Environ. Sci. Technol. 2004, 38, 6322-6327
Evaluation of the Model Anti-androgen Flutamide for Assessing the Mechanistic Basis of Responses to an Androgen in the Fathead Minnow (Pimephales promelas) GERALD T. ANKLEY,* DAVID L. DEFOE, MICHAEL D. KAHL, KATHLEEN M. JENSEN, AND ELIZABETH A. MAKYNEN U.S. Environmental Protection Agency, Mid-Continent Ecology Division, Duluth, Minnesota 55804 ANN MIRACLE U.S. Environmental Protection Agency, Ecological Exposure Research Division, Cincinnati, Ohio 45268 PHILLIP HARTIG, L. EARL GRAY, MARY CARDON, AND VICKIE WILSON U.S. Environmental Protection Agency, Reproductive Toxicology Division, Research Triangle Park, North Carolina 27711
In this study, we characterized the effects of flutamide, a model mammalian androgen receptor (AR) antagonist, on endocrine function in the fathead minnow (Pimephales promelas), a small fish species that is widely used for testing endocrine-disrupting chemicals (EDCs). Binding assays with whole cells transiently transfected with cloned fathead minnow AR indicated that flutamide binds competitively to the receptor. However, as is true in mammalian systems, a 2-hydroxylated metabolite of flutamide binds to the AR with a much higher affinity than the parent chemical. Mixture experiments with flutamide and the androgen 17β-trenbolone demonstrated that the antiandrogen effectively blocked trenbolone-induced masculinization (nuptial tubercle production) of female fathead minnows, indicating antagonism of an AR receptor-mediated response in vivo. Conversely, reductions in vitellogenin in trenbolone-exposed females were not blocked by flutamide, suggesting that the vitellogenin response is not directly mediated through the AR. The results of these studies provide data demonstrating the validity of using the fathead minnow as a model species for detecting EDCs that exert toxicity through interactions with the AR.
Introduction There is an international emphasis on development of testing programs for chemicals with the potential to affect reproduction and development in humans and wildlife through disruption of processes controlled by estrogens and androgens (1, 2). Several in vitro and in vivo systems have been * Corresponding author telephone: (218)529-5147; fax: (218)5295003; e-mail:
[email protected]. 6322 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 23, 2004
proposed for identifying endocrine-disrupting chemicals (EDCs), including short-term assays with three small fish modelssfathead minnow, Japanese medaka, and zebrafish (3-5). As part of characterizing and validating these assays, testing is conducted with model EDCs with known or suspected modes/mechanisms of action (MOA). A significant amount of the validation effort with the small fish has focused on chemicals that act as strong or weak agonists of the estrogen receptor (ER; 6-13). Comparatively less is known concerning responses of these model species to EDCs with other MOA, including chemicals that bind to the androgen receptor (AR). Further knowledge of the nature of interactions of xenobiotics with fish AR(s) is critical, as there are many environmental contaminants with the potential to interact with the receptor (14-24). Several natural and synthetic steroids have been used as model AR agonists in fish studies. Synthetic chemicals often are preferable to natural steroids for controlled experimentation because the former can be less prone to metabolism and excretion than hormones normally present in the animals. Historically, the most commonly used synthetic androgen in endocrinological research with fish has been 17R-methyltestosterone. Methyltestosterone has been used for decades to manipulate sex of fish in aquacultural settings (for review, see ref 25) and also has been frequently used in fish studies focused on toxicology of EDCs (9, 13, 26, 27). However, using methyltestosterone as a model androgen for EDC research is problematic because it can produce paradoxical responses indicative both of androgenic and estrogenic MOA (9, 27). This “mixed” response appears to be due to the conversion of methyltestosterone to the potent ER agonist 17R-methylestradiol by CYP19 aromatase, the same enzyme that converts testosterone to 17β-estradiol (28). An alternative model androgen that recently has been used in fish studies is 17β-trenbolone (20, 21). Trenbolone is a highaffinity ligand for the fathead minnow AR and is quite potent with regard to effects on reproductive endocrinology in this species (21). As opposed to methyltestosterone, trenbolone is not a suitable substrate for conversion to an estrogen analogue by CYP19 aromatase. Extensive studies evaluating trenbolone as a model EDC recently have been conducted with the fathead minnow, Japanese medaka, and zebrafish (29). Although trenbolone can affect many aspects of endocrine function (21), two responses that are quite similar in sexually mature females of all the species are masculinization (production of male secondary sex characteristics) and reductions in concentrations of the egg yolk protein precursor, vitellogenin (29). The masculinization response is consistent with activation of the fish AR (30, 31). However, it is uncertain how trenbolone causes effects on vitellogenin because production of the lipoprotein is thought to be controlled primarily through the ER (32). On the basis of studies in mammalian systems, the pharmaceutical flutamide has been proposed as a model AR antagonist for EDC research (33). Flutamide and/or its 2-hydroxylated metabolite bind with comparatively high affinity to mammalian AR(s) in vitro and block expression of AR-mediated responses in vivo (34-36). Flutamide has been used in EDC-oriented studies with fish under the assumption that it operates via the same MOA as in mammals (37-39). Bayley et al. (38) reported that guppies exposed to flutamide during early development were demasculinized, a response consistent with mammalian studies with antiandrogens (15). Jensen et al. (39) found that exposure of reproductively mature fathead minnows to flutamide affected endocrine function in a manner consistent with an AR 10.1021/es040022b Not subject to U.S. copyright. Publ. 2004 Am. Chem.Soc. Published on Web 07/14/2004
antagonist based, again, on studies in mammals. From the perspective of receptor binding, some studies indicate that flutamide can bind to fish (fathead minnow) AR(s) (40), while other studies indicate little, if any, specific binding of flutamide using fish (rainbow trout, goldfish) AR preparations (41). Although these in vivo and in vitro studies with fish are suggestive, they do not establish that flutamide elicits in vivo effects in fish through interactions with the AR. To achieve this, systematic studies (ideally with the same species) are required to (a) characterize binding of flutamide (and potentially active metabolites) to the AR and (b) assess the effects of flutamide on AR-mediated processes in vivo. The objective of the present study was to use two model EDCs, 17β-trenbolone and flutamide, to assess the mechanistic basis of responses of the fathead minnow reproductive endocrine system to the chemicals. In initial experiments, we characterized the binding affinity of flutamide and hydroxyflutamide to the fathead minnow AR. In subsequent experiments, we exposed fish to joint mixtures of flutamide and trenbolone to evaluate interactions between the two chemicals with respect to suppression of vitellogenin mRNA in and masculinization of female fathead minnows.
Materials and Methods Competitive Binding Assays. Although EDC studies typically use flutamide for in vivo exposures, in mammals it is the 2-hydroxylated metabolite most likely responsible for antiandrogenic properties of the chemical (36). A previous study with the fathead minnow documented, via mass spectroscopy, the occurrence of 2-hydroxyflutamide in fish exposed to flutamide (39). Hence, competitive binding assays using cloned fathead minnow AR expressed in an immortal cell line were conducted using both flutamide and hydroxyflutamide. Monolayer COS-1 cells (transformed African green monkey kidney cell line, ATCC, Rockville, MD) were maintained at 37 °C with 100% humidity and 5% CO2 in a commercially available growth medium (DMEM; Invitrogen/ GIBCO, Grand Island, NY) supplemented with 100 U penicillin, 100 mg/L streptomycin, 0.25 mg/L amphotericin B, 5.9 g/L HEPES buffer (all from Invitrogen/GIBCO), and 10% fetal bovine serum (FBS; Hyclone, Logan, UT). When confluent, the cells were suspended by treatment with trypsin and subcultured at 1:10 dilutions. For binding assays, the COS cells were plated at 200 000 cells/well in 12-well trays in growth medium. The next day the cells were transiently transfected with 0.5 µg of pFAR, a plasmid containing the fathead minnow AR (42), using the DEAE-dextran (diethylaminoethyl dextran) method of Wong et al. (43). Twenty four hours later, the cells were exposed to 0.5 nM [3H]R1881 (17R-[3H]-methyltrienolone; New England Nuclear Life Sciences Products, Inc., Boston, MA) along with varying amounts of test chemical in serum-free and phenol red-free DMEM (42). A 100-fold molar excess of unlabeled R1881 was used to determine nonspecific binding. The 2-hydroxyflutamide was provided by R. O. Neri (Schering-Plough Corp., Bloomfield, NJ), and flutamide was purchased from Sigma Chemical Co. (St. Louis, MO; >99% purity). Hydroxyflutamide was tested at concentrations ranging from 1 nM to 100 µM, and flutamide was tested from 10 nM to 300 µM. After adding the test materials, cells were incubated for 2 h at 37 °C, washed with phosphate-buffered saline, and lysed with 200 µL of ZAP (0.13 M ethyldimethylhexadecylammonium bromide with 3% glacial acetic acid). Liquid scintillation counting was used to determine radioactivity in the lysate. Specific binding was calculated by subtracting nonspecific binding from the total binding values, and IC50 (concentration required to inhibit 50% binding of the R1881) values were calculated using logistic regression (42).
Whole Animal Exposures. For the in vivo portion of this study, we used female fathead minnows from an established on-site culture unit at the Environmental Protection Agency facility in Duluth, MN. Flutamide (99% purity) and trenbolone (98% purity) were both from Sigma Chemical Co. Treatments consisted of (a) clean water (control), (b) flutamide (400 µg/ L), (c) trenbolone (0.5 µg/L), and (d) flutamide and trenbolone in combination (400 and 0.5 µg/L, respectively). Treatment concentrations for these exposures were derived from effects data from previous experiments in which effects of the two chemicals on fathead minnow reproductive endocrinology were assessed (21, 39). Solvent-free stock solutions of flutamide and trenbolone were generated as described in those previous papers. Sexually mature female fish (7-9 month old) came from maturation tanks in the culture facility in which they had been housed with males and allowed to spawn. The mean (SD, n ) 100) wet weight of the fish was 2.31 (0.75) g. The animals were placed in 5 L of water in glass test tanks 2 d prior to initiating chemical exposures. Duplicate tanks for each treatment contained 12 or 13 fish (a total of 25 fish per treatment). Exposures were conducted in a constant flow (ca. 60 mL/min) of Lake Superior water or test chemical dissolved in Lake Superior water. Fish were maintained at 25 ( 1.5 °C under a 16:8 light:dark photoperiod and fed brine shrimp ad libitum twice daily. The mean (SD, n ) 4) pH of the test system was 7.6 (0.09). Dissolved oxygen (DO) was measured weekly in each of the tanks; the mean (SD, n ) 32) DO concentration for the test was 6.29 (0.5) mg/L. After 1 d of exposure, five fish were removed from each treatment for collection of liver for vitellogenin mRNA measurements (described below). The animals were killed by placing them in 100 mg of MS-222/L buffered with 200 mg of NaHCO3/L. Livers were placed immediately into a solution of RNA Later (Ambion, Austin, TX) at 4 °C for 24 h and then stored at -20 °C until analyzed. After 1 week of exposure to the test chemicals, the fish were lightly anesthetized with the buffered MS-222 (ca. 2 min exposure; 44) and examined under a magnifying glass to determine degree of masculinization based on the number and degree of expression of nuptial tubercles on the heads of the fish (45). Nuptial tubercles are normally only found in reproductively active male fathead minnows but can be induced in females exposed to androgenic steroids, including trenbolone (9, 21, 46). After assessment, the fish were allowed to recover from anesthesia in clean water and replaced in the test tanks. Previous studies from our laboratory have shown that this handling regime should not adversely affect basic reproductive endocrinology of the fathead minnow (44). After 2 weeks of chemical exposure the fish were again evaluated for tubercle development; a subset of 10 of these animals from each treatment (five per tank) were maintained in clean water in the test system to determine persistence of the trenboloneinduced masculinization. The remaining fish were sacrificed, and livers were collected and stored as described above. Fish held in the clean water were anesthetized and assessed for tubercles 1 and 2 weeks after cessation of the chemical exposures. In conjunction with the final evaluation of the fish, livers from all remaining animals were collected and stored as previously described. During the assay, water concentrations of both flutamide and trenbolone were measured weekly in all of the test tanks from each of the four treatments using methods described in detail elsewhere (21, 39). Briefly, both chemicals were analyzed using reverse-phase high-pressure liquid chromatography with diode array detection for flutamide and fluorescence detection for trenbolone. Measured concentrations of the two chemicals were stable over the course of the test and deviated from the nominal (target) concentrations by less than 12% (data not shown). The detection limits for VOL. 38, NO. 23, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6323
flutamide and trenbolone were 100 and 0.2 µg/L, respectively; neither analyte was detected in control water. Vitellogenin mRNA Measurements. Changes in vitellogenin status can be determined through measurement either of transcribed mRNA or the translated protein (47, 48). To maximize sensitivity of the determination, particularly with respect to detecting rapid (1 d) changes in responses of treated fish, we chose to measure vitellogenin mRNA. Differential expression in liver of vitellogenin mRNA in the female fathead minnows was determined using semiquantitative reverse transcription-polymerase chain reaction (QRT-PCR; 49). Total RNA was isolated from liver using Tri Reagent (Molecular Research Center, Cincinnati, OH), and relative concentrations were determined by UV spectrophotometry. Primers for amplification of fathead minnow vitellogenin (GenBank Accession No. AF130354; 47) were designed using Oligo 6.65 software (Molecular Biology Insights, Cascade, CO). Primer performance and amplicon validations were achieved through examination of melt curve profiles to determine the presence of a single amplified product with subsequent isolation and DNA sequencing to confirm the identity of the amplicon as fathead minnow vitellogenin (data not shown). For normalization, amplification of 18S rRNA sequences were performed using Ambion’s Universal 18S Primer Mix (Austin, TX). The use of 18S has been documented as a reasonable approach (50-52) to normalization especially when no gene sequences identified as housekeeping genes in the fathead minnow were available. Briefly, 1 µg of total RNA from treated and control samples was reverse transcribed with 50 µM random hexamers and 50 µM oligo d(T)16VN, using reagents and the protocol recommended by Applied Biosystems (Foster City, CA). One-tenth of the cDNA was used for each PCR reaction along with a SYBR green master mix (MJ Research, Boston, MA) and 10 pmol of primers (data not shown). Cycling was carried out with 40 cycles of 95 °C for 20 s, 60 °C for 20 s, and 72 °C for 10 s. Target fluorescence was measured at the end of the 72 °C step for each cycle. Each gene assay included a standard curve of purified, template-specific cDNA in five serial dilutions for setting the cycle threshold. Amplification experiments were performed in triplicate. The mean cDNA expression levels for all samples were normalized to mean expression levels for 18S. Each reaction was tested for PCR efficiency using LinRegPCR software and the efficiency calculations used subsequently in assigning expression values for each sample using the eff∆Ct method (53), which has been demonstrated to be an effective method for calculating concentration estimates of mRNA using SYBR Green detection. The relative abundance of normalized vitellogenin mRNA for each treatment group was compared with timematched controls using a t-test statistic.
Results Figure 1 shows competitive binding of flutamide and hydroxyflutamide, relative to [3H]R1881, to the fathead minnow AR. In the assay system used for this study, a vector carrying the fathead minnow AR gene is introduced into the COS cells where the receptor is then expressed. Hence, as opposed to heterogeneous tissue preparations from animals, the system is devoid of other steroid receptors that might compete for the ligand and confound the results of the assay. In this system, 2-hydroxyflutamide was about 1.5 log units more potent (IC50 ) 348 nM) than flutamide (IC50 ) 5454 nM) in competing with R1881 binding to the AR (Figure 1). There was no mortality of any of the fish during the study. All fish behaved and fed normally. After 1 week of exposure to the trenbolone there was a significant masculinization of the females, with nuptial tubercles observed in 13 of 20 exposed fish (mean [SD]; 3.05[0.9] tubercles/fish). Consistent with previous studies, there were no nuptial tubercles 6324
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 23, 2004
FIGURE 1. Competitive binding of flutamide (9) and 2-hydroxyflutamide (b) with [3H]R1881 to the cloned fathead minnow androgen receptor expressed in COS cells. Data for each curve are mean values from four separate experiments. Bars indicate the standard error of the mean for data from the four determinations.
FIGURE 2. Temporal effects of exposure to flutamide, trenbolone, and flutamide plus trenbolone on hepatic vitellogenin mRNA in female fathead minnows. Day 1 samples represent the mean of five fish for each treatment with standard error bars. Day 14 and postexposure data are given as the mean of 10 fish for each treatment with standard error bars. Asterisks indicate significant (p e 0.05) differences from respective control values within that sampling period.
observed in control females or those exposed to flutamide. As hypothesized based on the anticipated MOA, there was no evidence of tubercles in female fish exposed to trenbolone in conjunction with flutamide. After 2 weeks, tubercles were observed in 19 of 20 females exposed to the androgen alone. The tubercle score had increased significantly (unpaired t-test, p e 0.05) in these fish (10.9[0.6] tubercles/fish) as compared to 1 week of exposure; however, joint treatment with flutamide continued to effectively block any evidence of masculinization caused by the trenbolone. The number of tubercles in females that had been exposed to trenbolone decreased comparatively rapidly after the animals had been placed in clean water; the mean (SD) number of tubercles in fish decreased significantly to 4.8 (1.1) and 3.6 (1.7), respectively, after 1 and 2 weeks in clean water. Expression of vitellogenin in female fathead minnows treated with trenbolone was significantly (p e 0.05) downregulated compared to the matching controls after only 1 d of treatment, while the combination of trenbolone and flutamide had a slightly less pronounced down-regulatory effect (Figure 2). Flutamide treatment alone at 1 d had no significant effect on vitellogenin mRNA expression. A general suppression of vitellogenin mRNA in trenbolone-exposed fish persisted by 14 d, although the treatments were not
significantly different from the matching control (Figure 2). At 14 d postexposure, all three treatment groups displayed a significant down-regulation of vitellogenin mRNA (Figure 2).
Discussion Small fish, such as the fathead minnow, have become important models for assessing the effects of EDCs in screening and testing programs throughout the world (for review, see ref 5). To date, there has been an emphasis on characterizing responses of fish to estrogenic EDCs (6-13). Although estrogens clearly are important both from human and ecological health perspectives, there is accumulating evidence that chemicals that affect processes controlled by the AR also are of concern. For example, there is evidence of the occurrence of environmental contaminants that act as agonists of the AR in complex mixtures associated with pulp and paper mill effluents and animal feeding operations (18, 19, 21, 22, 24). There also are a number of environmentally relevant pesticides that act as anti-androgens in mammals and probably also in fish (14-17, 23, 38, 40). Given these types of observations, it is critical that test systems developed to detect EDCs effectively capture not only chemicals that interact with the ER but also those that are antagonists or agonists of the AR. The purpose of this study, therefore, was to better characterize responses of the fathead minnow to EDCs that bind to the AR. Previous studies with fish have revealed conflicting results as to whether flutamide binds to AR(s). There was an indication of a low level of binding of flutamide to fathead minnow AR in heterogeneous tissue preparations (40); however, comparable studies in the rainbow trout and goldfish indicated no competitive binding of flutamide to the AR (41). We are aware of no studies with fish in which binding of 2-hydroxyflutamide to AR(s) has been assessed. In the present study, competitive binding data generated using the fathead minnow AR indicated that both flutamide and hydroxyflutamide bound to the receptor, but the metabolite bound with a much higher affinity than the parent compound. Fifty percent of the available receptors were bound with hydroxyflutamide at a concentration more than an order of magnitude lower than that required for binding of flutamide to a similar number of receptors. Studies in mammals suggest that 2-hydroxyflutamide is primarily responsible for the in vivo anti-androgenic properties of flutamide (34-36, 54). This is consistent with the results of Cardon et al. (55), who found that hydroxyflutamide bound with a higher affinity than flutamide to human AR expressed in the COS cells using the same protocol as in this study. Jensen et al. (39) conducted a study in which effects of exposure to flutamide (via water) on fathead minnow reproduction were assessed in a 21-d assay. Flutamide significantly decreased fecundity (egg production) in the fish and affected a number of other aspects of reproductive endocrine function in a manner consistent with exposure to an AR antagonist (39). Significantly, tissue extracts of fish exposed to flutamide in that study contained 2-hydroxyflutamide concentrations of approximately 10% of that of the parent compound. This observation in conjunction with the whole cell competitive binding data suggests that, as is true in mammals, it is the hydroxylated metabolite of flutamide responsible for anti-androgenic properties of the chemical. Thus, in species representative of two different vertebrate classes, hydroxyflutamide has been shown to have a greater affinity for the AR than flutamide. In related studies, it has been demonstrated that a number of steroidal androgens bind to the fathead minnow AR in the same relative manner as to the human AR, albeit with minor differences in potency (42). Collectively, these data suggest that chemicals that bind
to the AR in one vertebrate species are likely to bind to the AR in other vertebrates. This has important implications for the utility of receptor-based assays as screens for endocrineactive chemicals. That is, it may not be necessary to utilize AR(s) from multiple vertebrate phyla when screening for potential EDCs. Although the actual phenotype generated by xenobiotic modulation of the AR in the whole animal will be species-specific, demonstrated conservation of the endocrine system across vertebrate species, at least with respect to binding of chemicals to the AR, would seem to serve as a reasonable basis for initial extrapolation across species. Additional studies comparing and contrasting endocrine systems and modulation across species will allow for further refinement of extrapolation among species and chemicals. In studies conducted more than 30 yr ago, Smith (46) found that steroidal androgens stimulated production of nuptial tubercles in female fathead minnows. More recent research with this species has better characterized the masculinization response within the framework of formal EDC testing, using the synthetic steroids methyltestosterone and trenbolone (9, 21). It is widely accepted that expression of male secondary sexual characteristics in fish is under control of the AR (30, 31). In the fathead minnow, this assumption is directly supported by observed high-affinity binding of trenbolone to the AR in the COS whole cell binding assay (21, 42). The fact that flutamide (or hydroxyflutamide) effectively blocked expression of male secondary sex characteristics in trenbolone-exposed female fathead minnows provides powerful evidence of the anti-androgenic activity of flutamide in vivo. Ankley et al. (21) conducted a study in which reproductively active fathead minnows were exposed to trenbolone for 21 d; in addition to morphological masculinization of the females, there were significant reductions in their plasma concentrations of testosterone, estradiol, and vitellogenin protein. More recent studies also have shown significant reductions in vitellogenin protein concentration in sexually mature female fathead minnow, medaka, and zebrafish exposed to trenbolone for 21 d (29). The mechanism underlying this response is uncertain as it is generally accepted that vitellogenin is under control of the ER, not the AR, in oviporous animals (32). Data from the present study supports previous work (21, 29) in that significant reductions in vitellogenin mRNA were measurable in animals exposed to trenbolone within only 1 d of exposure. There was some degree of temporal variation in vitellogenin mRNA levels in controls reflecting, perhaps, cyclic regulation of vitellogenesis in the fish. However, after 14 d of exposure vitellogenin mRNA still was lower (albeit not significantly so) in trenbolonetreated fish than in matched controls. At 14-d postexposure to the androgen, hepatic vitellogenin mRNA was again significantly suppressed. Overall, co-treatment of the female fathead minnows with flutamide did not seem to ameliorate trenbolone-induced reductions in vitellogenin mRNA at any of the three sampling periods. This suggests that the vitellogenin reductions associated with trenbolone exposure probably do not occur via a mechanism(s) directly related to stimulation or inhibition of the AR. There are different possible explanations for the reductions in vitellogenin caused by trenbolone. First, it is possible that concentrations of estradiol were sufficiently reduced by trenbolone exposure to decrease stimulation of the ER and subsequent vitellogenin production (21). In that study, we speculated that the depressed estradiol concentrations occurred as a result of decreased concentrations of testosterone, the direct metabolic precursor of estradiol, caused perhaps by feedback inhibition of androgen production in the fish. A second possible explanation for the observed decreases of vitellogenin mRNA in females exposed to the VOL. 38, NO. 23, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6325
synthetic androgen is that trenbolone acts as an antagonist of the fathead minnow ER. In support of this hypothesis, recent research has demonstrated that 17β-trenbolone is a competitive inhibitor of binding of estradiol to the fathead minnow ER (T. Henry, U.S. Environmental Protection Agency, personal communication). Flutamide alone caused a decrease in vitellogenin mRNA expression at 14 d postexposure. In a 21-d reproduction assay, flutamide slightly increased levels of vitellogenin protein in plasma of female fathead minnows exposed to a flutamide water concentration about 50% higher than used in the present study (39). Hence, the two studies are somewhat at odds with one another with respect to flutamide effects on vitellogenin; however, it is difficult to directly compare them because of differences in the determination of vitellogenin status (mRNA vs protein), reproductive state of the animals (suppressed versus active spawning), and of course, temporal aspects of the flutamide exposure. Further work is needed to ascertain the endocrinological basis of flutamide-induced increases or decreases in vitellogenin in fish. This could have consequences in terms of using changes in vitellogenin to detect exposure to/effects of anti-androgens versus other classes of endocrine-disrupting chemicals. In summary, this study expands the current understanding of interactions of antagonists and agonists with AR(s) in fish. This type of information is critical not only to interpreting the results of tests with potential EDCs in a lab setting but also to understanding existing impacts of androgen agonists and antagonists on fish exposed to complex mixtures in the field.
Acknowledgments We thank Drs. Dan Villenueve and Tala Henry for comments on an earlier version of the manuscript. Diane Spehar and Roger LePage assisted in manuscript preparation. This paper has been reviewed in accordance with official EPA policy. Mention of products or trade names does not constitute endorsement by the Federal government.
Literature Cited (1) U.S. Environmental Protection Agency. Endocrine disrupter screening and testing advisory committee (EDSTAC) report; Office of Prevention, Pesticides and Toxic Substances: Washington, DC, 1998. (2) Huet, M.-C. Ecotoxicology 2000, 9, 77-84. (3) Organization for Economic Cooperation and Development. Report of the OECD expert consultation on testing in fish-EDF1, London, October; OECD: Paris, 1999. (4) Organization for Economic Cooperation and Development. Report on the OECD expert consultation on testing in fish-EDF2, Tokyo, March; OECD: Paris, 2000. (5) Ankley, G. T.; Johnson, R. D. Inst. Lab. Anim. Res. J. (in press). (6) Kramer, V. J.; Miles-Richardson, S.; Pierens, S. L.; Giesy, J. P. Aquat. Toxicol. 1998, 40, 335-360. (7) Nimrod, A. C.; Benson, W. H. Aquat. Toxicol. 1998, 44, 141-156. (8) Harries, J. E.; Runnalls, T.; Hill, E.; Harris, C. A.; Maddix, S.; Sumpter, J. P.; Tyler, C. R. Environ. Sci. Technol. 2000, 34, 30033011. (9) Ankley, G. T.; Jensen, K. M.; Kahl, M. D.; Korte, J. J.; Makynen, E. A. Environ. Toxicol. Chem. 2001, 20, 1276-1290. (10) Seki, M.; Yokota, H.; Matsubara, H.; Tsuruda, Y.; Maeda, M.; Tadokoro, H.; Kobayashi, K. Environ. Toxicol. Chem. 2002, 21, 1692-1698. (11) Seki, M.; Yokota, H.; Matsubara, H.; Maeda, M.; Tadokoro, H.; Kobayashi, K. Environ. Toxicol. Chem. 2003, 22, 1487-1496. (12) Segner, H.; Navas, J. M.; Scha¨fers, C.; Wenzel, A. Ecotoxicol. Environ. Saf. 2003, 54, 315-322. (13) O ¨ rn, S.; Holbech, H.; Madsen, T. H.; Norrgren, L.; Petersen, G. L. Aquat. Toxicol. 2003, 65, 397-411. (14) Gray, L. E., Jr.; Ostby, J. S.; Kelce, W. R. Toxicol. Appl. Pharmacol. 1994, 129, 46-52. (15) Gray, L. E., Jr.; Wolf, C.; Lambright, C.; Mann, P.; Price, M.; Cooper, R. L.; Ostby, J. J. Toxicol. Ind. Health 1999, 15, 94-118. (16) Kelce, W. R.; Wilson, E. M. J. Mol. Med. 1997, 75, 198-207. 6326
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 23, 2004
(17) Monosson, E.; Kelce, W. R.; Mac, M.; Gray, L. E. In Chemically Induced Alterations in Functional Development and Reproduction of Fishes; Rolland, R. M., Gilbertson, M., Peterson, R. E., Eds.; SETAC: Pensacola, FL, 1997; pp 53-60. (18) Jenkins, R.; Angus, R. A.; McNatt, H.; Howell, W. M.; Kemppainen, J. A.; Kirk, M.; Wilson, E. M. Environ. Toxicol. Chem. 2001, 20, 1325-1331. (19) Durhan, E. J.; Lambright, C.; Wilson, V.; Butterworth, B. C.; Kuehl, D. W.; Orlando, E. F.; Guillette, L. J., Jr.; Gray, L. E.; Ankley, G. T. Environ. Toxicol. Chem. 2002, 21, 1973-1976. (20) Peterson, B. N.; Foran, C. M.; Benson, W. H. Abstracts of the 22nd Annual Meeting of Society of Environmental Toxicology and Chemistry, Baltimore, MD, 2001; p 333. (21) Ankley, G. T.; Jensen, K. M.; Makynen, E. A.; Kahl, M. D.; Korte, J. J.; Hornung, M. W.; Henry, T. R.; Denny, J. S.; Leino, R. L.; Wilson, V. S.; Cardon, M. C.; Hartig, P. C.; Gray, L. E. Environ. Toxicol. Chem. 2003, 22, 1350-1360. (22) Hewitt, L. M.; Pryce, A. C.; Parrott, J. L,; Marlatt, V.; Wood, C.; Oakes, K.; Van Der Kraak, G. J. Environ. Toxicol. Chem. 2003, 22, 2890-2897. (23) Kiparissis, Y.; Metcalfe, T. L.; Balch, G. C.; Metcalfe, C. D. Aquat. Toxicol. 2003, 63, 319-403. (24) Orlando, E. F.; Kolok, A. S.; Binzcik, G. A.; Gates, J. L.; Horton, M. K.; Lambright, C. S.; Gary, L. E.; Soto, A. M.; Guillette, L. J. Environ. Health Perspect. 2004, 112, 353-358. (25) Hunter, G. A.; Donaldson, E. M. In Fish Physiology; Hoar, W. S., Randall, D. J., Donaldson, E. M., Eds.; Academic: New York, 1983; Vol. 9B, pp 223-303. (26) Papoulias, D. M.; Noltie, D. B.; Tillitt, D. E. Mar. Environ. Res. 2000, 51, 181-184. (27) Zerulla, M.; La¨nge, R.; Steger-Hartmann, T.; Panter, G.; Hutchinson, T.; Dietrich, D. R. Toxicol. Lett. 2002, 131, 51-63. (28) Hornung, M. W.; Jensen, K. M.; Korte, J. J.; Kahl, M. D.; Durhan, E. J.; Denny, J. S.; Henry, T. R.; Ankley, G. T. Aquat. Toxicol. 2004, 66, 15-23. (29) Organization for Economic Cooperation and Development. OECD draft report of the initial work towards the validation of the fish screening assay for the detection of endocrine active substances: Phase 1A; OECD: Paris, 2004. (30) Borg, B. Comp. Biochem. Physiol., Part C 1994, 109, 219-245. (31) Kime, D. E. Endocrine Disruption in Fish; Kluwer Academic: Norwell, MA, 1998. (32) Shapiro, D. CRC Crit. Rev. Biochem. 1982, 12, 187-203. (33) Ankley, G. T.; Black, M. C.; Garric, J.; Hutchinson, T.; Iguchi, T. In Science for Assessing the Impacts of Human Pharmaceuticals on Aquatic Systems; Williams, R. T., et al., Eds.; SETAC: Pensacola, FL, 2004 (in press). (34) Katchen, B.; Buxbaum, S. J. Clin. Endocrinol. Metab. 1975, 41, 373-379. (35) Neri, R.; Peets, E.; Watnick, A. Biochem. Soc. Trans. 1979, 7, 565-569. (36) Simard, J.; Singh, S. M.; Labrie, F. Urology 1997, 49, 580-589. (37) Hagino, S.; Kagoshima, M.; Asida, S. Environ. Sci. 2001, 8, 7587. (38) Bayley, M.; Junge, M.; Baatrup, E. Aquat. Toxicol. 2002, 56, 227239. (39) Jensen, K. M.; Kahl, M. D.; Makynen, E. A.; Korte, J. J.; Leino, R. L.; Butterworth, B. C.; Ankley, G. T. Aquat. Toxicol. (in press). (40) Makynen, E. A.; Kahl, M. D.; Jensen, K. M.; Tietge, J. E.; Wells, K. L.; Van Der Kraak, G.; Ankley, G. T. Aquat. Toxicol. 2000, 48, 461-475. (41) Wells, K.; Van Der Kraak, G. Environ. Toxicol. Chem. 2000, 19, 2059-2065. (42) Wilson, V.; Cardon, M.; Thornton, J.; Ankley, G. T.; Korte, J. J.; Welch, J.; Gray, L. E.; Hartig, P. Environ. Sci. Technol. (in press). (43) Wong, C.; Kelce, W. R.; Madhabanada, S.; Wilson, E. J. Biol. Chem. 1995, 270, 19998-20003. (44) Kahl, M. D.; Jensen, K. M.; Korte, J. J.; Ankley, G. T. J. Fish Biol. 2001, 59, 515-523. (45) Jensen, K. M.; Korte, J. J.; Kahl, M. D.; Ankley, G. T. Comp. Biochem. Physiol., Part C 2001, 128, 27-141. (46) Smith, R. J. F. Can. J. Zool. 1974, 52, 1031-1038. (47) Korte, J. J.; Kahl, M. D.; Jensen, K. M.; Pasha, M. S.; Parks, L. G.; LeBlanc, G. A.; Ankley, G. T. Environ. Toxicol. Chem. 2000, 19, 972-981. (48) Thomas-Jones, E.; Thorpe, K.; Harrison, N.; Thomas, G.; Morris, C.; Hutchinson, T.; Woodhead, S.; Tyler, C. Environ. Toxicol. Chem. 2003, 22, 3001-3008. (49) Freeman, W. M.; Walker, S. J.; Vrana, K. E. Biotechniques 1999, 26, 112-122, 124-125. (50) Chang, J. T.; Che, I.; Liao, C.; Wang, H.; Hsu, Y.; Hung, K.; Lin, C.; Hsieh, L.; Cheng, A. Clin. Biochem. 2001, 35, 591-596.
(51) Goidin, D.; Mamessier, A.; Staquet, M.; Schmitt, D.; BerthierVergnes, O. Anal. Biochem. 2001, 295, 17-21. (52) Schmittgen, T. D.; Zakrajsek, B. A. Biochem. Biophys. Methods 2000, 46, 69-81. (53) Ramakers, C.; Ruijter, J. M.; Deprez, R. H.; Moorman, A. F. Neurosci. Lett. 2003, 339, 62-66. (54) Moguilewsky, M.; Fiet, J.; Tournemine, C.; Raynaud, J. P. J. Steroid Biochem. 1986, 24, 139-146.
(55) Cardon, M. C.; Hartig, P. C.; Gray, L. E., Jr.; Wilson, V. S. Toxicol. Sci. 2003, 72 (S-1), 132.
Received for review February 17, 2004. Revised manuscript received May 5, 2004. Accepted May 17, 2004. ES040022B
VOL. 38, NO. 23, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6327
