Environ. Sci. Technol. 2009, 43, 2614–2619
Expression Signatures for a Model Androgen and Antiandrogen in the Fathead Minnow (Pimephales promelas) Ovary NATÀLIA GARCIA-REYERO,† DANIEL L. VILLENEUVE,‡ KEVIN J. KROLL,† LI LIU,§ EDWARD F. ORLANDO,| KAREN H. WATANABE,⊥ ´ LVEDA,# M A R ´I A S . S E P U GERALD T. ANKLEY,‡ AND N A N C Y D . D E N S L O W * ,† Department of Physiological Sciences and Center for Environmental and Human Toxicology, University of Florida, Gainesville, Florida 32611, U.S. EPA, ORD, NHEERL, MED, Duluth, Minnesota 55804, ICBR, University of Florida, Gainesville, Florida 32611, Department of Animal and Avian Sciences, University of Maryland, College Park, Maryland 20742, Division of Environmental and Biomolecular Systems, Oregon Health and Science University, Beaverton, Oregon 97006, and Department of Forestry and Natural Resources, Purdue University, Lafayette, Indiana 47907
Received September 5, 2008. Revised manuscript received January 21, 2009. Accepted January 27, 2009.
Trenbolone, an anabolic androgen, and flutamide, an antiandrogen, are prototypical model compounds for agonism and antagonism of the androgen receptor. We hypothesized that 48 h exposures of female fathead minnows (Pimephales promelas) to environmentally relevant concentrations of these chemicals would alter genes regulated by the androgen receptor and that a mixture of the two compounds would block the effects. Gene expression in the ovaries was analyzed using a fathead minnow-specific 22 000-gene microarray. Flutamide altered about twice the number of genes as trenbolone, most of which appeared to be through pathways not associated with the androgen receptor. A group of 70 genes, of which we could identify 37, were reciprocally regulated by trenbolone and flutamide. These are candidates for specific biomarkers for androgen receptor mediated gene expression. Four genes stand out as specifically related to reproduction: sperm associated antigen 8 (SPAG8), CASP8 and FADD-like apoptosis regulator (CFLAR), corticotropin releasing hormone (CRH), and 3β-hydroxysteroid dehydrogenases (3β-HSD). Three notable transcriptional regulators including myelocytomatosis viral oncogene homologue (MYC), Yin Yang 1 (YY1), and interferon regulator factor 1 (IRF1) may function as early molecular switches * Corresponding author phone: +1-352-392-2243, ext.5563; fax: +1-352-392-4707; e-mail:
[email protected]. † Department of Physiological Sciences and Center for Environmental and Human Toxicology, University of Florida. ‡ U.S. EPA, ORD, NHEERL, MED. § ICBR, University of Florida. | University of Maryland. ⊥ Oregon Health and Science University. # Purdue University. 2614
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 7, 2009
to control phenotypic changes in ovary tissue architecture and function in response to androgen or antiandrogen exposure.
Introduction Reproduction in vertebrates is controlled by the hypothalamic-pituitary-gonadal (HPG) axis through a cascade of hormones whose concentrations are tightly regulated by hormonal action and feedback control mechanisms. Disruption of normal reproductive function, either transient or permanent, has been linked to certain chemicals that modulate this axis, but the mechanisms involved are not always clear (reviewed in ref 1). Many endocrine disrupting compounds (EDCs) interact directly to activate or antagonize the sex hormone receptors, such as the estrogen receptor (ER) or androgen receptor (AR). The agonist-bound receptors regulate gene expression by binding to regulatory elements in promoters of susceptible genes, in a tissue-dependent manner (2). In this work, we focused on two compounds known to interact with the AR: 17β-trenbolone (TB) and flutamide (FLU). TB is produced from trenbolone acetate, a synthetic steroid administered to beef cattle as a growth promoter, and is relatively stable in animal waste (3). Ankley et al. (4) reported that TB is a potent AR agonist in the fathead minnow (Pimephales promelas, FHM), masculinizing females and reducing fecundity at water concentrations as low as 0.027 µg/L. TB also has a high affinity for the human AR (5). FLU is an antiandrogen used to treat prostate cancer. It competes with testosterone for binding to ARs in the prostate gland (6), thereby reducing the growth of cancer cells. FLU has been shown to reduce fecundity in the FHM, decreasing mature oocytes in females and causing spermatocyte degeneration and necrosis in testis (7). Theoretically, an AR agonist and antagonist should have opposing effects on gene expression patterns. We tested this hypothesis using microarrays. Specifically, we exposed female FHM to TB and FLU, both singly and in combination in water at concentrations known to affect reproduction (4, 7). The single chemical exposures allowed us to compare the gene expression signatures for each compound. The mixture study enabled us to ascertain whether the androgenic effects of TB might be reversed by the antiandrogenic effects of FLU.
Materials and Methods Fish Exposure and Sample Collection. Adult (ca. 6 month old) and reproductively mature female FHM from an on-site culture (U.S. Environmental Protection Agency, MidContinent Ecology Division Laboratory, Duluth, MN) were acclimated to test conditions (25 °C, 16:8 light:dark photoperiod, and fed adult brine shrimp twice daily) over a period of one week. Water quality conditions were monitored daily and were maintained within the guidelines established for FHM reproduction tests (8). The females had well developed ovaries and were undergoing active vitellogenesis, but were not actively spawning. Treatments included a control (Lake Superior water), three water concentrations of TB (0.05, 0.5, and 5 µg/L), three concentrations of FLU (50, 150, and 500 µg/L), and a mixture (0.5 µg TB /L and 500 µg FLU /L), with four replicate tanks per treatment, and six fish per tank, all under flow-through conditions, as previously described (4). No carrier solvents were used. The chemicals, dissolved in water at approximately their water solubility levels, were further diluted to achieve the target concentrations. Exposure times varied between 48 and 51 h for fish from replicate tanks. Water from each 10.1021/es8024484 CCC: $40.75
2009 American Chemical Society
Published on Web 02/23/2009
exposure tank was sampled just prior to addition of fish (0 h), after 24 h of exposure, and approximately 1 h before sampling (47 h). TB and FLU concentrations in the water were quantified as described elsewhere (4, 9). No TB or FLU was detected in the control tanks (method detection limits ) 0.03 µg TB/L; 20 µg FLU/L). Actual concentrations for each treatment were close to target (80-108%, Supporting Information (SI) S-Table 1) At the conclusion of the exposures, blood samples were collected for analysis of vitellogenin (VTG) and 17β-estradiol (E2) plasma concentrations. Ovaries were removed, weighed, flash frozen in liquid nitrogen, and stored at -80 °C. Vitellogenin Measurement. Plasma concentrations of VTG were determined by enzyme-linked immunosorbent assay (ELISA) using the monoclonal antibody, 2D3, as described previously (10). The limit of detection for FHM VTG ELISA was 0.5 µg/mL. All assays were performed in triplicate and reported as the mean of the three measurements. The coefficient of variation was
