BIOLOGY OF REPRODUCTION 61, 51±60 (1999)

Binding Characteristics of Estrogen (ER) in Atlantic Croaker (Micropogonias undulatus) Testis: Different Affinity for Estrogens and Xenobiotics from that of Hepatic ER 1

Anna Katrina Loomis and Peter Thomas2 Department of Marine Science, Marine Science Institute, University of Texas at Austin, Port Aransas, Texas 78373

ABSTRACT spermatogenesis and fertility [10]. However, although the existence of estrogens and ERs in males is well recognized An (ER) was identified in cytosolic and nu- throughout the vertebrates, their precise physiological roles Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 clear fractions of the testis in a marine teleost, Atlantic croaker in male reproduction remain unclear. (Micropogonias undulatus). A single class of high affinity, low capacity, and displaceable binding sites was identified by satu- The ER in the testis is also a potential target for endo- ration analysis, with a K of 0.40 nM in cytosolic extracts and a crine disruption by xenoestrogens. Evidence is accumulat- d ing which suggests that estrogenic compounds, including Kd of 0.33 nM in nuclear extracts. Competition studies demon- strated that the receptor was highly specific for estrogens (di- pesticides and industrial chemicals, impair male reproduc- estrone) and also bound tive function in wildlife [11, 12]. Rainbow trout exposed to ؍ ethylstilbestrol Ͼ estradiol k estriol several antiestrogens. Testosterone and 5␣-dihydrotestosterone estrogenic alkylphenolic chemicalsÐnonionic surfactants had much lower affinities for the receptor, whereas no displace- that are often detected in sewage ef¯uentÐexhibit inhibi- ment of specific binding occurred with 11-ketotestosterone or tion of testicular growth [13]. Moreover, increased exposure any of the C21 maturation-inducing steroids. A variety of xen- to environmental xenoestrogens has been suggested as a oestrogens, including o,p؅-dichlorodiphenyltrichloroethane (DDT), possible cause of the apparent increase in male reproductive chlordecone (Kepone), nonylphenol, hydroxylated polychlori- disorders and purported decrease in sperm count in humans nated biphenyls (PCBs), and the mycotoxin zearalenone, bound over the past several decades [14±16]. to the receptor with relatively low binding affinities, 10Ϫ3 to Although the hepatic ER has been characterized in a 10Ϫ5 that of estradiol. A comparison of the binding affinities of wide variety of teleosts [17±19], until now a testicular ER various ligands for the testicular ER and the hepatic ER in this has not been identi®ed in any teleost species. The purpose species revealed that the testicular ER was saturated at a lower of this study, therefore, was to identify the ER in the testis [3H]estradiol concentration (1 nM vs. 4 nM). The binding affin- of the marine teleost, Atlantic croaker (Micropogonias un- ities of several compounds, including testosterone and nafoxi- dulatus), in order to begin de®ning the role of estrogens in dine, exhibited marked differences for the two ERs; and most of male reproductive physiology in teleosts. The hormone the estrogens and xenoestrogens tested had higher binding af- binding characteristics of the nuclear ER in the Atlantic finities for the testicular receptor. Minor amounts of estradiol (0.12 ng/g tissue/h) were produced by testicular tissue fragments croaker testis were examined as well as the binding af®n- incubated in vitro, and estradiol was detected in male Atlantic ities of several xenoestrogens. Synthesis of estradiol was croaker plasma. The identification of a testicular ER and evi- also investigated to con®rm testicular production of the nat- dence that estradiol is produced by the testes in croaker suggest ural ligand of the ER in this species. that estrogens participate in the hormonal control of testicular Recently a second ER, ER␤, has been identi®ed and function in teleosts. characterized in mammals [20]. There are differences in the binding af®nities of various ligands for the ␣ and ␤ forms INTRODUCTION of the mammalian ER. Moreover, they have different tissue distributions, with ER␤ showing high expression levels in Although estrogens are considered to be primarily fe- the testis [21]. Preliminary evidence has also been obtained male reproductive hormones with numerous targets and ac- for multiple forms of the ER in Atlantic croaker, one form tions in the female reproductive system, recent evidence predominating in the gonads and another in the liver [22]. suggests that they may also have important roles in the Therefore, the binding af®nities of a variety of estrogens regulation of gonadal function in males. Estrogens have and xenoestrogens for the testicular and hepatic ERs in At- been detected in the testes and male plasma of several ver- lantic croaker were compared to determine whether similar tebrate species [1, 2]. In addition, classical nuclear estrogen differences in ERs from different tissues exist in a teleost receptors (ERs) have been identi®ed in the male reproduc- species. tive systems of both mammalian and nonmammalian ver- tebrates, including the rat testis, mouse epididymis and ef- MATERIALS AND METHODS ferent ductules, human epididymis and seminal vesicle, and testes of the Urodele amphibian, Necturus maculosus, the Fish and Tissue Collection freshwater turtle, Chrysemys picta, and the elasmobranch, One-year-old Atlantic croaker (ϳ35 g) were collected by Squalus acanthias [3±9]. Studies on ER knock-out mice otter trawl from the bays near Port Aransas, Texas. Fish demonstrate that a functional ER is required for normal were caught in September, at the beginning of gonadal re- crudescence, and were maintained in 4200-liter circular, re- Accepted February 5, 1999. circulating tanks at a temperature of 22±25ЊC under an 11L: Received December 3, 1998. 13D photoperiod, and fed a mixed diet of commercial pel- 1This study was funded by Texas Sea Grant College Program, no. R/ lets and shrimp (3% of BW/day). Spermiating males and MBT-3 and EPA STAR grant no. R826125. 2Correspondence: Peter Thomas, University of Texas Marine Science mature, vitellogenic females were rapidly killed by severing Institute, 750 Channel View Drive, Port Aransas, TX 78373. FAX: 512 749 the spinal cord, and testicular and liver tissues, respectively, 6777; e-mail: [email protected] were collected. Blood was collected from the caudal vein 51 52 LOOMIS AND THOMAS of spermiating males. Tissue for the ER assays and plasma tant was discarded. The washing step was repeated 3 times, for estradiol measurement were stored at Ϫ80ЊC and and the remaining pellet was extracted with 3 ml TEDGK Ϫ20ЊC, respectively, whereas testicular estradiol production for 1 h with vigorous mixing every 10±15 min, and then in vitro was measured immediately after tissue collection. centrifuged at 12 800 ϫ g for 45 min. Frozen samples showed negligible loss of receptor binding when stored at Ϫ80ЊC for at least 6 mo. Measurement of Kd and Bmax Saturation analysis was performed by incubating 100 ␮l Chemicals of cytosolic or nuclear extract with 100 ␮l of varying con- [2,4,6,7-3H]Estradiol-17␤ ([3H]estradiol, 84 Ci/mmol) centrations of [3H]estradiol in TEDG, with ®nal concentra- was purchased from New England Nuclear (Boston, MA). tions ranging from 0.125 to 4 nM. Nonspeci®c binding was Nonradioactive steroids were purchased from Sigma Chem- determined by incubating extracts with [3H]estradiol and ical Company (St. Louis, MO) or Steraloids Inc. (Wilton, 100-fold excess concentrations of DES. Speci®c binding was de®ned as the fraction of total binding displaced by NH). Diethylstilbestrol (DES), tamoxifen citrate, and naf- Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 oxidine hydrochloride were purchased from ICN Biomed- 100-fold excess DES. Extracts (1:20 w:v, cytosol; 1:6 w:v, icals Inc. (Aurora, OH). Zearalenone was obtained from nuclear) were incubated for 18 h at 4ЊC. Free steroid was Sigma Chemical Company. Hydroxylated polychlorinated removed by incubation with 0.5 ml DCC for 20 min fol- biphenyls (PCBs: 4,4Ј-PCB-3-OH, 4,4Ј-dichloro-3-biphen- lowed by centrifugation at 3200 ϫ g for 20 min. Bound ylol; 2Ј,5Ј-PCB-3-OH, 2Ј,5Ј-dichloro-3-biphenylol; 2,2Ј,5Ј- [3H]estradiol was measured in a liquid scintillation counter PCB-4-OH, 2,2Ј,5Ј-trichloro-4-biphenylol; 2Ј,3Ј,4Ј,5Ј-PCB- for 5 min (LS 6000SC; Beckman Instruments Inc., Fuller- 4-OH, 2Ј,3Ј,4Ј,5Ј-tetrachloro-4-biphenylol) were purchased ton, CA). The equilibrium dissociation constant (Kd) and from Ultra Scienti®c (North Kingston, RI). Dichlorodi- binding capacity (Bmax) were calculated from Scatchard phenyltrichloroethane (DDT) compounds and chlordane analysis of the speci®c binding data [23]. Protein content were purchased from Chem Service (West Chester, PA). of tissue extracts was determined using the method of Brad- Chlordecone (Kepone) was obtained from the the National ford [24]. Institute of Environmental Health Sciences Repository. 4- Nonylphenol was obtained from the Huntsman Corporation Variation of Kd and Bmax during Testicular Recrudescence (Port Neches, TX). Aroclor 1254 was purchased from the Saturation analyses were performed on cytosolic frac- Foxboro Co. (North Haven, CT). ICI 164384 and ICI tions of testes collected throughout gonadal recrudescence 182780 were gifts from Dr. A.E. Wakeling at Zeneca Phar- from 17 individuals with gonado-somatic indexes (GSIs) maceuticals (Cheshire, England). All other chemicals were ranging from 1.62% to 12.7%. GSI was calculated as (go- reagent grade and purchased from general laboratory sup- nad weight ϫ 100)/(total weight Ϫ gonad weight). K and pliers. d Bmax were determined for individual ®sh in most cases. Tes- tes were pooled from ®sh with GSIs lower than 1.95%. ER Assay Buffers Homogenization buffer (TEDG) consisted of Tris base Association and Dissociation Kinetics of the Testicular ER (50 mM), Na2 EDTA (1.5 mM), dithiothreitol (1.0 mM), To determine the association kinetics of receptor bind- and glycerol (30% v:v), pH ϭ 7.4. Nuclear wash buffer ing, 0.5 ml of cytosolic extract was incubated in triplicate (TMDS) comprised Tris base (50 mM), MgCl2 (10.7 mM), with 1 nM [3H]estradiol in the presence or absence of 100 sucrose (250 mM), and dithiothreitol (1.0 mM), pH ϭ 7.4. nM estradiol. At speci®c time points between 15 min and Nuclear extraction buffer (TEDGK) consisted of TEDG 24 h, the reaction was stopped with the addition of 0.5 ml buffer ϩ KCl (600 mM). Dextran-coated charcoal (DCC) DCC. The extracts were incubated with DCC for 20 min, was 0.4% w:v activated charcoal, 0.1% w:v Dextran T-70 and then centrifuged at 3200 ϫ g for 20 min. in TEDG (for saturation analysis, association and dissoci- Dissociation kinetics of receptor binding were deter- ation kinetics, and steroid competition curves); or 0.6% w:v mined after a 14-h incubation of cytosolic extract with 1 activated charcoal, 0.15% w:v Dextran T-70 in TEDG (for nM [3H]estradiol. Dissociation began with the addition of xenobiotic competition curves). 100 nM estradiol, and speci®c binding was measured at time points between 15 min and 24 h. Incubations were Preparation of Tissue Extracts terminated with the addition of 0.5 ml DCC. Dissociation was expressed as a percentage of the speci®c binding pre- Liver and testicular tissue extracts were prepared using sent in control incubations at each time point. the protocol described for the spotted seatrout hepatic ER [18]. All procedures were performed on ice or at 4ЊC. One Steroid Specificity and Xenobiotic Binding to Testis and gram of tissue was homogenized in 9 ml TEDG with 4 passes of a polytetra¯uoroethylane (Te¯on) pestle in a glass Liver ER homogenizer. The homogenate was centrifuged for 20 min Two hundred microliters of testicular cytosolic extract at 820 ϫ g. The supernatant was centrifuged for 1 h at 103 was incubated with 300 ␮lof[3H]estradiol (®nal concen- 000 ϫ g to obtain a cytosolic extract, which was assayed tration 1 nM) in TEDG for 18 h with or without competi- for receptor content. Lipid layers were aspirated from the tors. Concentrations of competitors ranged from 10 pM to liver tissue extracts after each centrifugation step. The cy- 1 mM. All competitors were dissolved in ethanol and added tosolic fraction was assayed immediately or frozen at to the cytosolic extract for a ®nal ethanol concentration of Ϫ80ЊC until use (binding remained constant in cytosolic 1%. Free steroid was removed with the addition of 0.5 ml and nuclear extracts for at least 1 mo). DCC followed by a 20-min incubation before centrifugation To obtain a nuclear fraction, the pellet from the initial at 3200 ϫ g. Maximum speci®c binding was de®ned as the centrifugation was washed with 10 ml TMDS, the mixture fraction of total [3H]estradiol binding suppressed by 100- was centrifuged at 1150 ϫ g for 20 min, and the superna- fold excess DES (100 nM). Competitor binding was ex- TESTICULAR ER IN ATLANTIC CROAKER 53 pressed as a percentage of maximum speci®c binding. Five concentrations were tested for each competitor, and the as- say was repeated with 3 different cytosolic extracts for each competitor. The hepatic ER is an intermediary in the regulation of vitellogenesis by estradiol in females and is usually not present in high concentrations in males [25]. Therefore, liv- ers from mature females were used in these investigations. The liver ER saturates between 4 and 8 nM estradiol, and 4nM[3H]estradiol was used for all liver competition as- says. Maximum speci®c binding was determined by incu- bation with 400 nM DES.

Measurement of Testicular Estradiol Production In Vitro Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 Testes were removed from spermiating Atlantic croaker and kept on ice. Approximately 100-mg tissue fragments were incubated in 24-well polystyrene incubation plates with 1-ml aliquots of Dulbecco's Modi®ed Eagle's medium lacking any pH indicator dye, supplemented with sodium bicarbonate (1.2 g/L), penicillin (60 mg/L), and strepto- mycin (100 mg/L), at a pH of 7.4. After a 30- to 60-min preincubation, tissue was coarsely chopped, medium was removed, and 1 ml of fresh medium was added to each incubation well. Tissue fragments were incubated for 9 h in a Dubnoff metabolic shaking incubator (Precision Sci- enti®c, Chicago, IL) in a humid environment at 25ЊC under an atmosphere of oxygen. At the end of the incubation pe- riod, media were removed and frozen at Ϫ20ЊC until as- sayed for estradiol. Because estradiol production was ex- pected to be low, samples were concentrated by ®ltering several milliliters of incubation medium through Sep-Pak C18 cartridges (Waters Chromatography Division, Milli- pore Corporation, Milford, MA). Media from several in- cubations were combined, and analysis was performed on 2 or 2.8 ml of media. Steroids were eluted from Sep-Paks with 2 ml water followed by 2 ml 10% methanol, and ®- nally 2 ml 100% methanol, which was collected and dried down under a stream of nitrogen. Extracts were resuspend- FIG. 1. A) Representative saturation plot of [3H]estradiol binding in tes- ticular cytosolic extract. Extract was incubated for 18 h at 4ЊC with a ed in a phosphate buffer and subsequently analyzed for es- range of [3H]estradiol concentrations. TB, Total binding; SB, specific bind- tradiol by RIA according to the protocol of Smith and ing; NSB, nonspecific binding. B) Scatchard plot of specific binding. Kd

Thomas [25]. The overall mean steroid recovery ef®ciency ϭ 0.46 nM; Bmax ϭ 0.03 nM, equivalent to 10 pmol/g protein. was 85%.

Measurement of Estradiol in Male Plasma and B). Scatchard analysis showed a single class of high af®nity, saturable binding sites with a mean K of 0.33 Ϯ Two hundred-microliter aliquots of plasma from 12 ma- d 0.01 nM (n ϭ 16). The mean Bmax for nuclear extracts was ture early recrudescing male Atlantic croaker (mean GSI ϭ 0.028 Ϯ 0.002 nM, equal to 0.083 Ϯ 0.007 pmol/g testis, 5.28 Ϯ 0.56%) were assayed for estradiol after solvent ex- for individuals with a GSI greater than 3% (n ϭ 11). traction (hexane/ethyl acetate, 70:30) by speci®c RIA [25]. Mean steroid recovery ef®ciency was 76%. The antiserum shows negligible cross-reaction with teleostean androgens Variation in Kd and Bmax during Gonadal Recrudescence (Ͻ 0.02%) and recognizes primarily estradiol-17␤ (cross- The Kd of testicular cytosolic extracts varied from 0.21 reactivity with estrone and estriol Ͻ 0.7%) [25]. to 0.73 nM, but no consistent trend was apparent during the period of testicular growth. The Bmax ranged from non- RESULTS detectable to 0.043 nM, and on a per-testis basis increased during recrudescence from below the assay detection limits Saturation Analysis of Testicular ER 2 to 3.04 pmol per testis (r ϭ 0.72, data not shown). The Kd Saturation and Scatchard analysis of binding to testicular for nuclear fractions varied little during gonadal growth, cytosolic extracts showed a single class of high-af®nity ranging from 0.25 to 0.44 nM. The receptor content of the binding sites that saturated at 1±2 nM [3H]estradiol and had nuclear extracts ranged from 0.013 to 0.071 nM during re- a mean equilibrium dissociation constant (Kd) of 0.40 Ϯ crudescence, with higher values in individuals with GSIs 0.04 nM (n ϭ 13) and a mean maximum binding capacity below 3%, and fewer binding sites as GSIs increased. This (Bmax) of 0.023 Ϯ 0.004 nM, equal to 0.23 Ϯ 0.04 pmol/g trend was consistent when Bmax was calculated in terms of testis (n ϭ 17, Fig. 1, A and B). moles per gram of protein, but no trend was apparent when 3 Speci®c binding of [ H]estradiol to testicular nuclear ex- the Bmax was calculated on a per-testis basis (data not tracts was also saturated between 1 and 2 nM (Fig. 2, A shown). 54 LOOMIS AND THOMAS Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020

FIG. 2. A) Representative saturation plot of [3H]estradiol binding in tes- ticular nuclear extract. Extract was incubated for 18 h at 4ЊC with a range of [3H]estradiol concentrations. TB, Total binding; SB, specific binding;

NSB, nonspecific binding. B) Scatchard plot of specific binding. Kd ϭ 0.35 nM; Bmax ϭ 0.04 nM, equivalent to 26 pmol/g protein.

FIG. 4. Ligand binding specificity of testicular cytosolic extracts. Extracts were incubated for 18 h with 1 nM [3H]estradiol and 10 pM-1 mM com- petitor. Steroid binding is expressed as a percentage of the maximum specific binding. Each point represents the mean of 3 observations. SEM Ͻ 10% of the mean. A) Estrogens and antiestrogens. B) Other steroids: 11-KT, 11-ketotestosterone; 5␣-DHT, 5␣-dihydrotestosterone; 20␤-S,

3 17,20␤,21-trihydroxy-4-pregnen-3-one; 17,20␤-P, 17,20␤-dihydroxy-4- FIG. 3. Time course of association and dissociation of [ H]estradiol pregnen-3-one. C) Xenobiotics: PCB-OH, 2Ј,3Ј,4Ј,5Ј-tetrachloro-4-bi- Њ binding to testicular cytosolic extracts at 4 C. Each point represents the phenylol. mean Ϯ SEM of 3 estimates, each derived from triplicate measurements. Circles, association; squares, dissociation. TESTICULAR ER IN ATLANTIC CROAKER 55

TABLE 1. EC50 and RBA of various competitors for testicular and liver cytosolic ERs (each value is the mean of 3 competition assays).* Testis Liver

a a Competitor EC50 (M) RBA (%) EC50 (M) RBA (%) Steroids DES 1.3 ϫ 10Ϫ10 100 2.3 ϫ 10Ϫ9† 100 Estradiol-17␤ 7.0 ϫ 10Ϫ10 17.9 4.5 ϫ 10Ϫ9 50 Estriol 1.7 ϫ 10Ϫ8 7.4 ϫ 10Ϫ1 3.5 ϫ 10Ϫ8 6.4 Estrone 1.5 ϫ 10Ϫ8 8.3 ϫ 10Ϫ1 2.0 ϫ 10Ϫ8 11.3† 5␣-dihydrotestosterone 1.7 ϫ 10Ϫ6 7.4 ϫ 10Ϫ3 4.0 ϫ 10Ϫ6 5.6 ϫ 10Ϫ2 Testosterone 9.0 ϫ 10Ϫ6 1.4 ϫ 10Ϫ3 ND,† ND, † 11-ketotestosterone ND ND ND ND Cortisol ND ND ND ND 17,20␤-dihydroxy-4-pregnen-3-one ND ND ND ND 17,20␤,21-trihydroxy-4-pregnen-3-one ND ND ND ND Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 Antiestrogensb Tamoxifen citrate 5.0 ϫ 10Ϫ7 1.4 ϫ 10Ϫ2 4.0 ϫ 10Ϫ6 7.5 ϫ 10Ϫ2 Nafoxidine hydrochloride 3.5 ϫ 10Ϫ6 2.0 ϫ 10Ϫ3 L,† L,† ICI 164384 7.0 ϫ 10Ϫ9 1.0 4.5 ϫ 10Ϫ8 6.7 ICI 182780 2.5 ϫ 10Ϫ9 2.8 8.0 ϫ 10Ϫ9 37.5† Xenobioticsb o,pЈ-DDT 3.0 ϫ 10Ϫ5 2.3 ϫ 10Ϫ4 1.5 ϫ 10Ϫ4 2.0 ϫ 10Ϫ3 o,pЈ-DDE 3.0 ϫ 10Ϫ5 2.3 ϫ 10Ϫ4 4.0 ϫ 10Ϫ4† 7.5 ϫ 10Ϫ4 o,pЈ-DDD 4.0 ϫ 10Ϫ5 1.8 ϫ 10Ϫ4 5.0 ϫ 10Ϫ4† 6.0 ϫ 10Ϫ4 p,pЈ-DDT L L L L p,pЈ-DDE L L L L p,pЈ-DDD L L L L 4,4Ј-PCB-3-OH L L ND,† ND,† 2Ј,5Ј-PCB-3-OH 1.5 ϫ 10Ϫ5 4.7 ϫ 10Ϫ4 L,† L,† 2,2Ј,5Ј-PCB-4-OH 1.0 ϫ 10Ϫ5 7.0 ϫ 10Ϫ4 L,† L,† 2Ј,3Ј,4Ј,5Ј-PCB-4-OH 6.0 ϫ 10Ϫ7 1.2 ϫ 10Ϫ2 2.0 ϫ 10Ϫ6 1.5 ϫ 10Ϫ1† Kepone 4.0 ϫ 10Ϫ5 1.8 ϫ 10Ϫ4 1.0 ϫ 10Ϫ4 3.0 ϫ 10Ϫ3† Nonylphenol 1.3 ϫ 10Ϫ6 5.6 ϫ 10Ϫ3 1.5 ϫ 10Ϫ5† 2.0 ϫ 10Ϫ2 Aroclor 1254 L L L L Zearalenone 1.5 ϫ 10Ϫ6 4.7 ϫ 10Ϫ3 1.7 ϫ 10Ϫ6 1.8 ϫ 10Ϫ1† Chlordane 2.5 ϫ 10Ϫ5 2.8 ϫ 10Ϫ4 L,† L,† * ND, no displacement; L, less than 50% displacement; †, Ͼ 1 order of magnitude difference between testis and liver values. a 3 EC50 is the competitor concentration that causes 50% displacement of [ H]estradiol. Testicular cytosol fractions were incubated with 1 nM and liver cytosolic fractions were incubated with 4 nM [3H] estradiol. b Antiestrogens and xenobiotics were assayed separately from steroids. Different activated charcoal concentrations were used, and RBAs are calculated Ϫ11 Ϫ9 from a DES EC50 ϭ 7 ϫ 10 for testis, 3 ϫ 10 for liver.

Association and Dissociation Kinetics of Testicular ER (20␤-S) and 17,20␤-dihydroxy-4-pregnen-3-one (17,20␤- 3 P)Ðonly two of the androgens, testosterone and 5␣-dihy- The association and dissociation kinetics of [ H]estradiol drotestosterone, were able to displace speci®c binding (Fig. speci®c binding to the testicular ER are shown in Figure 3. 4B). At 4ЊC, association occurred with a T1/2 of 1 h, and - imum association was reached after 5 h. Binding remained constant up to 16 h, and approximately 36 Ϯ 21% of spe- Xenobiotic Binding to the Testicular ER ci®c binding was lost by 24 h. Dissociation of [3H]estradiol occurred with a T1/2 of 8 h, and after 24 h 84 Ϯ 2% of the Zearalenone, 4-nonylphenol, and one hydroxylated PCB [3H]estradiol had dissociated from its binding site. (2Ј,3Ј,4Ј,5Ј-tetrachloro-4-biphenylol) were all able to dis- place 50% of the speci®c binding at concentrations around Binding Specificity of Testicular ER 1 ␮M (Fig. 4C). All o,pЈ congeners of the DDT compounds were able to displace 50% of the speci®c binding at ap- The synthetic estrogen DES bound to the ER with ap- proximately 30 ␮M (Fig. 4C, Table 1), whereas the p,pЈ proximately 5 times higher af®nity than estradiol (EC50 1.3 Ϫ10 Ϫ10 congeners of DDT bound with less af®nity and did not ϫ 10 vs. 7.0 ϫ 10 for estradiol, Table 1), and this displace 50% of the speci®c binding. The chlorinated pes- compound was used for calculating relative binding af®n- ticide Kepone was able to displace more than 80% of the ities (RBAs). The natural estrogens estradiol, estriol, and speci®c binding at a concentration of 100 ␮M. estrone all bound with high af®nity to the testicular ER (Fig. 4A). Estriol and estrone had 20 times less af®nity for the receptor than estradiol on the basis of RBA values (Ta- Saturation Analysis of Hepatic ER ble 1). The antiestrogens ICI 164384 and ICI 182780 bound with 1% and 3% of the af®nity for DES, respectively. The Saturation and Scatchard analysis of binding to hepatic antiestrogens tamoxifen citrate and nafoxidine hydrochlo- cytosolic extracts showed a single class of binding sites ride also bound, but with 100- to 1000-fold less af®nity with a mean Kd of 2.75 Ϯ 0.53 nM and a mean Bmax of than the ICI compounds. 0.46 Ϯ 0.11 nM, equal to 4.6 Ϯ 1.1 pmol/g liver (n ϭ 5; Among the other steroids assayed for binding af®nity, Fig. 5). Saturation of this receptor occurred between 4 and including androgens, cortisol, and the two maturation-in- 8nM[3H]estradiol, a concentration four times higher than ducing steroidsÐ17,20␤,21-trihydroxy-4-pregnen-3-one that required to saturate the testicular ER. 56 LOOMIS AND THOMAS

®nity of estrone for the two receptors was similar (Fig. 6B). Nafoxidine displaced only 15% of the speci®c binding in the liver extracts, whereas it caused 75% displacement in the testicular extracts (Fig. 6C). The differences in binding af®nities for the two receptors of the antiestrogen ICI 164384 and the xenobiotic o,pЈ-DDT were similar to those observed with estradiol (Fig. 6, A, D, and E). A variety of xenobiotics, including 2,2Ј,5Ј-PCB-4-OH, did not cause 50% displacement of [3H]estradiol from the hepatic ER at the highest concentration tested (100 ␮M, Fig. 6F). The nonionic surfactant 4-nonylphenol also showed a distinct difference in binding af®nities between tissues (Fig. 6G). Interestingly, the mycotoxin zearalenone was the only com-

pound for which the binding curves in the two tissues over- Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 lapped (Fig. 6H).

Measurement of Testicular and Plasma Estradiol Measurable amounts of estradiol were produced by tes- ticular fragments in vitro in 2 out of 4 experiments when the incubation media were concentrated. Testicular estradiol production was 0.12 and 0.07 ng/g tissue/h in the two ex- periments. Estradiol was also detected in all of the plasma samples collected from mature male Atlantic croaker at a midpoint in gonadal recrudescence, with a mean concen- tration of 0.28 Ϯ 0.03 ng/ml (n ϭ 12).

DISCUSSION The results of this study demonstrate the presence of a nuclear ER in the testis of Atlantic croaker, the ®rst testic- ular ER reported in a teleost species. This binding com- ponent exhibits the characteristics of a classic nuclear ste- roid receptor. Estradiol was able to bind to both cytosolic and nuclear extracts of testicular tissue. Binding was satu- rable and demonstrated steroid speci®city for both natural and synthetic estrogens. Binding to the receptor was re- versible and exhibited high af®nity and low capacity, with FIG. 5. A) Representative saturation plot of [3H]estradiol binding in he- patic cytosolic extract. Extract was incubated for 18 h at 4ЊC with a range a Kd of 0.4 nM and a Bmax of 10 pmol/g protein in cytosolic of [3H]estradiol concentrations. TB, Total binding; SB, specific binding; extracts. These values are comparable to those of testicular

NSB, nonspecific binding. B) Scatchard plot of specific binding. Kd ϭ 2.05 ERs in other vertebrate species; a rat testicular ER was nM; Bmax ϭ 0.31 nM. described with a Kd of 0.43 nM and a Bmax of 15 pmol/g cytosol protein, and an amphibian testicular ER was re- ported with a Kd of 0.1 nM and a Bmax of 2±5 pmol/g Testicular and Hepatic ER Binding Comparisons protein [3, 7]. The binding characteristics of the Atlantic A comparison of hepatic and testicular ER binding croaker ER differ from those of the sex-steroid binding pro- showed several differences. The testicular extracts had a tein, which has a higher binding capacity and lower binding higher af®nity than the liver extracts for the majority of af®nity, and binds both estradiol and testosterone [26]. competitors tested, as shown by the concentrations required Moreover, EDTA and dithiothreitol, which destroy the to displace 50% of the speci®c binding (EC50s, Table 1). binding activity of the sex-steroid binding protein, were The EC50s of DES, testosterone, nafoxidine, and several included in the homogenization buffer used in this study. xenobiotics were more than an order of magnitude higher We also report here that Atlantic croaker testes synthe- in the liver extracts. However, this was not the case for all size estradiol in vitro. Synthesis of estradiol and aromatase compounds tested. The EC50s of ICI 182780 and estrone activity in the testis have been described previously in the were similar, and those for zearalenone were nearly iden- rainbow trout (Oncorhynchus mykiss) [27], and expression tical. The RBAs, which compare the af®nity of a competitor of the aromatase mRNA has been described in the testis of with the af®nity of DES at the EC50, were also different the channel cat®sh (Ictalurus punctatus) [28]. Testicular for many compounds. The RBAs of testosterone, the anti- synthesis of estradiol has been reported in a variety of other estrogens nafoxidine hydrochloride and ICI 182780, all hy- vertebrate species, including amphibians and humans [2, droxylated PCBs tested, Kepone, zearalenone, and chlor- 29, 30], although it typically represents a minor component dane were all more than an order of magnitude different of testicular steroidogenesis (Ͻ 2% of testosterone produc- for the two receptors. tion in humans). Most evidence suggests that the Leydig The competition curves for the binding of several rep- cell is the site of estradiol synthesis in the adult testis of resentative compounds to ERs in the two tissues are com- both mammalian and nonmammalian vertebrates [1, 31]. pared in Figure 6. Estradiol, the natural ligand, caused 50% Estradiol is also present in male Atlantic croaker plasma, displacement at nearly an order of magnitude lower con- and levels are comparable to those reported in several other centration in testicular extracts (Fig. 6A), whereas the af- teleost species (0.1±0.7 mg/ml) including tilapia (Sarother- TESTICULAR ER IN ATLANTIC CROAKER 57 Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020

FIG. 6. Comparison of competitor binding to liver and testicular cytosolic ERs. Extracts were incubated with saturating concentrations of [3H]estradiol: 1 nM for testis, 4 nM for liver. Each point represents the mean of 3 observations Ϯ SEM. Circles, testis; squares, liver. 58 LOOMIS AND THOMAS odon melanotheron), the tropical freshwater teleost Pygo- liver ER). Differences in the relative binding af®nities of centrus cariba, and the weakly electric ®sh Sternopygus several synthetic estrogens (moxestrol, ICI 164384) to ER␣ macrurus [32±34]. The presence of estradiol in the plasma and ␤ have also been observed in mice [21]. Of all the and testes is further evidence that estrogens have a physi- compounds tested in this study, only zearalenone had over- ological role in male Atlantic croaker. lapping binding curves for the two receptors. Interestingly, Varriale and coworkers suggested that estradiol has a the binding curves of zearalenone for human ER␣ and ␤ role in seasonal reproduction of the male frog (Rana es- also overlap [43]. culenta) from observations that estradiol peaks in both the The ER in the testis is not only a site of estrogen action plasma and testis in early spring, when plasma testosterone but is also a likely target for endocrine disruption by xeno- levels are decreasing and spermatogenesis is initiated [29]. estrogens. Testicular growth was impaired in male rainbow Moreover, the peak in binding capacity of the testicular ER trout chronically exposed to several estrogenic alkylphen- correlated with the beginning of spermatogenesis in this olic compounds [13]. A recent study has shown that a large species [35]. In the present study, the ER was identi®ed in proportion of the male ®sh collected from several European both cytosolic and nuclear fractions of Atlantic croaker tes- rivers in the vicinity of sewage outfalls were intersex, with Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 tis throughout the period of gonadal recrudescence. Recep- ovarian as well as testicular tissue in their gonads, and also tor content per testis increased in cytosolic fractions as GSI had elevated plasma vitellogenin levels, evidence of xen- increased, while in nuclear fractions receptor concentrations oestrogen exposure [12]. Rats exposed to the estrogenic were higher per gram of protein during early recrudescence. chemicals DES, octylphenol, or butyl benzyl phthalate dur- Additional data on seasonal changes in plasma estradiol ing gestation or lactation exhibited reduced testicular levels and testicular estradiol production, as well as ER weight and reduced sperm production [44]. Although the localization within the testis, will be required to interpret site of interference by these chemicals was not determined, the signi®cance of these changes in ER content during the the testis appears to be sensitive to xenoestrogens, either Atlantic croaker reproductive cycle. However, these prelim- directly or indirectly, during gonadal development and neo- inary data indicate that estradiol may be active throughout natal life. the reproductive season in male Atlantic croaker. The data presented in this study show that a variety of A number of other actions of testicular estradiol have xenobiotics bind to the testicular ER of a teleost and there- been described in several vertebrate species. Estradiol de- fore have the potential to interfere with male reproductive creases serum testosterone levels and gonadotropin-stimu- function. Many of the compounds tested in this study which lated testicular testosterone production in vitro in adult bound to the ER, including o,pЈ-DDT, Kepone, and nonyl- male rats [36, 37]. Involvement of estradiol has also been phenol, have been shown to have estrogenic effects in suggested in the process of Leydig cell development in rats mammals, birds, or ®sh [13, 45, 46]. Hydroxylated PCBs, [38]. Estradiol also in¯uences the spermatogonial mitotic which have been identi®ed in ®sh-eating birds, marine index in the frog, Rana esculenta [39] and spermatogenic mammals, and humans, also have the potential for estro- progression in the dog®sh shark, Squalus acanthias [40]. genic and antiestrogenic activity [47]. However, ®sh do not In addition, estradiol has been shown to regulate the reab- appear to metabolize PCBs as readily as do other verte- sorption of luminal ¯uid in the head of the epididymis in brates [48], although hydroxylated metabolites of PCBs mice [41]. It is likely that many of these actions of estro- have been identi®ed in some species [49]. gens are mediated by binding to a nuclear ER. ER knock- All of the DDT congeners tested bound to the Atlantic out male mice were infertile, had abnormal spermatogen- croaker ERs. The o,pЈ derivatives had higher af®nities than esis, reduced testis size, and dismorphogenic seminiferous the p,pЈ derivatives, in agreement with the results of pre- tubules; and the motility of their sperm was decreased [10, vious studies [50, 51]. In general, the differences in binding 42]. af®nities of the o,pЈ and p,pЈ derivatives for the Atlantic The recent discovery of ER␤ in the mouse, a second croaker ERs were smaller than those reported for other spe- form of the ER with different tissue distribution and ligand cies. In rat uterine receptors, the o,pЈ-isomers (approxi- binding from those of ER␣, has indicated an additional el- mately 300 ␮M) caused 50±75% inhibition of [3H]estradiol ement of differential transcriptional control by estrogens at binding, whereas the p,pЈ-isomers caused 20% or less in- the level of the receptor [20, 21, 43]. The results of the hibition of binding [50]. In the kelp bass hepatic ER, o,pЈ- present study suggest that tissue differences in ER binding 1, 1-dichloro-2,2Ј-bis(p-chlorophenyl)ethylene (DDE) dis- characteristics also occur in Atlantic croaker. Differential placed more than 70% of the speci®c binding, whereas at binding af®nities of several ligands to the hepatic and tes- similar concentrations p,pЈ-DDE displaced only 10% [51]. ticular ERs were identi®ed. The testicular ER had a higher Similarly, o,pЈ-DDT (100 ␮M) caused a 60% inhibition of af®nity for nearly all of the compounds tested and also binding to the rabbit uterine ER, whereas p,pЈ-DDT and saturated at a lower estradiol concentration. While these p,pЈ-DDE caused only a 10% inhibition of [3H]estradiol results do not prove the existence of two forms of the ER, binding [52]. In contrast, our results show that the o,pЈ- preliminary data from our laboratory suggest that multiple DDT inhibited 60% of binding and p,pЈ-DDT inhibited al- forms of the ER are present in this species, with different most 50% of binding in both the testicular and liver ERs forms dominant in the gonads and liver [22]. The relative at a concentration of 100 ␮M. Thus, it is important to test binding af®nities of several natural estrogens were similar potential endocrine-disrupting chemicals on a broad range in both tissues, whereas the relative binding af®nities for of vertebrate species in view of possible species differences two of the antiestrogens, nafoxidine hydrochloride and ICI in their receptor binding. The competitive binding assays 182780, differed markedly. In addition, the order of binding with Atlantic croaker tissues show that a variety of envi- af®nity of estrogens and anti-estrogens for the two ERs ronmental chemicals including pesticides, nonionic surfac- were different (DES Ͼ estradiol k ICI 182780 Ͼ ICI tants, and hydroxylated PCBs have the potential to interfere 164384 ϭ estrone ϭ estriol Ͼ tamoxifen Ͼ nafoxidine, for with the reproductive function of both the liver and testis the testis ER; DES Ͼ estradiol k ICI 182780 Ͼ estrone Ͼ in this species. We have observed estrogenic feedback ef- estriol ϭ ICI 164384 Ͼ tamoxifen Ͼ nafoxidine, for the fects of o,pЈ-DDT on gonadotropin secretion in croaker, TESTICULAR ER IN ATLANTIC CROAKER 59 with tissue concentrations in the low per-million range, sim- 14. Sharpe RM, Skakkebaek NE. Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 1993; 341: ilar to the EC50 of the pesticide for the testicular ER in this species [53, 54]. 1392±1395. 15. Sharpe RM. Could environmental, oestrogenic chemicals be respon- The differences in binding of xenobiotics to the testicular sible for some disorders of human male reproductive development? and hepatic ER observed in this study may have important Curr Opin Urol 1994; 4:295±301. implications for the pattern of endocrine disruption in dif- 16. Carlsen E, Giwercman A, Keiding N, Skakkebaek NE. Evidence for ferent tissues. [3H]Estradiol binding is displaced from the decreasing quality of semen during past 50 years. Br Med J 1992; testicular ER by xenobiotic concentrations only 10±20% of 305:609±613. those necessary to displace it from the hepatic ER. More- 17. Lazier CB, Lonergan K, Mommsen TP. Hepatic estrogen receptors and plasma estrogen-binding activity in the Atlantic salmon. Gen over, the number of cytosolic receptors in the liver is 20 Comp Endocrinol 1985; 57:234±245. times higher than that in the testis on a per gram of tissue 18. Smith JS, Thomas P. Binding characteristics of the hepatic estrogen basis, and the testicular ER saturates at one-fourth of the receptor of the spotted seatrout, Cynoscion nebulosus. Gen Comp En- [3H]estradiol concentration required to saturate the hepatic docrinol 1990; 77:29±42.

ER. These results suggest that the testis may be more sen- 19. Pakdel F, Feon S, Le Gac F, Le Menn F, Valotaire Y. In vivo estrogen Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 sitive than the liver to environmental estrogens. Therefore, induction of hepatic estrogen receptor mRNA and correlation with investigations of the effects of xenoestrogens in males vitellogenin mRNA in rainbow trout. Mol Cell Endocrinol 1991; 75: 205±212. should include an examination of testicular structure and 20. Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J- function. Xenoestrogen-induced changes in the testis may A. Cloning of a novel estrogen receptor expressed in rat prostate and be discernible at concentrations below those required to in- ovary. Proc Natl Acad Sci USA 1996; 93:5925±5930. duce vitellogenin production. 21. Kuiper GGJM, Carlsson B, Grandien K, Enmark E, Haggblad J, Nils- In conclusion, we report the ®rst evidence for the exis- son S, Gustafsson J-A. Comparison of the ligand binding speci®city tence of a testicular ER in a teleost species. Differences in and transcript tissue distribution of estrogen receptors ␣ and ␤. En- docrinology 1997; 138:863±870. binding characteristics between the hepatic and testicular 22. Hawkins MB, Skipper JK, Crews D, Thomas P. Cloning of three es- ER of this species were observed, suggesting that the testis trogen receptor mRNAs from Atlantic croaker. In: Xth International may be more sensitive to estrogens than the liver. We also Congress on Hormonal Steroids; 1998; Quebec City. Abstract 6. demonstrate that several xenobiotics can bind to both the 23. Scatchard G. The attractions of proteins for small molecules and ions. testicular and hepatic ERs and therefore could potentially Ann NY Acad Sci 1949; 51:660±672. disrupt estrogen action. 24. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72:248±254. ACKNOWLEDGMENT 25. Smith JS, Thomas P. Changes in hepatic estrogen-receptor concentra- The authors would like to thank the members of the P.T. lab for help tions during the annual reproductive and ovarian cycles of a marine with ®sh and tissue collection and for reviewing the manuscript. teleost, the spotted seatrout, Cynoscion nebulosus. Gen Comp Endo- crinol 1991; 81:234±245. REFERENCES 26. Laidley CW, Thomas P. Partial characterization of a sex-steroid bind- ing protein in the spotted seatrout (Cynoscion nebulosus). Biol Reprod 1. Callard GV, Canick JA, Pudney J. Estrogen synthesis in Leydig cells: 1994; 51:982±992. structural-functional correlations in Necturus testis. Biol Reprod 1980; 27. Fostier A, Le Gac F, Loir M. Steroids in male reproduction. In: Pro- 23:461±479. ceedings of the Third International Symposium on the Reproductive 2. Winters SJ, Troen P. Testosterone and estradiol are co-secreted epi- Physiology of Fish; 1987; St. Johns, Newfoundland, Canada. pp. 239± sodically by the human testis. J Clin Invest 1986; 78:870±873. 245. 3. Kato J, Onouchi T, Okinaga S, Ito N. Estradiol receptor in rat testis. 28. Trant JM, Lehrter J, Gregory T, Nunez S, Wunder J. Expression of Endocrinology 1974; 94:902±907. cytochrome P450 aromatase in the channel cat®sh, Ictalurus puncta- 4. Abney TO. A cytoplasmic estradiol receptor in rat testicular tissue. tus. J Steroid Biochem Mol Biol 1997; 61:393±397. Endocrinology 1976; 99:555±566. 29. Varriale B, Pierantoni R, Di Matteo L, Minucci S, Fasano S, 5. Cooke PS, Young P, Hess RA, Cunha GR. Estrogen receptor expres- D'Antonio M, Chief® G. Plasma and testicular estradiol and plasma sion in developing epididymis, efferent ductules, and other male re- androgen pro®le in the male frog Rana esculenta during the annual productive organs. Endocrinology 1991; 128:2874±2879. cycle. Gen Comp Endocrinol 1986; 64:401±404. 6. Murphy JB, Emmott RC, Hicks LL, Walsh PC. Estrogen receptors in 30. Longcope C, Widrich W, Sawin CT. The secretion of estrone and the human prostate, seminal vesicle, epididymis, testis, and genital estradiol-17␤ by human testis. Steroids 1972; 20:439±448. skin: a marker for estrogen-responsive tissues? J Clin Endocrinol Me- 31. Payne AH, Perkins LM, Georgiou M, Quin PG. Intratesticular site of tab 1980; 50:938±948. 7. Mak P, Callard IP, Callard GV. Characterization of an estrogen recep- aromatase activity and possible function of testicular estradiol. Ste- tor in the testis of the Urodele amphibian Necturus maculosus. Biol roids 1987; 50:435±448. Reprod 1983; 28:261±270. 32. Kishida M, Tyler III WA, Specker JL. Changes in plasma estradiol 8. Mak P, Ho S, Callard IP. Characterization of an estrogen receptor in and testosterone concentrations during a brooding cycle of female the turtle testis. Gen Comp Endocrinol 1983; 52:182±189. mouthbrooding tilapia, Oreochromis mossambicus, and male mouth- 9. Callard GV, Mak P. Exclusive nuclear location of estrogen receptors brooding tilapia, Sarotherodon melanotheron. In: Proceedings of the in Squalus testis. Proc Natl Acad Sci USA 1985; 82:1336±1340. Fifth International Symposium on the Reproductive Physiology of 10. Eddy EM, Washburn TF, Bunch DO, Goulding EH, Gladen BC, Lu- Fish; 1995; Austin, TX. p. 236. bahn DB, Korach KS. Targeted disruption of the estrogen receptor 33. Guerrero HY, Caceres-Dittmar G, Marcano D. Seasonal changes in in male mice causes alteration of spermatogenesis and infertility. plasma levels of sexual hormones in the tropical freshwater teleost, Endocrinology 1996; 137:4796±4805. Pygocentrus cariba (formerly P. notatus, Teleostei; Characidae). In: 11. Colborn T, vom Saal FS, Soto AM. Developmental effects of endo- Proceedings of the Fifth International Symposium on the Reproductive crine-disrupting chemicals in wildlife and humans. Environ Health Physiology of Fish; 1995; Austin, TX. p. 239. Perspect 1993; 101:378±384. 34. Zakon HH, Thomas P, Yan H-Y. Electric organ discharge frequency 12. Jobling S, Nolan M, Tyler CR, Brighty G, Sumpter JP. Widespread and plasma sex steroid levels during gonadal recrudescence in a nat- sexual disruption in wild ®sh. Environ Sci Technol 1998; 32:2498± ural population of the weakly electric ®sh Sternopygus macrurus. J 2506. Comp Physiol A 1991; 169:493±499. 13. Jobling S, Sheahan D, Osborne JA, Matthiessen P, Sumpter JP. Inhi- 35. Fasano S, Pierantoni R, Minucci S, Di Matteo L, Chief® G. Seasonal bition of testicular growth in rainbow trout (Oncorhynchus mykiss) ¯uctuations of estrogen-binding activity in the testis of the frog, Rana exposed to estrogenic alkylphenolic chemicals. Environ Toxicol Chem esculenta. Gen Comp Endocrinol 1989; 75:157±161. 1996; 15:194±202. 36. Melner MH, Abney TO. The direct effect of 17␤-estradiol on LH- 60 LOOMIS AND THOMAS

stimulated testosterone production in hypophysectomized rats. J Ste- bertson M, Peterson RE (eds.), Chemically Induced Alterations in roid Biochem 1980; 13:203±210. Functional Development and Reproduction of Fishes. Pensacola, FL: 37. Moger WH. In vitro inhibitory effect of estradiol on testosterone pro- SETAC Press; 1997: 29±51. duction. J Steroid Biochem 1980; 13:61±66. 47. Kramer VJ, Helferich WG, Bergman A, Klasson-Wehler E, Geisy JP. 38. Zhai J, Lanclos KD, Abney TO. Estrogen receptor messenger ribonu- Hydroxylated polychlorinated biphenyl metabolites are anti-estrogenic cleic acid changes during Leydig cell development. Biol Reprod 1996; in a stably transfected human breast adenocarcinoma (MCF7) cell line. 55:782±788. Toxicol Appl Pharmacol 1997; 144:363±376. 39. Minucci S, Di Matteo L, Chief® P, Pierantoni R, Fasano S. 17␤-Es- 48. Hutzinger O, Nash DM, Safe S, DeFreitas ASW, Norstrom RJ, Wild- tradiol effects on mast cell number and spermatogonial mitotic index ish DJ, Zitko V. Polychlorinated biphenyls: metabolic behavior of pure in the testis of the frog, Rana esculenta. J Exp Zool 1997; 278:93± isomers in pigeons, rats, and brook trout. Science 1972; 178:312±314. 100. 49. Melancon MJ Jr, Lech JJ. Isolation and identi®cation of a polar me- 40. Betka M, Callard GV. Negative feedback control of the spermatogenic tabolite of tetrachlorobiphenyl from bile of rainbow trout exposed to progression by testicular oestrogen synthesis: insights from the shark 14C-tetrachlorobiphenyl. Bull Environ Contam Toxicol 1976; 15:181± testis model. APMIS 1998; 106:252±258. 188. 41. Hess RA, Bunick D, Lee K-H, Bahr J, Taylor JA, Korach KS, Lubahn 50. Nelson JA. Effects of dichlorodiphenyltrichloroethane (DDT) analogs 3

DB. A role for oestrogens in the male reproductive system. Nature and polychlorinated biphenyl (PCB) mixtures on 17␤-[ H]estradiol Downloaded from https://academic.oup.com/biolreprod/article/61/1/51/2734627 by guest on 31 December 2020 1997; 390:509±512. binding to rat uterine receptor. Biochem Pharmacol 1974; 23:447±451. 42. Korach KS. Insights from the study of animals lacking functional 51. Spies RB, Thomas P. Reproductive and endocrine status of female estrogen receptor. Science 1994; 266:1524±1527. kelp bass from a contaminated site in the southern California bight 43. Kuiper GGJM, Lemmen JG, Carlsson B, Corton JC, Safe SH, van der and estrogen receptor binding of DDTs. In: Rolland RM, Gilbertson Saag PT, van der Burg B, Gustafsson J-A. Interaction of estrogenic M, Peterson RE (eds.), Chemically Induced Alterations in Functional chemicals and phytoestrogens with estrogen receptor ␤. Endocrinol- Development and Reproduction of Fishes. Pensacola, FL: SETAC ogy 1998; 139:4252±4263. Press; 1997: 113±133. 44. Sharpe RM, Fisher JS, Millar MM, Jobling S, Sumpter JP. Gestational 52. Danzo BJ. Environmental xenobiotics may disrupt normal endocrine and lactational exposure of rats to xenoestrogens results in reduced function by interfering with the binding of physiological ligands to testicular size and sperm production. Environ Health Perspect 1995; steroid receptors and binding proteins. Environ Health Perspect 1997; 103:1136±1143. 105:294±301. 45. Bulger WH, Kupfer D. Estrogenic activity of pesticides and other 53. Khan IA, Thomas P. Estradiol-17␤ and o,pЈ-DDT stimulate gonado- xenobiotics on the uterus and male reproductive tract. In: Thomas JA, tropin release in Atlantic croaker. Mar Environ Res 1998; 46:149± Korach KS, McLachlan JA (eds.), Endocrine Toxicology. New York: 152. Raven Press; 1985: 1±33. 54. Ungerer J, Thomas P. Role of the very low density lipoproteins in the 46. Thomas P, Khan IA. Mechanisms of chemical interference with re- accumulation of o,pЈ-DDT in ®sh ovaries during gonadal recrudes- productive endocrine function in sciaenid ®shes. In: Rolland RM, Gil- cence. Aquat Toxicol 1996; 35:183±195.