SOT
Society of Toxicology www.toxsci.oxfordjournals.org
OXFORD
doi: 10.1093/toxsci/kfu212
Advance Access Publication Date: October 15, 2014
Antiandrogenic Mechanisms of Pesticides in Human LNCaP Prostate and H295R Adrenocortical Carcinoma Cells
Christina N. Robitaille, Patricia Rivest, and J. Thomas Sanderson1
INRS-Institut Armand-Frappier, Université du Québec, 531 boulevard des Prairies, Laval, Quebec, Canada H7V 1B7
1To whom correspondence should be addressed at INRS-Institut Armand-Frappier, Édifice 18, PRF-K110, 531 boulevard des Prairies, Laval, Quebec, Canada H7V 1B7. Fax: +1 450 687 5566. E-mail: thomas.sanderson@iaf.inrs.ca.
ABSTRACT
Several pesticides suspected or known to have endocrine disrupting effects were screened for pro- or antiandrogenic properties by determining their effects on proliferation, prostatic-specific antigen (PSA) secretion and androgen receptor (AR) expression, and AR phosphorylation in androgen-dependent LNCaP human prostate cancer cells, as well as on the expression and catalytic activity of the enzyme CYP17 in H295R human adrenocortical carcinoma cells, an in vitro model of steroidogenesis. Effects on SRD5A gene expression were determined in both cell lines. Benomyl, vinclozolin, and prochloraz, but not atrazine, concentration dependently (1-30 uM) decreased dihydrotestosterone (DHT)-stimulated proliferation of LNCaP cells. All pesticides except atrazine decreased DHT-stimulated PSA secretion, AR nuclear accumulation, and AR phosphorylation on serines 81 and 213 in LNCaP cells. Benomyl and prochloraz, but not vinclozolin or atrazine, decreased levels of CYP17 gene and protein expression, as well as catalytic activity in H295R cells. In the case of prochloraz, some of these effects corresponded with cytotoxicity. H295R cells expressed AR protein and SRD5A1, but not SRD5A2 transcripts. SRD5A1 gene expression in H295R cells was increased by 10 nM DHT, whereas in LNCaP cells significant induction was observed by 0.1 nM DHT. AR protein expression in H295R cells was not increased by DHT. Vinclozolin decreased DHT-induced SRD5A1 gene expression in LNCaP, but not H295R cells, indicating a functional difference of AR between the cell lines. In conclusion, pesticides may exert antiandrogenic effects through several mechanisms that are cell type-specific, including AR antagonism and down-regulation or catalytic inhibition of androgen biosynthetic enzymes, such as CYP17 and SRD5A1.
Key words: androgen receptor; pesticides; LNCaP; H295R; CYP17; SRD5A1
Downloaded from http://tox sci.oxfordjournals.org/ at Carleton University on December 1, 2014
Accumulating evidence implicates certain pesticides in causing endocrine disrupting effects that can be associated with repro- ductive and developmental dysfunction as well as with an in- creased risk of development of hormone-dependent cancers (Brevini et al., 2005; Schug et al., 2011). Many studies focus on estrogenic effects, but pro- or antiandrogenic mechanisms of action of endocrine disruptors have not been investigated in the same detail. A notable exception is the fungicide vinclozolin (Kavlock and Cummings, 2005), which has been reported to directly antagonize the androgen receptor (AR) (Wong et al., 1995), thus preventing its activation by endogenous androgens such as testosterone and dihydrotestosterone (DHT).
The consequence of this disruption is a demasculinization of male offspring in rats that were exposed to 100 and 200 mg vin- clozolin/kg body weight/day from gestational day 14 until post- natal day 3 (Gray et al., 1994) and alterations in androgen- dependent developmental processes such as prostate growth and androgen-mediated gene expression in this tissue in adult male rats (Kelce et al., 1997). A study in female offspring of rats exposed to 1 mg vinclozolin/kg body weight/day for the duration of pregnancy showed alterations in female mammary glands, including an increase in epithelial branching and loss of epithe- lial cell polarization at postnatal day 35, but which were less pronounced at postnatal day 50 (El Sheikh Saad et al., 2013).
@ The Author 2014. Published by Oxford University Press on behalf of the Society of Toxicology.
All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
Such effects were consistent with altered mammary gland de- velopment in female mice lacking AR (Yeh et al., 2003).
Pesticides may act through various potential mechanisms to cause disruption of androgenic function, which beside an inter- action with the AR, as exemplified by vinclozolin, may include interferences with the function of steroidogenic enzymes or “cross-talk” pathways that influence AR signaling by altering AR phosphorylation or degradation, or its affinity for coregulators (Balk and Knudsen, 2008). The AR is a key target for pro- or anti- androgenic compounds and is involved in (sexual) development in both males and females, as well as in the progression of certain hormone-dependent cancers such as that of prostate. Key ste- roidogenic enzymes that may be targeted by androgen-disrupting chemicals are cytochrome P450 17 (CYP17), which conveys steroid 17a-hydroxylase and 17,20-lyase activities and is responsible for the synthesis of precursors for all androgens, and steroid 5x-re- ductase (SRD5A), which converts testosterone to the significantly more potent androgen DHT (Sanderson, 2006).
Given the time-consuming and expensive aspects of in vivo experiments to determine disruptive effects of chemicals on androgenic function, a number of in vitro tools have been developed to rapidly screen for pro- or antiandrogenic effects of compounds of interest to be selected for in vivo studies (Beck et al., 2008; Kolle et al., 2010; Sohoni and Sumpter, 1998).
In the present study, we used LNCaP human prostate cancer cells, which express functional AR (Horoszewicz et al., 1983; van Steenbrugge et al., 1989), to determine the effects of several pes- ticides suspected or known to have endocrine disrupting prop- erties on androgen-dependent cell proliferation, prostate- specific antigen (PSA) secretion, SRD5A expression, and phos- phorylation and nuclear accumulation of AR protein. We used H295R human adrenocortical carcinoma cells as a model of ste- roidogenesis (Gazdar et al., 1990; Hilscherova et al., 2004; Sanderson et al., 2000; Sanderson, 2006; Staels et al., 1993) to de- termine effects of pesticides on the function of the androgen synthesizing enzyme CYP17 and on SRD5A gene expression. Our aim was to use these cell-based assays to delineate differential mechanisms of (anti)androgenic action of atrazine, benomyl, vinclozolin, and prochloraz.
MATERIALS AND METHODS
Pesticides. All pesticides were purchased from Sigma-Aldrich (St. Louis, MO). Atrazine and benomyl were Fluka PESTANAL grade with purities of 97.4% and 99.0%, respectively. Vinclozolin and prochloraz were Riedel deHaen analytical grade with purities of 99.6% and 99.1%, respectively. Their chemical structures are shown in Table 1.
| Pesticide (CAS Number) | Structure | Classification | LNCaP Cell Bioassay | H295R Cell Bioassay | |
|---|---|---|---|---|---|
| CI | Cell proliferation | n.e. | CYP17 gene expression n.e. | ||
| PSA secretion | n.e. | CYP17 protein expression n.e. | |||
| N N | AR accumulation | n.e. | CYP17 catalytic activity n.e. | ||
| Atrazine (1912-24-9) | Triazine herbicide | AR phosphorylation | n.e. | ||
| CH3CH2NH N NHCH(CH3)2 | |||||
| Benomyl (17804-35-2) | O | Cell proliferation | ; | CYP17 gene expression ; | |
| N N H O | Benzimidazole fungicide | PSA secretion | CYP17 protein expression 1 | ||
| AR accumulation AR phosphorylation | 44 ; | CYP17 catalytic activity ; | |||
| CI CI | Cell proliferation | 44 | CYP17 gene expression n.e. | ||
| PSA secretion | 444 | CYP17 protein expression n.e. CYP17 catalytic activity n.e. | |||
| AR accumulation | 444 | ||||
| Vinclozolin (50471-44-8) | O O H3C CH CH2 | Dicarboximide fungicide | AR phosphorylation SRD5A1 gene expression | 44 44 | SRD5A1 gene expression n.e. |
| Prochloraz (67747-09-5) | H. H3C CH2 CI CI H. O O H2 CI N N | Triazole fungicide | Cell proliferation PSA secretion AR accumulation AR phosphorylation | 44 444 ; | CYP17 gene expression 11 CYP17 protein expression 44 CYP17 catalytic activity 444 |
Notes: The effects of vinclozolin on DHT-mediated induction of SRD5A1 gene expression were studied in both cell lines. Abbreviations and symbols: (n.e.) no effect. Inhibition observed at (III) ≥ 1 uM; (11) ≥ 10 AM; (1) = 30 µM.
Downloaded from http://toxsci.oxfordjournals.org/ at Carleton University on December 1, 2014
Cell culture. LNCaP human prostate carcinoma cells were obtained from American Type Culture Collection (ATCC, Manassas, Virginia) and grown in RPMI 1640 medium (Life Technologies; Cat. No. 21870-076) supplemented with 1% L-glu- tamine (21051-024), 1% penicillin/streptomycin (15070-063), 1% HEPES buffer (15630-080), 10% fetal bovine serum (FBS) (12483- 020), and 1% sodium pyruvate (11360-070). Cells were exposed to pesticides in phenol red-free medium (11835-030), supple- mented with 2% dextran/charcoal-stripped FBS (12676-011). Pesticides and DHT (A2570-000; Steraloids, Newport, Rhode Island) were dissolved in dimethyl sulfoxide (DMSO) (D4540; Sigma-Aldrich) as 1000-fold stock solutions. DMSO was used as solvent control and final DMSO concentrations in culture medium were either 0.1% or 0.2% dependent on whether the experiment required single or dual compound exposures. H295R human adrenocortical carcinoma cells (ATCC) were grown in Dulbecco’s Modified Eagle’s medium with phenol red, complemented with 1% L-glutamine, 1% penicillin/streptomy- cin, 1% insulin-transferrin/sodium selenite/bovine serum albu- min/linoleic acid (ITS+1) (Sigma-Aldrich), and 2.5% Nu-serum (BD Biosciences). H295R cells were seeded in Dulbecco medium without supplement, when exposed with hormones treatments.
Cell proliferation and viability. LNCaP cells were seeded in 96-well plates (5 × 103 cells/well) in phenol red- and steroid-free medium and acclimatized for 24h to arrest cell proliferation. Cells were then exposed to fresh medium containing 1, 3, 10, or 30 uM pesticides in presence or absence of 0.1 nM DHT; cells were re-exposed to identical treatments in fresh medium 24h later and after another 72h the number of viable cells was determined using a WST-1 cell viability kit (Roche, Mississauga, Canada). Cells were incubated with WST-1 substrate for 2h, after which the formation of formezan was quantified by meas- uring its absorbance at a wavelength of 440 nm and a reference wavelength of 630 nm using a SpectraMax M5 spectrophotome- ter (Molecular Devices, Sunnyvale, California). H295R cell viabil- ity was determined in 24-well plates (5 x 104 cells/well) using a WST-1 assay after a 24-h exposure to 1, 3, 10, 30, or 100 µM of each pesticide.
PSA secretion. LNCaP cells (0.3 × 104 cells/well) in 12-well plates (Corning) were exposed to pesticides with or without 0.3 nM DHT for 24h, after which medium was collected and centri- fuged. PSA levels were determined using a commercial enzyme- linked immunoabsorbance assay kit (No. EL10050; Anogen, Mississauga, Ontario, Canada). Media from cells treated without DHT were diluted 10 times and from those treated with DHT 20 times prior to assay. Cellular protein concentrations were determined using a bicinchoninic acid (BCA) protein assay kit (Pierce Biotechnology, Rockford, Illinois). This colorimetric test is based on the reduction of Cu2+ to Cu+ by proteins in alkaline medium. Proteins were quantified by measuring the absorbance of Cu+ at a wavelength of 562 nm using a SpectraMax M5 spectrophotometer. PSA secretion levels were expressed in ng per 100 ng cellular protein.
AR protein expression and phosphorylation. LNCaP cells (5 x 105 cells/well) in 6-well plates (Corning) were exposed to the pesti- cides in the presence or absence of 10nM DHT for 24h, after which cytoplasmic and nuclear proteins were extracted using a NE-PER extraction kit (Pierce Biotechnology). Nuclear proteins (25 µg) were separated by sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) by migration for 1 h through a 10%
SDS-PA gel at 150 V, after which proteins were transferred over a 10-min period to a polyvinylidene fluoride membrane using a Trans-Blot Turbo system (Bio-Rad, Mississauga, Ontario, Canada). Membranes were blocked with Tris-buffered saline con- taining 5% milk powder for 1h and then incubated overnight in the same blocking buffer containing primary antibody. H295R cells (5.5 × 105 cells/well) were treated with various concentra- tions of DHT, and with 10 uM vinclozolin in the presence or absence of 10 nM DHT, after which AR protein levels were deter- mined in total lysates, as described for LNCaP cells above. Phosphorylated AR levels were determined in total lysates of LNCaP or H295R cells exposed for 1, 6, or 24 h to pesticides (H295R cells were exposed to vinclozolin only) with or without DHT. When using phospho-antibodies, the blocking buffer contained bovine serum albumin instead of milk. Primary antibodies against AR were PG-21 (Cat. No. FCABS415PE, Millipore, Billerica, Massachusetts), p-AR-Ser81 (07-1375, Millipore), and p-AR- Ser210/213 (ab45089, Abcam, Cambridge, UK). Secondary antibod- ies were horse radish peroxidase-conjugated antirabbit (31210, Pierce Biotechnology) for AR and p-AR-Ser81 and HRP-conjugated antimouse (31172, Pierce Biotechnology) for p-AR-Ser213. Protein bands were made visible by immunofluorescence with Clarity Western ECL substrate (170-5061) and a ChemiDoc MP Imager (Bio-Rad). Densities of the protein bands were analyzed by Image Lab Software (Bio-Rad).
CYP17 and SRD5A gene expression. Gene expression was quanti- fied by real-time reverse transcriptase-polymerase chain reac- tion (RT-qPCR). Effects on CYP17 gene expression were studied in H295R cells only, because LNCaP cells were found not to express CYP17. H295R cells (5.5 x 105 cells/well) were added to 6-well plates in 4 ml of culture medium and left to adhere for 24 h. Cells were then exposed to fresh medium containing 30 uM of the pesticides atrazine, benomyl, or vinclozolin, or 10 uM pro- chloraz. Forskolin (F6886; Sigma-Aldrich) at a final concentra- tion of 10 uM was used as a positive control for CYP17 gene induction to verify the responsiveness of the H295R cells. To determine effects on SRD5A1/2 expression, H295R and LNCaP cells (5.5 x 105 cells/well) that had acclimatized in steroid-free medium were exposed to various concentrations of DHT to determine the induction response of SRD5A1 in each cell type; SRD5A2 was not expressed in either cell line. H295R and LNCaP cells were then exposed to vinclozolin (10 uM) in the presence or absence of an optimally inducing concentration of DHT. For all gene expression experiments, wells without cells were used as negative controls. Total RNA was extracted using a High Pure RNA Isolation kit (Cat. No. 11828665001, Roche). RNA was quan- tified by measuring its absorbance at a wavelength of 260 and 280 nm using a SpectraMax M5 spectrophotometer. After reverse transcription using a Thermocycler T3000 (Biometra, Goettingen, Germany), 1 µg of the resultant cDNA was amplified by PCR using a Thermal Cycler C1000 (Bio-Rad) and primers that selectively recognized either CYP17 (fw: 5’-GCC-TTC-CTG-CTG- CAC-AAT-CCT-3’; rv: 5’-AAA-CTC-ACC-GAT-GCT-GGA-GTC- AAC-3’; amplicon: 207 bp), SRD5A1 (fw: 5’-GCG-AGG-AGG-AAA- GCC-TAT-GC-3’; rv: 5’-CAG-GGC-ATA-GCC-ACA-CCA-CT-3’; amplicon: 316 bp), or SRD5A2 (fw: 5’-ATT-GCG-CCA-GCT-CAG- GAA-G-3’; rv: 5’-TGG-AAT-AAG-GGC-TTT-CCG-AGA-T-3’; ampli- con: 256 bp) (Soderstrom et al., 2001). Beta-actin and glyceralde- hyde-3-phosphate dehydrogenase were verified to be suitable reference genes and used to normalize gene expression.
CYP17 protein expression. H295R cells (5.5 x 105 cells/well) were added to 6-well plates in 4ml of medium and acclimatized for
A
150
Atrazine
Vinclozolin
Cell proliferation (% DMSO control)
Benomyl
Prochloraz
T
T
100-
T
T
T
T
I
T
Z
Y
T
T
T
I
T
T
ST
I
T
T
T
T
50-
0
DMSO
1 µM
3 µM
10 µM
30 µM
B
#
Atrazine
Vinclozolin
600-
Benomyl
Prochloraz
T
Cell proliferation (% DMSO control)
T
T
Y
I
T
T
400
I
T
T
T.
T
¥
*
I
T
T
*
*
T
₭
I
200
T
0
DMSO
0 µM
1 µM
3 µM
10 µM
30 µM
1
+ 0.1 nM DHT
24h. Cells were exposed to fresh medium containing 30 uM atrazine, benomyl, or vinclozolin, or 10 uM prochloraz. Forskolin (10 µM) was used as positive control and cells exposed to DMSO acted as a solvent control. Proteins were extracted using a radioimmunoprecipitation assay lysis buffer (Sigma-Aldrich) with 1% anti-protease and 1% anti-phosphatase. Total protein (50 µg) underwent SDS-PAGE and immunoblotting using the same protocol as for AR, except with CYP17-selective antirabbit antibodies at a 1:100 dilution (Cat. No. 14447-1-AP; Acris, San Diego, California) and HRP-conjugated secondary antirabbit antibodies 1:5000 dilution (Cat. No. 31210, Pierce Biotechnology).
CYP17 catalytic activity. H295R cells (5x104 cells/well) were added to 24-well plates in 1ml medium and left to adhere for 24 h. Cells were then exposed to pesticides in fresh medium
A
200-
Atrazine
Vinclozolin
Benomyl
Prochloraz
PSA secretion (% DMSO control)
T
150-
T
T
I
T
T
T
T
T
1
100-
T
T
T
T
I
T
I
T
T
I
=
50-
0
DMSO
0.3 µM
1 μ.Μ
3 µM
10 μ.Μ
30 μΜ
B
#
Atrazine
Vinclozolin
400-
Benomyl
Prochloraz
I
PSA secretion (% DMSO control)
300-
I
I
I
1
1
200-
Z
*
*
*
*
*
*
*
3.
*
*
I
1
I
T
1
*
100-
T
I
*
I
T
1
T
T
×
0
DMSO
0 HM
0.3 µM
1 µM
3 µM
10 µM
30 μΜ
1
1
+ 0.3 nM DHT
for another 24h, after which the medium was replaced with fresh medium without supplements, containing 100 nM of the CYP17 substrate pregnenolone and 5 uM of the 3ß-hydroxys- teroid dehydrogenase (3ß-HSD) inhibitor trilostane. After a 90- min incubation, the reaction medium was removed and dehy- droepiandrosterone (DHEA) content was quantified using an ELISA kit (EIA3415; DRG Intl., New Jersey). This assay design allowed for the integrated measurement of both the 17a- hydroxylase and 17,20-lyase activity of CYP17 without the rapid loss of either substrate or product via 3B-HSD-mediated reac- tions, which are known to be strong in H295R cells (Canton et al., 2006).
Statistical analysis. Results were presented as means with stand- ard errors. Statistically significant differences (*P <0.05) from control were determined by Student’s t test or by one-way anal- ysis of variance (ANOVA) and a Dunnett’s post hoc test to correct for multiple comparisons to control. All statistical analyses were performed using GraphPad Prism 5.04 (GraphPad Software, San Diego, California).
A
#
B
#
Relative band intensity (% DMSO control)
500-
Relative band intensity (% DMSO control)
800
400-
600-
300-
400
200-
100-
200
0
DMSO
Atrazine 1 µM
Atrazine 10 µM
DHT (10 nM)
Atrazine 1 µM + DHT
Atrazine 10 µM +DHT
0
DMSO
Benomyl 1 µM
Benomyl 10 UM
DHT (10 nM)
Benomyl 1 µM + DHT
Benomyl 10 µM +DHT
AR
AR
ß-actin
ß-actin
C
#
D
#
Relative band intensity (% DMSO control)
600-
Relative band intensity (% DMSO control)
400
400-
300-
200-
200
*
*
100-
0
DMSO
Vinclozolin 1 µM
Vinclozolin 10 µM
DHT (10 nM)
Vinclozolin 1 µM + DHT
Vinclozolin 10 µM +DHT
0
DMSO
Prochloraz 1 µM
Prochloraz 10 µM
DHT (10 nM)
Prochloraz 1 µM + DHT
Prochloraz 10 µM +DHT
AR
ß-actin
AR
ß-actin
RESULTS
Antiandrogenic Effects in LNCaP Cells
None of the pesticides affected basal LNCaP cell proliferation under steroid-deprived conditions at concentrations between 1 and 30 µM (Fig. 1A). An incubation of LNCaP cells with 0.1nM DHT increased cell proliferation 5- to 6-fold above control (Fig. 1B) that corresponded to a 75% effective concentration (EC75) for this response, which was consistent with our previous studies (Rivest et al., 2011). Passage number had a strong effect
on androgen-stimulated cell proliferation, with passage num- bers above 18 resulting in strongly diminished stimulation, as reported previously (Sanderson et al., 2013). DHT-stimulated LNCaP cell proliferation was reduced concentration depend- ently and statistically significantly by benomyl (30 uM), vinclo- zolin (≥10µM), and prochloraz (≥10 uM), but not by atrazine (Fig. 1B). At 30 uM, benomyl, vinclozolin, and prochloraz reduced DHT-stimulated cell proliferation by 55%, 60%, and 88%, respectively.
#
250
T
p-AR-Ser81
1
p-AR-Ser213
Band Instensity (% DMSO)
200-
T
T
T
150-
T
1
T
*
*
*
*
T
*
T
T
T
T
100-
T
*
T
50-
* *
T
0
DMSO
DHT 10 nM
Atrazine 10 uM
Benomyl 10 µM
Benomyl 30 uM
Prochloraz 10 µM
Prochloraz 30 uM
Vinclozolin 10 µM
Vinclozolin 30 µM
1
+ 10 nM DHT
p-AR-Ser81
p-AR-Ser210/213
amido black
None of the pesticides at concentrations between 1 and 30 µM affected basal secretion of PSA by steroid-deprived LNCaP cells (Fig. 2A). DHT increased PSA secretion by LNCaP cells about 3-fold at an EC75 of 0.3 nM for this particular response (Fig. 2B). DHT-stimulated PSA secretion decreased concentration dependently after exposure to benomyl, prochloraz, and vinclo- zolin, but not atrazine (Fig. 2B). These decreases were statisti- cally significant at concentrations of 1 uM and above, with inhibition between 90% and 100% at each concentration.
DHT (24 h exposure) increased levels of AR in the nucleus of LNCaP cells about 4- to 6-fold at 10 nM, which was determined to be the EC75 for this response (Fig. 3). Benomyl (10 uM), vinclo- zolin (1 and 10 µM), and prochloraz (1 and 10 „M) significantly reduced the DHT-stimulated increase in nuclear AR levels after a 24-h exposure (Figs. 3B-D), whereas atrazine had no effect (Fig. 3A). To assess whether DHT could stimulate the phosphor- ylation of AR at positions known to convey transcriptional activity to the receptor (Gioeli et al., 2002; Gioeli and Paschal, 2012), we determined the effects of DHT and the pesticides on levels of AR phosphorylated at serine positions 81 (p-AR-Ser81) and 213 (p-AR-Ser213). A preliminary concentration and time- course experiment showed that levels of phosphorylated AR were highest after a 1-h exposure to 10 nM DHT (not shown), which was consistent with other reports (Gioeli et al., 2002). Upon exposure, the fungicides vinclozolin (10 and 30 uM), benomyl (30 µM), and prochloraz (30 µM), but not atrazine, decreased the DHT-mediated phosphorylation of AR on Ser81 and Ser213 in LNCaP cells (Fig. 4).
CYP17 Expression and Catalytic Activity in H295R Cells
As an alternate mechanism of pro- or antiandrogenicity, the effects of the pesticides on CYP17 expression and catalytic
125-
☒ Atrazine
Cell viability (% DMSO control)
100
T
T
Benomyl
T
T
T
T
T
I
Vinclozolin
I
I
T
I
75-
I
Prochloraz
T
X
T
50-
*
*
T
25-
T
*
T
*
0
DMSO
1 µM
3 µM
10 µM
30 μΜ
100 μΜ
activity were evaluated in H295R cells; we did not detect CYP17 expression or activity in LNCaP cells (not shown). To differenti- ate inhibitory effects from possible cytotoxicity, the viability of H295R cells was determined after a 24 h exposure to each pesti- cide (1-100 uM). Prochloraz was cytotoxic at concentrations greater than 10 µM; vinclozolin was cytotoxic at 100 uM (Fig. 5). Exposure to benomyl (30 µM) or prochloraz (10 uM) decreased CYP17 gene expression in H295R cells (Fig. 6A). These two pesti- cides also decreased CYP17 protein expression (Fig. 6B). Pregnenolone-stimulated DHEA production, our measure of CYP17 activity, was decreased in H295R cells exposed to benomyl (30 uM) and prochloraz (1 and 10 uM) (Fig. 6C). Neither atrazine nor vinclozolin had an effect on CYP17 gene and pro- tein expression or its catalytic activity.
SRD5A Expression in H295R and LNCaP Cells
Another possible steroidogenic mechanism of pro- or antian- drogenicity is a chemical-induced change in the expression of SRD5A genes. We found similar levels of SRD5A1 gene expres- sion in H295R and LNCaP cells; neither cell line expressed SRD5A2 (not shown). In both cell lines, SRD5A1 gene expression was inducible by DHT, although statistically significant increases occurred at a 10-fold lower concentration in LNCaP than H295R cells (Fig. 7A). DHT-mediated induction of SRD5A1 expression was decreased statistically significantly by 10 UM vinclozolin in LNCaP but not H295R cells (Fig. 7B). Similarly, nei- ther atrazine (30 „M), benomyl (30 µM), nor prochloraz (1 µM) had an effect on SRD5A1 expression in H295R cells (not shown).
AR Expression in H295R Cells
The similar induction of SRD5A1 expression by DHT in LNCaP and H295R cells suggests that H295R cells contain functional AR. We found that AR protein was expressed at relatively high basal levels in H295R cells, but that its expression was not increased by a 24-h exposure to 10 nM DHT. Neither did a 24-h exposure to 1 or 10 uM vinclozolin in the presence or absence of DHT have an effect on AR expression relative to control (Fig. 8). In addition, we did not observe increased phosphorylation of AR on Ser81 in response to DHT and failed to detect any phosphor- ylation of AR on Ser213 (not shown).
DISCUSSION
Antiandrogenic Effects in LNCaP Cells
Our results using LNCaP cells confirm that certain pesticides, in particular vinclozolin, exert antiandrogenic effects, such as
A
#
B
#
700-
600-
500
CYP17 gene expression (% DMSO)
500-
400
CYP17 protein expression
400
300
200
300-
150
200
(% DMSO)
150
100-
100-
*
50-
*
50-
*
*
0
Forskolin 10 µM
DMSO
Atrazine 30 µM
Benomyl 30 µM
Vinclozolin 30 µM
Prochloraz 10 uM
0
Forskolin 10 µM
DMSO
Atrazine 30 µM
Benomyl 30 µM
Vinclozolin 30 uM
Prochloraz 10 uM
C
150-
DHEA production (% DMSO)
T
100
*
50
*
*
0
DMSO (0.2%)
Atrazine 3 µM
Atrazine 30 uM
Benomyl 3 uM
Benomyl 30 µM
Vinclozolin 3 µM
Vinclozolin 30 µM
Prochloraz 1 µM
Prochloraz 10 µM
reduced DHT-stimulated LNCaP cell proliferation (Fig. 1), PSA secretion (Fig. 2), and nuclear AR accumulation (Fig. 3). Prochloraz and benomyl appeared to act similar to vinclozolin, although benomyl was less potent. The observed antiandro- genic effects of vinclozolin are consistent with other studies of its endocrine disrupting properties, which have been attributed specifically to an interference with AR ligand binding, resulting in disrupted male development and demasculinization (Gray et al., 1994, 1999; Kavlock and Cummings, 2005; Kubota et al., 2003). Prochloraz has also been reported to have antiandrogenic effects in vivo and in vitro (Vinggaard et al., 2002, 2005), but benomyl has not been associated with any antiandrogenic activity, neither in vitro nor in vivo (Scippo et al., 2004; Yamada et al., 2005). However, we find that benomyl, at least at 30 uM, reduces DHT-stimulated LNCaP cell proliferation (Fig. 1) and PSA secretion (Fig. 2), as well as decreasing nuclear AR accumu- lation (Fig. 3) and AR phosphorylation on serine positions 81
and 213 (Fig. 4), indicating it is antiandrogenic at high concen- trations. This appears to be corroborated by a study showing that at a high dose of 100mg/kg body weight for 10 days, benomyl reduced the number of seminal vesicles and weight of the levator ani-bulbocavernosus muscles in castrated rats sup- plemented with testosterone (Hershberger assay) (Yamada et al., 2005).
Atrazine did not exert any antiandrogenic effects in LNCaP cells. This is also consistent with the known behavior of atra- zine, which has clear endocrine disrupting properties in vivo, but that are estrogenic and feminizing in nature (Belloni et al., 2011; Hayes et al., 2002; Roy et al., 2009; Wilhelms et al., 2006) and not related to an interaction with the AR signaling pathway (Vinggaard et al., 2008).
The antiandrogenic potencies of the pesticides appeared to be dependent on the nature of the DHT-stimulated response in question. This may in part have to do with the amount of DHT
A
Relative SRD5A1 gene expression (% DMSO)
500
H295R
LNCaP
400
*
300
*
*
*
200
100
0
DMSO
DHT 0.1 nM
DHT 1 nM
DHT 10 nM
B
Relative SRD5A1 gene expression (% DMSO)
400
#
H295R
LNCaP
300
200
*
100
0
DMSO
DHT
Vinclozolin (10 µM)
Vinclozolin + DHT
required to exert a specific androgenic effect in LNCaP cells (under our conditions). Far less DHT was required to induce LNCaP cell proliferation or PSA secretion, than to observe signif- icant nuclear AR accumulation. It is probable that far greater receptor occupancy is required for AR protein stabilization, which is the main mechanism of androgen-mediated increases in cellular and nuclear AR levels (Zhou et al., 1995), than for other more sensitive AR signaling responses. Among the anti- androgenic responses elicited by the pesticides in LNCaP cells, the DHT-stimulated secretion of PSA appears to be the most sensitive endpoint as it was decreased by each of the three anti- androgenic pesticides at a concentration as low as 1 uM.
+ DHT (10 nM)
DMSO
DHT
(10 nM)
Vinclozolin (1 μΜ)
Vinclozolin (10 μΜ)
Vinclozolin (1 μΜ)
Vinclozolin (10 μ.Μ)
AR
Beta-actin
AR Phosphorylation in LNCaP Cells
We are the first to report that certain pesticides can inhibit the androgen-induced phosphorylation of AR at sites known to be associated with increased translational activity (Gioeli et al., 2002; Gioeli and Paschal, 2012; Koryakina et al., 2014). The DHT- induced phosphorylation of AR on positions serine 81 and 213, which reside in the N-terminal domain of the receptor, is reported to convey transactivational activity, possibly by enhancing the interaction of this domain with AR coactivating factor, AF-1 (Gioeli, 2005). It is thought that the binding of DHT to the AR induces a conformational change that increases expo- sure of the phosphorylation site(s) to certain kinases. Cyclin- dependent kinases CDK1 (Chen et al., 2006), CDK9 (Gordon et al., 2010), and CDK5 (Hsu et al., 2011) have been reported to phos- phorylate AR on serine 81, which is the highest stochiometric phosphorylation site stimulated by DHT. Serine 81 phosphory- lation results in reduced ubiquitination of AR, and subsequent protein stabilization and rapidly increased intracellular levels of AR in the absence of increased AR gene expression (Chen et al., 2006; Hsu et al., 2011). The Ser213 position is phosphorylated by Akt (protein kinase B), which has an established role in sus- tained proliferation of prostate cancer cells when overexpressed due to a loss of phosphatase and tensin homolog (PTEN) activ- ity, which normally keeps levels of the PI3K product phosphati- dylinositol-3,4,5-triphosphate low enough to suppress PI3K/ Akt-mediated proliferative signaling (Mulholland et al., 2005).
Benomyl, vinclozolin, and prochloraz reduced levels of phos- phorylated p-AR-Ser81 and p-AR-213 (Fig. 4), indicating that these pesticides are effectively reducing the transactivational capacity of the AR. This is likely caused by preventing DHT from interacting with the ligand-binding domain of the AR, given both AR phosphorylation sites are reduced to a similar extent. However, other possible mechanisms could play a role, such as inhibition by the pesticides of various kinase activities or the stimulation of specific phosphatases, such as PPA2, which is known to dephosphorylate AR at position Ser81 (Yang et al., 2007). In fact, these mechanisms are likely interdependent, as pesticide-mediated blockade of androgen binding to the AR would reduce exposure of the phospho-sites to kinases as well as enhance the recruitment of phosphatases, which would not normally be able to target androgen-bound AR (Yang et al., 2007).
Steroidogenic Effects in H295R Cells
We demonstrated that benomyl decreased CYP17 gene expres- sion as well as protein levels and catalytic activity in H295R cells (Fig. 6). The genetic down-regulation of this enzyme by benomyl has not been reported previously. CYP17 expression in human adrenal cortex and in H295R cells is, in part, regulated by the cyclic AMP/protein kinase A signaling pathway, and the CYP17 promoter is under a complex control by various transcription
factors and coregulators, such as steroidogenic factor-1 (SF-1), GATA-6, and Sp1 among others (Sewer and Jagarlapudi, 2009). Any number of these functions could be disrupted by benomyl and the mechanisms of benomyl-mediated down-regulation of CYP17 deserve further study.
Prochloraz was a strong inhibitor of CYP17 gene and protein expression (Figs. 6A and 6B), although this occurred at a concen- tration (10 µM) that was partially toxic to the cells. However, prochloraz inhibited the catalytic activity of CYP17 by 73% at a noncytotoxic concentration of 1 µM (Fig. 6C), which is consistent with previous reports (Ohlsson et al., 2009).
To our knowledge, we report for the first time that H295R cells express SRD5A1 and AR. SRD5A1 is an AR-responsive gene, which is induced by potent androgens, as we found previously in LNCaP cells (Gasmi and Sanderson, 2010). Here, we show that SRD5A1 gene expression is induced in both LNCaP and H295R cells (Fig. 7A), suggesting that AR may control similar genes in both cell types. However, the functions of AR in H295R cells are unknown and the receptor does not respond to androgens with increased intracellular protein levels (Fig. 8). Nor did the AR antagonist vinclozolin reduce the induction of SRD5A1 in H295R cells, whereas it did so in LNCaP cells (Fig. 7B). The apparently high constitutive level of p-AR-Ser81 and lack of p-AR-Ser213 suggest different kinases play a role in AR phosphorylation and subsequent AR-signaling in H295R cells. The relevance of AR expression in H295R cells given the role of this cell line as a screening tool for chemical interferences with steroidogenesis is, thus far, unknown.
In conclusion, endocrine disrupting pesticides may act via multiple mechanism to exert antiandrogenic effects as sum- marized in Table 1. Benomyl appears to be an effective down- regulator of CYP17 gene expression, vinclozolin acts principally as an AR antagonist, and prochloraz exerts combined mecha- nisms of antiandrogenic action. Atrazine was consistently inac- tive in our in vitro models of pro- or antiandrogenicity. The effects of endocrine disrupting chemicals on kinase pathways that may phosphorylate AR, thus either activating or deactivat- ing it androgen independently, as well as the role of AR in H295R cells, warrant further investigation.
FUNDING
Natural Sciences and Engineering Research Council (NSERC) of Canada grant (313313-2012 to J.T.S.) and a student bursary from the Fondation Armand-Frappier (to P.R.).
ACKNOWLEDGMENTS
The authors would like to thank Sébastien Alguerola for his technical assistance. The authors have no conflict of inter- est to disclose.
REFERENCES
Balk, S. P., and Knudsen, K. E. (2008). AR, the cell cycle, and pros- tate cancer. Nucl. Recept. Signal. 6, e001.
Beck, V., Reiter, E., and Jungbauer, A. (2008). Androgen receptor transactivation assay using green fluorescent protein as a re- porter. Anal. Biochem. 373, 263-271.
Belloni, V., Dessi-Fulgheri, F., Zaccaroni, M., Di Consiglio, E., De Angelis, G., Testai, E., Santochirico, M., Alleva, E., and Santucci, D. (2011). Early exposure to low doses of atrazine affects behavior in juvenile and adult CD1 mice. Toxicology 279, 19-26.
Brevini, T. A., Zanetto, S. B., and Cillo, F. (2005). Effects of endo- crine disruptors on developmental and reproductive func- tions. Curr. Drug Targets Immune, Endocr. Metabol. Disord. 5, 1-10.
Canton, R. F., Sanderson, J. T., Nijmeijer, S., Bergman, A., Letcher, R. J., and van den Berg, M. (2006). In vitro effects of bromi- nated flame retardants and metabolites on CYP17 catalytic activity: a novel mechanism of action? Toxicol. Appl. Pharmacol. 216, 274-281.
Chen, S., Xu, Y., Yuan, X., Bubley, G. J., and Balk, S. P. (2006). Androgen receptor phosphorylation and stabilization in prostate cancer by cyclin-dependent kinase 1. Proc. Natl Acad. Sci. U. S. A. 103, 15969-15974.
El Sheikh Saad, H., Toullec, A., Vacher, S., Pocard, M., Bieche, I., and Perrot-Applanat, M. (2013). In utero and lactational expo- sure to vinclozolin and genistein induces genomic changes in the rat mammary gland. J. Endocrinol. 216, 245-263.
Gasmi, J., and Sanderson, J. T. (2010). Growth inhibitory, antian- drogenic, and pro-apoptotic effects of punicic acid in LNCaP human prostate cancer cells. J. Agric. Food Chem. 58, 12149-12156.
Gazdar, A. F., Oie, H. K., Shackleton, C. H., Chen, T. R., Triche, T. J., Myers, C. E., Chrousos, G. P., Brennan, M. F., Stein, C. A., and La Rocca, R. V. (1990). Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res. 50, 5488-5496.
Gioeli, D. (2005). Signal transduction in prostate cancer progres- sion. Clin. Sci. (Lond) 108, 293-308.
Gioeli, D., Ficarro, S. B., Kwiek, J. J., Aaronson, D., Hancock, M., Catling, A. D., White, F. M., Christian, R. E., Settlage, R. E., Shabanowitz, J., et al. (2002). Androgen receptor phosphoryla- tion. Regulation and identification of the phosphorylation sites. J. Biol. Chem. 277, 29304-29314.
Gioeli, D., and Paschal, B. M. (2012). Post-translational modifi- cation of the androgen receptor. Mol. Cell Endocrinol. 352, 70-78.
Gordon, V., Bhadel, S., Wunderlich, W., Zhang, J., Ficarro, S. B., Mollah, S. A., Shabanowitz, J., Hunt, D. F., Xenarios, I., Hahn, W. C., et al. (2010). CDK9 regulates AR promoter selectivity and cell growth through serine 81 phosphorylation. Mol. Endocrinol. 24, 2267-2280.
Gray, L. E., Jr, Ostby, J., Monosson, E., and Kelce, W. R. (1999). Environmental antiandrogens: low doses of the fungicide vinclozolin alter sexual differentiation of the male rat. Toxicol. Ind. Health 15, 48-64.
Gray, L. E., Jr, Ostby, J. S., and Kelce, W. R. (1994). Developmental effects of an environmental antiandrogen: the fungicide vin- clozolin alters sex differentiation of the male rat. Toxicol. Appl. Pharmacol. 129, 46-52.
Hayes, T., Haston, K., Tsui, M., Hoang, A., Haeffele, C., and Vonk, A. (2002). Herbicides: feminization of male frogs in the wild. Nature 419, 895-896.
Hilscherova, K., Jones, P. D., Gracia, T., Newsted, J. L., Zhang, X., Sanderson, J. T., Yu, R. M., Wu, R. S., and Giesy, J. P. (2004). Assessment of the effects of chemicals on the expression of ten steroidogenic genes in the H295R cell line using real-time PCR. Toxicol. Sci. 81, 78-89.
Horoszewicz, J. S., Leong, S. S., Kawinski, E., Karr, J. P., Rosenthal, H., Chu, T. M., Mirand, E. A., and Murphy, G. P. (1983). LNCaP model of human prostatic carcinoma. Cancer Res. 43, 1809-1818.
Hsu, F. N., Chen, M. C., Chiang, M. C., Lin, E., Lee, Y. T., Huang, P. H., Lee, G. S., and Lin, H. (2011). Regulation of androgen
Downloaded from http://toxsci.oxfordjournals.org/ at Carleton University on December 1, 2014
receptor and prostate cancer growth by cyclin-dependent ki- nase 5. J. Biol. Chem. 286, 33141-33149.
Kavlock, R., and Cummings, A. (2005). Mode of action: inhibition of androgen receptor function-vinclozolin-induced malfor- mations in reproductive development. Crit. Rev. Toxicol. 35, 721-726.
Kelce, W. R., Lambright, C. R., Gray, L. E., Jr, and Roberts, K. P. (1997). Vinclozolin and p,p’-DDE alter androgen-dependent gene expression: in vivo confirmation of an androgen recep- tor-mediated mechanism. Toxicol. Appl. Pharmacol. 142, 192-200.
Kolle, S. N., Kamp, H. G., Huener, H. A., Knickel, J., Verlohner, A., Woitkowiak, C., Landsiedel, R., and van Ravenzwaay, B. (2010). In house validation of recombinant yeast estrogen and androgen receptor agonist and antagonist screening as- says. Toxicol. in vitro 24, 2030-2040.
Koryakina, Y., Ta, H. Q., and Gioeli, D. (2014). Androgen receptor phosphorylation: biological context and functional conse- quences. Endocr. Relat. Cancer 21, T131-T145.
Kubota, K., Ohsako, S., Kurosawa, S., Takeda, K., Qing, W., Sakaue, M., Kawakami, T., Ishimura, R., and Tohyama, C. (2003). Effects of vinclozolin administration on sperm pro- duction and testosterone biosynthetic pathway in adult male rat. J. Reprod. Dev. 49, 403-412.
Mulholland, D. J., Dedhar, S., Coetzee, G. A., and Nelson, C. C. (2005). Interaction of nuclear receptors with the Wnt/beta- catenin/Tcf signaling axis: Wnt you like to know? Endocr. Rev. 26, 898-915.
Ohlsson, A., Ulleras, E., and Oskarsson, A. (2009). A biphasic ef- fect of the fungicide prochloraz on aldosterone, but not corti- sol, secretion in human adrenal H295R cells-underlying mechanisms. Toxicol. Lett. 191, 174-180.
Rivest, P., Renaud, M., and Sanderson, J. T. (2011). Proliferative and androgenic effects of indirubin derivatives in LNCaP hu- man prostate cancer cells at sub-apoptotic concentrations. Chem. Biol. Interact. 189, 177-185.
Roy, J. R., Chakraborty, S., and Chakraborty, T. R. (2009). Estrogen-like endocrine disrupting chemicals affecting puberty in humans-a review. Med. Sci. Monit. 15, RA137-145. Sanderson, J. T. (2006). The steroid hormone biosynthesis path- way as a target for endocrine-disrupting chemicals. Toxicol. Sci. 94, 3-21.
Sanderson, J. T., Clabault, H., Patton, C., Lassalle-Claux, G., Jean- Francois, J., Pare, A. F., Hebert, M. J., Surette, M. E., and Touaibia, M. (2013). Antiproliferative, antiandrogenic and cy- totoxic effects of novel caffeic acid derivatives in LNCaP hu- man androgen-dependent prostate cancer cells. Bioorg. Med. Chem. 21, 7182-7193.
Sanderson, J. T., Seinen, W., Giesy, J. P., and 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, 121-127.
Schug, T. T., Janesick, A., Blumberg, B., and Heindel, J. J. (2011). Endocrine disrupting chemicals and disease susceptibility. J. Steroid Biochem. Mol. Biol. 127, 204-215.
Scippo, M. L., Argiris, C., Van De Weerdt, C., Muller, M., Willemsen, P., Martial, J., and Maghuin-Rogister, G. (2004). Recombinant human estrogen, androgen and progesterone receptors for detection of potential endocrine disruptors. Anal. Bioanal. Chem. 378, 664-669.
Sewer, M. B., and Jagarlapudi, S. (2009). Complex assembly on the human CYP17 promoter. Mol. Cell Endocrinol. 300, 109-114.
Soderstrom, T. G., Bjelfman, C., Brekkan, E., Ask, B., Egevad, L., Norlen, B. J., and Rane, A. (2001). Messenger ribonucleic acid levels of steroid 5 alpha-reductase 2 in human prostate predict the enzyme activity. J. Clin. Endocrinol. Metab. 86, 855-858.
Sohoni, P., and Sumpter, J. P. (1998). Several environmental oestrogens are also anti-androgens. J. Endocrinol. 158, 327-339.
Staels, B., Hum, D. W., and Miller, W. L. (1993). Regulation of ste- roidogenesis in NCI-H295 cells: a cellular model of the hu- man fetal adrenal. Mol. Endocrinol. 7, 423-433.
van Steenbrugge, G. J., Groen, M., van Dongen, J. W., Bolt, J., van der Korput, H., Trapman, J., Hasenson, M., and Horoszewicz, J. (1989). The human prostatic carcinoma cell line LNCaP and its derivatives. An overview. Urol. Res. 17, 71-77.
Vinggaard, A. M., Christiansen, S., Laier, P., Poulsen, M. E., Breinholt, V., Jarfelt, K., Jacobsen, H., Dalgaard, M., Nellemann, C., and Hass, U. (2005). Perinatal exposure to the fungicide prochloraz feminizes the male rat offspring. Toxicol. Sci. 85, 886-897.
Vinggaard, A. M., Nellemann, C., Dalgaard, M., Jorgensen, E. B., and Andersen, H. R. (2002). Antiandrogenic effects in vitro and in vivo of the fungicide prochloraz. Toxicol. Sci. 69, 344-353.
Vinggaard, A. M., Niemela, J., Wedebye, E. B., and Jensen, G. E. (2008). Screening of 397 chemicals and development of a quantitative structure-activity relationship model for androgen receptor antagonism. Chem. Res. Toxicol. 21, 813-823.
Wilhelms, K. W., Fitzpatrick, K. F., Scanes, C. G., and Anderson, L. L. (2006). In ovo exposure to a triazine herbicide: effects of at- razine on circulating reproductive hormones and gonadal histology in young Japanese quail. Arch. Environ. Contam. Toxicol. 51, 117-122.
Wong, C., Kelce, W. R., Sar, M., and Wilson, E. M. (1995). Androgen receptor antagonist versus agonist activities of the fungicide vinclozolin relative to hydroxyflutamide. J. Biol. Chem. 270, 19998-20003.
Yamada, T., Sumida, K., Saito, K., Ueda, S., Yabushita, S., Sukata, T., Kawamura, S., Okuno, Y., and Seki, T. (2005). Functional genomics may allow accurate categorization of the benz- imidazole fungicide benomyl: lack of ability to act via ste- roid-receptor-mediated mechanisms. Toxicol. Appl. Pharmacol. 205, 11-30.
Yang, C. S., Xin, H. W., Kelley, J. B., Spencer, A., Brautigan, D. L., and Paschal, B. M. (2007). Ligand binding to the androgen re- ceptor induces conformational changes that regulate phos- phatase interactions. Mol. Cell. Biol. 27, 3390-3404.
Yeh, S., Hu, Y. C., Wang, P. H., Xie, C., Xu, Q., Tsai, M. Y., Dong, Z., Wang, R. S., Lee, T. H., and Chang, C. (2003). Abnormal mam- mary gland development and growth retardation in female mice and MCF7 breast cancer cells lacking androgen recep- tor. J. Exp. Med. 198, 1899-1908.
Zhou, Z. X., Lane, M. V., Kemppainen, J. A., French, F. S., and Wilson, E. M. (1995). Specificity of ligand-dependent andro- gen receptor stabilization: receptor domain interactions in- fluence ligand dissociation and receptor stability. Mol. Endocrinol. 9, 208-218.
Downloaded from http://toxsci.oxfordjournals.org/ at Carleton University on December 1, 2014