced enzyme activity of aromatase (CYP19A1), which has been reported after treatment of H295R cells with, for example atrazine and the flavonoids quercetin and genistein [40]. Increased secretion of estradiol could also result from inhibition of the downstream metabo- lism of estradiol to estradiol sulfate, a step catalyzed by SULT1E1. Formation of estradiol sulfate is an inactivation of estradiol, as the sulfate conjugate is unable to bind to the estrogen receptor. We found that the increase in estradiol secretion by RSV and PIC was associated with a decrease in estrone sulfate secretion (measured as an indicator of estradiol sulfate), indicat- ing an inhibition of SULT1E1 activity. Inhibition of SULT1E1 by resveratrol has previously been reported by Otake et al. [41] in human mammary epithelial cells and by Furimsky et al. [42] in S9 fractions from human liver and jejunum. In the present study we found an up regulation of the gene expression of SULT1E by RSV and PIC, possibly as a feedback response to the decreased enzyme activity. Our results emphasize that test substances, causing increased estrogen secretion in the H295R cells, should also be tested for down-stream effects, to avoid misclassifica- tion as CYP19A1 activators.
Of the synthetic analogs, the acetylated RSVTA and RSVDA had similar effects on steroid secretion pattern and were the most potent CYP17A1 inhibitors with a steep concentration response relationship. The methyl- ated analog RSVTM was a potent inhibitor of DHEA and testosterone secretion, a less potent inhibitor of cortisol secretion and a weak inducer of progesterone secretion, indicating a stronger inhibition on the 17,20- lyase than on the hydroxylase activity of CYP17A1. Furthermore, RSVTM did not affect aldosterone secre- tion, in contrast to the other compounds and did not increase estradiol secretion compared to controls, although compared to testosterone secretion estradiol was slightly increased. It has been demonstrated that substitution of hydroxyl groups in RSV with acetyl or
methoxy groups, such as in RSVTM, increases the metabolic stability and bioactivity [19,21]. According- ly, tumor formation and growth in mice with prostate cancer xenographs was more efficiently suppressed and inhibited by oral treatment with RSVTM than with PIC and RSV [9]. After 52 days oral administra- tion of 50 mg/kg body weight per day of RSV and RSVTM to mice with prostate cancer xenografts, serum levels of the respective chemical were 0.1 and 3.5 MM, respectively. The tumors had considerably higher concentrations of RSVTM compared to RSV; 3.324 versus 0.014 µg/g. PIC administered in a similar way was not detected in serum or in tumor tissue [9].
Due to the low bioavailability of RSV doubts have been expressed on its efficiency in vivo. In several studies where human volunteers consumed moderate amounts of red wine, very low concentrations of RSV, if any, were detected in blood [43,44]. Due to the increasing interest in the beneficial health effects of RSV, food supplements with higher amounts of RSV are commercially available. The high volume of distribution of RSV may result in a large part of the absorbed RSV bound to cell membranes or lipophilic tissues, such as the adrenal glands, which could explain a bioactive potential of RSV and PIC [17]. However, probably more promising candidates for prostate cancer chemoprevention are the synthetic RSV analogs, due to improved pharmacokinetic prop- erties and more potent CYP17A1 inhibition compared to RSV.
CONCLUSIONS
In summary, our study demonstrates that RSV, RSVTA, RSVDA, RSVTM, and PIC exert distinct effects on steroidogenesis and suppress secretion of androgens, mediated by CYP17A1 inhibition, in the human adrenocortical cells, H295R. The trimethylated resveratrol analog RSVTM, which recently has been demonstrated to inhibit tumor progression in prostate cancer xenografts, has a preferential inhibition of the 17,20-lyase activity of the CYP17A1 enzyme. Our data reveals a novel mechanism for resveratrol and resvera- trol analogs in chemoprevention of prostate cancer.
REFERENCES
1. Chang X, Heene E, Qiao F, Nick P. The phytoalexin resveratrol regulates the initiation of hypersensitive cell death in Vitis cell. PLOS ONE 2011;6(10):e26405.
2. Catalgol B, Batirel S, Taga Y, Ozer NK. Resveratrol: French paradox revisited. Front Pharmacol 2012;3:141.
3. Vang O. What is new for resveratrol? Is a new set of recommendations necessary? Ann N Y Acad Sci 2013;1290:1-11.
4. Cottart CH, Nivet-Antoine V, Beaudeux JL. Review of recent data on the metabolism, biological effects, and toxicity of resveratrol in humans. Mol Nutr Food Res 2014;58(1):7-21.
5. Piotrowska H, Kucinska M, Murias M. Biological activity of piceatannol: Leaving the shadow of resveratrol. Mutat Res 2012;750(1):60-82.
6. Potter GA, Patterson LH, Wanogho E, Perry PJ, Butler PC, Ijaz T, Ruparelia KC, Lamb JH, Farmer PB, Stanley LA, Burke MD. The cancer preventative agent resveratrol is converted to the antican- cer agent piceatannol by the cytochrome P450 enzyme CYP1B1. Br J Cancer 2002;86(5):774-778.
7. Piver B, Fer M, Vitrac X, Merillon JM, Dreano Y, Berthou F, Lucas D. Involvement of cytochrome P450 1A2 in the biotransforma- tion of trans-resveratrol in human liver microsomes. Biochem Pharmacol 2004;68(4):773-782.
8. Shankar S, Singh G, Srivastava RK. Chemoprevention by resveratrol: Molecular mechanisms and therapeutic potential. Front Biosci 2007;12:4839-4854.
9. Dias SJ, Li K, Rimando AM, Dhar S, Mizuno CS, Penman AD, Levenson AS. Trimethoxy-resveratrol and piceatannol adminis- tered orally suppress and inhibit tumor formation and growth in prostate cancer xenografts. Prostate 2013;73(11):1135-1146.
10. Ratan HL, Steward WP, Gescher AJ, Mellon JK. Resveratrol-A prostate cancer chemopreventive agent? Urol Oncol 2002;7(6): 223-227.
11. Svechnikov K, Spatafora C, Svechnikova I, Tringali C, Soder O. Effects of resveratrol analogs on steroidogenesis and mitochon- drial function in rat Leydig cells in vitro. J Appl Toxicol 2009; 29(8):673-680.
12. Ortega I, Villanueva JA, Wong DH, Cress AB, Sokalska A, Stanley SD, Duleba AJ. Resveratrol reduces steroidogenesis in rat ovarian theca-interstitial cells: The role of inhibition of Akt/ PKB signaling pathway. Endocrinology 2012;153(8):4019-4029.
13. Zhu H, Garcia JA. Targeting the adrenal gland in castration- resistant prostate cancer: A case for orteronel, a selective CYP-17 17,20-lyase inhibitor. Curr Oncol Rep 2013;15(2):105-112.
14. Vasaitis TS, Bruno RD, Njar VC. CYP17 inhibitors for prostate cancer therapy. J Steroid Biochem Mol Biol 2011;125(1-2):23-31.
15. Cottart CH, Nivet-Antoine V, Laguillier-Morizot C, Beaudeux JL. Resveratrol bioavailability and toxicity in humans. Mol Nutr Food Res 2010;54(1):7-16.
16. Walle T. Bioavailability of resveratrol. Ann N Y Acad Sci 2011; 1215:9-15.
17. Roupe KA, Yanez JA, Teng XW, Davies NM. Pharmacokinetics of selected stilbenes: Rhapontigenin, piceatannol and pinosylvin in rats. J Pharm Pharmacol 2006;58(11):1443-1450.
18. Basini G, Tringali C, Baioni L, Bussolati S, Spatafora C, Grasselli F. Biological effects on granulosa cells of hydroxylated and methylated resveratrol analogues. Mol Nutr Food Res 2010;54 (Suppl 2):S236-S243.
19. Lin HS, Ho PC. A rapid HPLC method for the quantification of 3,5,4’-trimethoxy-trans-stilbene (TMS) in rat plasma and its application in pharmacokinetic study. J Pharm Biomed Anal 2009;49(2):387-392.
20. Mazue F, Colin D, Gobbo J, Wegner M, Rescifina A, Spatafora C, Fasseur D, Delmas D, Meunier P, Tringali C, Latruffe N. Structural determinants of resveratrol for cell proliferation inhibition potency: Experimental and docking studies of new analogs. Eur J Med Chem 2010;45(7):2972-2980.
21. Liang L, Liu X, Wang Q, Cheng S, Zhang S, Zhang M. Pharmacokinetics, tissue distribution and excretion study of resveratrol and its prodrug 3,5,4’-tri-O-acetylresveratrol in rats. Phytomedicine 2013;20(6):558-563.
22. Sarpietro MG, Spatafora C, Tringali C, Micieli D, Castelli F. Interaction of resveratrol and its trimethyl and triacetyl deriva- tives with biomembrane models studied by differential scanning calorimetry. J Agric Food Chem 2007;55(9):3720-3728.
23. Cardile V, Chillemi R, Lombardo L, Sciuto S, Spatafora C, Tringali C. Antiproliferative activity of methylated analogues of E- and Z-resveratrol. Z Naturforsch C 2007;62(3-4):189-195.
24. Cardile V, Lombardo L, Spatafora C, Tringali C. Chemo- enzymatic synthesis and cell-growth inhibition activity of resveratrol analogues. Bioorg Chem 2005;33(1):22-33.
25. Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, Chrousos GP, Brennan MF, Stein CA, La Rocca RV. Establish- ment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthe- sis. Cancer Res 1990;50(17):5488-5496.
26. Rainey WE, Bird IM, Mason JI. The NCI-H295 cell line: A pluripotent model for human adrenocortical studies. Mol Cell Endocrinol 1994;100(1-2):45-50.
27. Oskarsson A, Ulleras E, Plant KE, Hinson JP, Goldfarb PS. Steroidogenic gene expression in H295R cells and the human adrenal gland: Adrenotoxic effects of lindane in vitro. J Appl Toxicol 2006;26(6):484-492.
28. Ulleras E, Ohlsson A, Oskarsson A. Secretion of cortisol and aldosterone as a vulnerable target for adrenal endocrine disrup- tion-Screening of 30 selected chemicals in the human H295R cell model. J Appl Toxicol 2008;28(8):1045-1053.
29. OECD. OECD Guideline for the testing of chemicals. Test no. 456: H295R Steroidogenesis Assay. 2011.
30. Nicolosi G, Spatafora C, Tringali C. Chemo-enzymatic pre- paration of resveratrol derivatives. J Mol Catal B Enzym 2002; 16(5-6):223-229.
31. Ohlsson A, Ulleras E, Oskarsson A. A biphasic effect of the fungicide prochloraz on aldosterone, but not cortisol, secretion in human adrenal H295R cells-underlying mechanisms. Toxicol Lett 2009;191(2-3):174-180.
32. Athar M, Back JH, Kopelovich L, Bickers DR, Kim AL. Multiple molecular targets of resveratrol: Anti-carcinogenic mechanisms. Arch Biochem Biophys 2009;486(2):95-102.
33. Bommareddy A, Eggleston W, Prelewicz S, Antal A, Witczak Z, McCune DF, Vanwert AL. Chemoprevention of prostate cancer by major dietary phytochemicals. Anticancer Res 2013;33(10): 4163-4174.
34. Mitchell SH, Zhu W, Young CYF. Resveratrol inhibits the expression and function of the androgen receptor in LNCaP prostate cancer cells. Cancer Res 1999;59(23):5892-5895.
35. Swanson HI, Njar VC, Yu Z, Castro DJ, Gonzalez FJ, Williams DE, Huang Y, Kong AN, Doloff JC, Ma J, Waxman DJ, Scott EE. Targeting drug-metabolizing enzymes for effective chemopreven- tion and chemotherapy. Drug Metab Dispos 2010;38(4):539-544.
36. Goldberg T, Berrios-Colon E. Abiraterone (zytiga), a novel agent for the management of castration-resistant prostate cancer. P T 2013;38(1):23-26.
37. Soifer HS, Souleimanian N, Wu S, Voskresenskiy AM, Collak FK, Cinar B, Stein CA. Direct regulation of androgen receptor activity by potent CYP17 inhibitors in prostate cancer cells. J Biol Chem 2012;287(6):3777-3787.
38. Hara T, Kouno J, Kaku T, Takeuchi T, Kusaka M, Tasaka A, Yamaoka M. Effect of a novel 17,20-lyase inhibitor, orteronel (TAK-700), on androgen synthesis in male rats. J Steroid Biochem Mol Biol 2013;134:80-91.
39. Yamaoka M, Hara T, Hitaka T, Kaku T, Takeuchi T, Takahashi J, Asahi S, Miki H, Tasaka A, Kusaka M. Orteronel (TAK-700), a novel non-steroidal 17,20-lyase inhibitor: Effects on steroid synthesis in human and monkey adrenal cells and serum steroid levels in cynomolgus monkeys. J Steroid Biochem Mol Biol 2012;129(3-5):115-128.
40. Sanderson JT, Hordijk J, Denison MS, Springsteel MF, Nantz MH, van den Berg M. Induction and inhibition of aromatase (CYP19) activity by natural and synthetic flavonoid compounds in H295R human adrenocortical carcinoma cells. Toxicol Sci 2004;82(1):70-79.
41. Otake Y, Nolan AL, Walle UK, Walle T. Quercetin and resveratrol potently reduce estrogen sulfotransferase activity in normal human mammary epithelial cells. J Steroid Biochem Mol Biol 2000;73(5):265-270.
42. Furimsky AM, Green CE, Sharp LEH, Catz P, Adjei AA, Parman T, Kapetanovic IM, Weinshilboum RM, Iyer LV. Effect of resveratrol on 17 beta-estradiol sulfation by human hepatic and jejunal S9 and recombinant sulfotransferase 1E1. Drug Metab Dispos 2008;36(1):129-136.
43. Vitaglione P, Sforza S, Galaverna G, Ghidini C, Caporaso N, Vescovi PP, Fogliano V, Marchelli R. Bioavailability of trans- resveratrol from red wine in humans. Mol Nutr Food Res 2005; 49(5):495-504.
44. Urpi-Sarda M, Jauregui O, Lamuela-Raventos RM, Jaeger W, Miksits M, Covas MI, Andres-Lacueva C. Uptake of diet resveratrol into the human low-density lipoprotein. Identifica- tion and quantification of resveratrol metabolites by liquid chromatography coupled with tandem mass spectrometry. Anal Chem 2005;77(10):3149-3155.