Abstract
Two distinct androgen receptors (ARs) with different characteristics were identified in the brain and ovary of Atlantic croaker, Micropogonias undulatus. A nuclear AR, AR1, was identified in the brain that had high affinity binding sites for testosterone (T; Kd, 1.1 ± 0.15 nm; binding capacity, 1.4 ± 0.14 pmol/g tissue; n = 16). A second nuclear AR, AR2, was found in the ovary that had high affinity binding sites for 5α-dihydrotestosterone (DHT; Kd, 0.62 ± 0.1 nm; binding capacity, 0.38 ± 0.06 pmol/g tissue; n = 14). AR2 has physiochemical properties similar to those of other vertebrate ARs. AR2 has high affinity binding for a broad spectrum of natural and synthetic androgens, including 17α-methyl-5α-dihydrotestosterone, which has a relative binding affinity of DHT = 100% > T > mibolerone > 11-ketotestosterone = 16%, a rapid association (t1/2, 44 min) and a slow dissociation (t1/2, 45 h) rate, as well as specific binding to purified DNA. The cytosolic AR2 interacts with heat shock proteins in a manner similar to other steroid receptors, as sodium molybdate stabilizes the receptor, and it has a 7.4–7.8S sedimentation coefficient in a 5–20% sucrose gradient. In contrast, AR1 is highly specific for only a few androgens, with T= 100% relative binding affinity ≫ DHT ≫ 11-ketotestosterone> mibolerone > 17α-methyl-5α-dihydrotestosterone = 0, has rapid association (t1/2, 15 min) and dissociation (t1/2, 2.6 ± 0.7 h) rates, and has specific binding to purified DNA upon heat activation. The cytosolic binding component sediments at 5.6–5.7S in a 5–20% sucrose gradient and is not affected by sodium molybdate, which suggests that AR1 does not interact with heat shock proteins in the usual manner. This is the first report of the presence of two different nuclear ARs displaying markedly different steroid binding specificities within a single vertebrate species.
THE NUCLEAR receptor superfamily is comprised of ligand-activated transcription factors with characteristic structures containing an hypervariable N-terminal trans-activation domain, a conserved DNA-binding domain, and a variable C-terminal ligand-binding domain (1). This modular structure has facilitated the evolution of diverse families of nuclear receptors that recognize a wide variety of different ligands, such as retinoid, steroid, and thyroid hormones (2–4). The hormonal responses are pleiotropic, coordinating a broad suite of tissue- and cell-specific developmental, metabolic, behavioral, and reproductive processes and involve interactions among the ligands, receptors, and cell-specific factors such as corepressors and coactivators (5). Recently, new subtypes and isoforms of nuclear receptors have been identified, which suggests that a single receptor type is not sufficient to mediate the full range of physiological actions of many of these hormones. The differential distribution and activity of receptor subtypes and isoforms allow cognate ligands to bind and act in different manners depending upon the tissue or cell (5). In addition, some receptor subtypes and isoforms demonstrate different binding characteristics (6–8), suggesting that ligand-based selectivity may play a role in controlling the pleiotropic responses of nuclear steroid receptors.
The progesterone receptor (PR) isoforms PR-A and PR-B differentially regulate transcription (9) as a result of differences within their A/B regions. Furthermore, despite similarities in ligand specificity (10), the two isoforms possess different progesterone binding properties, which may be due to their different N-terminal regions differentially affecting the conformation of the ligand-binding domain (6). Additionally, ligand-dependent cross-talk can occur between the human PR-A, but not PR-B, and the human ER (11). Together, these studies suggest that these two isoforms can mediate functionally different responses to different ligands.
The estrogen receptor (ER) is unique within the subfamily of nuclear steroid receptors in that there are two subtypes, ERα and ERβ, with different binding affinities and distributions (7, 12, 13). Rat ERα and ERβ demonstrate similar relative binding affinities (RBAs) for natural estrogens, but different affinities for some synthetic estrogens (7). Paech et al. (14) showed that ligands differentially induce the trans-activation properties of ERα and ERβ depending upon the type of DNA response element the receptor complex interacts with. Recently, an isoform of theβ -subtype, β2, with unique binding characteristics has been characterized by Peterson et al. (8). Taken together, these studies suggest that multiple receptor subtypes can differentially mediate the responses of steroids on a tissue-specific or a cell-specific level and that the ligand may play a role in defining the specific response (15).
Multiple isoforms of the androgen receptor (AR) have also been described (16, 17), but have not been as fully characterized as the PR isoforms. Like the PR isoforms, the human AR-A isoform is an N-terminal truncated version of the full-length AR-B receptor (16). AR-A is expressed at low levels in many androgen-responsive tissues (18); however, it appears to have functions similar to those of the full-length AR-B isoform (19). Likewise, isoforms of the glucocorticoid receptor have been found, but are not well characterized (20). The mineralocorticoid receptor has also been shown to possess multiple isoforms, α, β, and γ, which demonstrate differential expression in the brain of developing rats (21).
There is circumstantial evidence that teleost fish possess multiple ARs. ARs have been biochemically characterized in only a few species of fish, and no nucleotide sequences for fish AR messenger RNAs have been published. The putative fish AR, first described in brown trout (Salmo trutta) skin (22) and subsequently in goldfish (Carassius auratus) brain tissue (23) and rainbow trout (Oncorhynchus mykiss) lymphocytes (24), has high affinity for testosterone (T) and low affinity for dihydrotestosterone (DHT) and the important fish androgen, 11-ketotestosterone (11-KT). Recently, an AR was partially characterized in coho salmon (Oncorhynchus kisutch) ovaries that demonstrated higher affinity for DHT and 11-KT than T (25), a difference that may explain the contradiction between the biological activity of 11-KT and its low binding affinity for the putative fish AR (22–24, 26).
Physiological investigations of androgen action in teleosts also suggest that teleosts have multiple ARs. These studies show that T, an important androgen in male and female fish, and 11-KT, a predominantly male-specific androgen, have different physiological actions and roles (26). T plays an important role in steroid feedback control of gonadotropin secretion in both male and female fish (27, 28). 11-KT is believed to control gonadal differentiation (29), sexual dimorphism (30), and spermatogenesis (31), whereas in females its physiological role is unknown. The presence of two putative fish ARs could explain these differences in androgen action between T and 11-KT.
In the present study, we investigated whether multiple ARs with different ligand binding specificities are present in a single vertebrate species, the Atlantic croaker, Micropogonias undulatus, a well developed model of teleost reproduction. The biochemical characteristics and binding affinities of both a high affinity nuclear AR in the brain, which we have called AR1, and a second high affinity nuclear AR in the ovary, AR2, are described.
Materials and Methods
Materials
[1,2,6,7-3H]T (92.0 Ci/mmol),[ 1,2,4,5,6,7-3H]DHT (127.0 Ci/mmol), and[ 14C]BSA (0.024 mCi/mg) were purchased from New England Nuclear (Boston, MA) and stored at −20 C. The unlabeled steroids were purchased from either Steraloids, Inc. (Wilton, NH), or Sigma Chemical Co. (St. Louis, MO). Mibolerone was a gift from Upjohn Laboratories (Kalamazoo, MI). All steroids were stored in 95% ethanol at −20 C. The chemicals and salts used for making the buffers were purchased from Sigma Chemical Co. and Fisher Scientific International, Inc. (Pittsburgh, PA). The scintillation cocktail was a mixture of 4 liters toluene, 16 g 7,5-diphenyl-oxazole, 0.4 g 1,4-bis[5-phenyl-2-oxazolyl]-benzene, and 400 ml methanol.
Buffers
Assay buffers for AR1 measurement include H-1, the homogenization buffer [50 mm Tris-HCl, 1 mm EDTA, 12 mm monothioglycerol, and 30% glycerol (vol/vol), pH 7.5, at 4 C]; W, the washing buffer [10 mm Tris-HCl, 2 mm MgCl2, 2 mm monothioglycerol, 250 mm sucrose, and 10% glycerol (vol/vol), pH 7.5, at 4 C]; and E-1, the extraction buffer (H-1 and 0.7 m KCl). The assay buffers for AR2 measurement include H-2, the homogenization buffer [50 mm Tris-HCl, 10 mm sodium molybdate, 1 mm EDTA, 12 mm monothioglycerol, and 10% glycerol (vol/vol), pH 7.4, at 4 C]; W, the washing buffer; S-P, the sucrose pad buffer (50 mm Tris-HCl, 1 mm EDTA, and 1.2 m sucrose, pH 7.4, at 4 C); and N, the nuclear buffer (50 mm Tris-HCl, 1 mm EDTA, 0.2 mm mersalyl acid, and 10% glycerol (vol/vol), pH 7.4, at 4 C).
Animals and tissue sampling
Adult Atlantic croaker, between 14–28 cm in size, were collected during the fall by either gill net or trawling from the bays near Port Aransas, TX. Fish were maintained in circular recirculating tanks under constant photoperiod and temperature regimens and were fed a commercial fish food diet daily. Fish were acclimated in the laboratory for at least 1 month before any tissue collection or experimentation. Fish were killed by decapitation, and tissues were removed, placed on ice, and used immediately or frozen on dry ice and stored at −80 C for up to 1 yr.
Preparation of brain tissue for AR1 assays
Whole brain tissue, comprising the olfactory bulbs and everything posterior to the medulla oblongata including the pituitary, was homogenized in H-1 buffer (1:10, wt/vol) at 4 C with five passes of a glass Tenbroeck tissue grinder (Wheaton Science Products, Millville, NJ), and the homogenate was centrifuged at 2,500 × g for 15 min. The supernatant was spun at 160,000 × g for 1 h and charcoal-stripped (0.25%, vol/vol) to remove endogenous steroids to obtain the cytosolic fraction. The crude nuclear pellet was washed three times with buffer W. The final pellet was resuspended in E-1 buffer (1:10 initial tissue wt/vol) and incubated for 1 h, with vortexing at 15-min intervals. The nuclear suspension was centrifuged at 160,000 × g for 1 h. The tissue preparations were either used immediately or frozen at −80 C, where binding activity remained constant for at least 1 week.
Preparation of ovarian tissue for AR2 assays
A homogenate was prepared from ovarian tissue using a Tissuemizer (Tekmar Co., Cincinnati, OH) at medium power for 20 sec in H-2 buffer (1:10, wt/vol) at 4 C followed by three passes of a glass Tenbroeck tissue grinder. All subsequent steps were identical to those for preparing brain cytosolic fractions. Binding activity remained constant for at least 1 week at −80 C.
Nuclear pellets were prepared as described by Peck and Kelner (32) using a sucrose pad/exchange assay. The crude tissue homogenate was layered onto an equal volume S-P buffer, and the resulting nuclear pellets were washed once with buffer W. The pellets were then incubated with various concentrations of [3H]DHT in buffer N for 1 h at 4 C, followed by the addition of 25 mm monothioglycerol, and were further incubated at 4 C for 12 h before separating bound from free steroid with 500 μl of a 60% hydroxyapatite (HAP) slurry. This exchange method was adapted from that reported by Traish et al. (33).
Saturation kinetics and Scatchard analysis
Except where noted all tissue preparations were incubated at 4 C between 14–20 h for measurement of AR1 and between 8–14 h for AR2 measurement. The dissociation constant (Kd) and receptor concentration (Bmax) of AR1 and AR2 were estimated by Scatchard (34) plots using the program Deltagraph (DeltaPoint, Inc., Monterey, CA). To measure total binding, 50 μl [3H]T or[ 3H]DHT (0.1–6 nm) dried under N2 and redissolved in buffer were added to duplicate tubes containing 250-μl aliquots of tissue. In parallel sets of tubes a 100-fold excess of either unlabeled T or DHT was added to determine nonspecific binding.
To determine whether the binding assays were specific for either AR1 or AR2, 100 nm unlabeled 17α-methyl-5α-dihydrotestosterone (MDHT), which binds with high affinity to AR2, but not AR1, was added along with unlabeled and labeled T or DHT, and saturation analyses were performed.
Separation of bound from free steroid
Free steroid was separated from bound using dextran-coated charcoal (DCC) or, where noted, HAP. Equal volumes of DCC [50 mm Tris-HCl, 1 mm EDTA, 10% glycerol (vol/vol), 1% Norit-A charcoal (wt/vol), and 0.1% dextran T-70 (wt/vol), pH 7.5, at 4 C] and sample were incubated for 5 min before centrifugation at 3000 × g for 5 min at 4 C. The supernatants were collected, the standard scintillation cocktail was added, and radioactivity was measured. Alternatively, equal volumes of a 60% slurry of HAP in buffer W-2 [50 mm Tris-HCl, 1 mm EDTA, and 10% glycerol (vol/vol), pH 7.5, at 4 C] were added to the samples and incubated for 20 min with vortexing every 5 min and then centrifuged at 200 × g for 2 min. The supernatant was poured off, and the pellet was washed twice with 1 ml W-2 buffer. One milliliter of ethanol was added to the pellet and incubated for 20 min at 25 C, with vortexing every 5 min. The ethanol supernatant was collected for determination of radioactivity after centrifuging the samples at 1000 × g for 2 min. The radioactivity was measured by counting for 5 min in an LS 6000SC scintillation counter (Beckman Coulter, Inc., Fullerton, CA).
Metabolism of 3H-labeled androgens
To determine whether metabolism of the 3H-labeled androgen ligands occurred during the receptor assays, pooled whole brain and ovarian homogenates were incubated for 24 h at 4 C with 6 nm [3H]T or 6 nm[ 3H]DHT, respectively. The radioactive products were separated by TLC before measurement of radioactivity as described by Smith and Thomas (35), except that dichloromethane was used as the extraction solvent. There was no metabolism (<1.0%) of either[ 3H]T in brain homogenates or [3H]DHT in ovarian cytosolic homogenates (results not shown).
Steroid specificity
The unlabeled competitors, dissolved in 95% ethanol, were pipetted into test tubes and dried under N2. The[ 3H]ligand was added to the tissue extracts immediately before adding the sample to the competitors. Samples were vortexed and incubated for 12–16 h for AR1 and for 8–12 h for AR2. Competition curves of unlabeled T were run in each of the AR1 assays, and those of unlabeled DHT were run in the AR2 assays as standards. The RBAs of steroids for AR1 and AR2 were expressed as a percentage of the maximum specific binding of T and DHT, respectively.
Sucrose gradient centrifugation
A linear 5–20% sucrose gradient (1.7 ml) was prepared by layering equal volumes of 2.2%, 12.1%, and 22% sucrose solutions and allowing the layers to vertically diffuse for 6 h at room temperature followed by 2 h at 4 C (36). The sucrose solutions were prepared in H-1 buffer for AR1 measurement and in H-2 buffer for AR2 measurement. Three hundred-microliter aliquots of tissue preparations containing unlabeled AR1 or labeled AR2 were added to the top of the gradient, and the gradient was centrifuged at 55,000 rpm at 4 C for the appropriate length of time in an SW 55 Ti rotor (Beckman Coulter, Inc., Palo Alto, CA).[ 14C]BSA was used as a standard and was loaded onto a separate gradient during each centrifuge spin. Gradient fractions (∼100 μl) were collected from the bottom of the centrifuge tubes using a peristaltic pump.
Binding of [3H]T to AR1 was measured in individual fractions, using 50 μl 6 nm [3H]T with or without 500 nm T in buffer H-1, after centrifuging the gradient for 9.5 h loaded with cytosolic extracts of brain tissue at 55,000 rpm in a SW 55 Ti rotor from Beckman Coulter, Inc. MDHT (100 nm) was included in both sets of assay tubes for all of the fractions to verify binding to AR1. The HAP method was used to separate bound from free steroid.
Binding of [3H]DHT to AR2 was measured in individual fractions after an 8-h centrifugation of the gradient loaded with cytosolic extracts of ovarian tissue preincubated with 6 nm[ 3H]DHT for 8–12 h with or without 500 nm DHT. The unbound steroid was removed by DCC before loading samples onto the gradient. Specific binding was determined by subtracting the nonspecific binding fractions from the total binding fractions, which were determined from separate gradients.
Statistical analysis
The mean, sem, and sample number are reported. The statistical analyses were performed using two-tailed Student’s t test, and the equality of the variance was determined using an F test.
Results
AR1 and AR2 saturation analysis
Saturation analysis of [3H]T binding to AR1 in the nuclear fraction of pooled, gonadally recrudesced (gonadosomatic index of 10–15%) male and female whole brain homogenates demonstrated high affinity and saturable binding (Fig. 1). Saturation was reached with 2–4 nm [3H]T, and binding remained constant up to a concentration of 40 nm (results not shown). Scatchard plots (Fig. 1, inset) of the saturation data showed a single class of high affinity AR, with a Kd of 1.1 ± 0.15 nm and a Bmax of 1.4± 0.14 pmol/g tissue (n = 16), more than twice the Bmax of [3H]T binding in the nuclear fraction of pooled brain tissue from gonadally regressed (gonadosomatic index of 0.5%) females (Table 1). Specific binding of[ 3H]T to AR1 was also present in the nuclei isolated from whole brain tissue (Kd = 0.46 ± 0.18 nm; Bmax, 0.095 ± 0.04 pmol/g tissue; n = 2), using the method described by Wray et al. to isolate nuclei from mouse brain tissue (37). The binding of [3H]T to AR1 in both the crude nuclear pellet and the pure nuclei was salt extractable and was maximal using a 0.7-m KCl buffer. The Kd of [3H]T binding to AR1 in the cytosolic fraction was similar to that in the nuclear fraction, but the Bmax was lower. Seasonal differences were also observed in the Bmax of cytosolic AR1, which was 35% higher in gonadally regressed fish than in the fully recrudesced fish (Table 1). No specific binding of [3H]T to AR1 in cytosolic fractions of the testes was detected (Table 1). The addition of 10 mm sodium molybdate, Na2MoO4, to the H-1 buffer did not alter the binding kinetics or increase the Bmax (1.1 pmol/g tissue) of AR1.
![A representative saturation curve of[ 3H]T binding to AR1 in a nuclear fraction of whole brain homogenates. Specific binding (•) was determined by subtracting nonspecific (▵; 1 μm T plus [3H]T) from total binding (□). Inset, Scatchard analysis of the specific binding.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631001.jpeg?Expires=1710102366&Signature=BLepCoSFK9OUFcTYVLzQbCHRO9hZdrmqhthKccaZRBq9SxGkqF~ykA4ZEVwYhqryX~VnhfPtq8G3T5GfveG8oglzuFWN3EYMnpPi8OtqmSi8uQsrLkQNo16vwfgx-tSaCCm6afWLC1QLQAIdxOat78bTjCH5AmdWG6Z1j-b-HIkNMm7hEqKo8M9oqcdgLzj6wGuopX3AQXCHIN~7tyE1ylw3ZjR6Txv3cJBsM3J0RzDwCQ4hilgpIm4pNWlf2KVrbHeJs~x4yHvPm69La3B5FA6kMFm6VtLux51QOP1VcV~DyZhz80ae33CtxWtJCuHrk-QcM6m7zAnoQ1R3wx4C0w__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
A representative saturation curve of[ 3H]T binding to AR1 in a nuclear fraction of whole brain homogenates. Specific binding (•) was determined by subtracting nonspecific (▵; 1 μm T plus [3H]T) from total binding (□). Inset, Scatchard analysis of the specific binding.
Dissociation constants (Kd) and receptor concentrations (Bmax) for both AR1 and AR2 in brain, ovarian, and testicular tissues
| Tissue | Reproductive stage | AR1 (binding [3H]T) | AR2 (binding [3H]DHT) | ||
|---|---|---|---|---|---|
| Kd (nm) | Bmax (pmol/g tissue) | Kd (nm) | Bmax (pmol/g tissue) | ||
| Brain | |||||
| Nuclear | Recrudesced | 1.1 ± 0.15 (16) | 1.4 ± 0.14 (16)a | —b | — |
| Cytosolic | 0.95 ± 0.15 (16) | 1.02 ± 0.07 (16)c | 1.3d | 0.02 | |
| Nuclear | Regressed | 6.2 ± 1.4 (4) | 0.66 ± 0.09 (4)a | — | — |
| Cytosolic | 5.7 ± 0.74 (3) | 1.55 ± 0.45 (3)c | — | — | |
| Ovary | |||||
| Nuclear | Recrudesced | — | — | 4.5 ± 2.6 (2) | 2.3 ± 0.2 (2) |
| Cytosolic | ND | ND | 0.62 ± 0.1 (14) | 0.38 ± 0.06 (14) | |
| Nuclear | Regressed | — | — | — | — |
| Cytosolic | — | — | 0.17 ± 0.03 (4) | 0.47 ± 0.1 (4) | |
| Testesc | |||||
| Cytosolic | Recrudesced | ND | ND | 0.94 ± 0.4 (2) | 0.06 ± 0.01 (2) |
| Tissue | Reproductive stage | AR1 (binding [3H]T) | AR2 (binding [3H]DHT) | ||
|---|---|---|---|---|---|
| Kd (nm) | Bmax (pmol/g tissue) | Kd (nm) | Bmax (pmol/g tissue) | ||
| Brain | |||||
| Nuclear | Recrudesced | 1.1 ± 0.15 (16) | 1.4 ± 0.14 (16)a | —b | — |
| Cytosolic | 0.95 ± 0.15 (16) | 1.02 ± 0.07 (16)c | 1.3d | 0.02 | |
| Nuclear | Regressed | 6.2 ± 1.4 (4) | 0.66 ± 0.09 (4)a | — | — |
| Cytosolic | 5.7 ± 0.74 (3) | 1.55 ± 0.45 (3)c | — | — | |
| Ovary | |||||
| Nuclear | Recrudesced | — | — | 4.5 ± 2.6 (2) | 2.3 ± 0.2 (2) |
| Cytosolic | ND | ND | 0.62 ± 0.1 (14) | 0.38 ± 0.06 (14) | |
| Nuclear | Regressed | — | — | — | — |
| Cytosolic | — | — | 0.17 ± 0.03 (4) | 0.47 ± 0.1 (4) | |
| Testesc | |||||
| Cytosolic | Recrudesced | ND | ND | 0.94 ± 0.4 (2) | 0.06 ± 0.01 (2) |
The Kd and Bmax values were determined from Scatchard analyses of saturation binding curves.
ND, Nondetectable binding. The number of determinations is in parentheses.
Values significantly different from one another (P < 0.05).
Binding not tested.
Values significantly different from one another (P < 0.05).
The cytosolic fractions of testicular tissue were prepared using the standard protocols for AR1 and AR2 binding analyses.
Dissociation constants (Kd) and receptor concentrations (Bmax) for both AR1 and AR2 in brain, ovarian, and testicular tissues
| Tissue | Reproductive stage | AR1 (binding [3H]T) | AR2 (binding [3H]DHT) | ||
|---|---|---|---|---|---|
| Kd (nm) | Bmax (pmol/g tissue) | Kd (nm) | Bmax (pmol/g tissue) | ||
| Brain | |||||
| Nuclear | Recrudesced | 1.1 ± 0.15 (16) | 1.4 ± 0.14 (16)a | —b | — |
| Cytosolic | 0.95 ± 0.15 (16) | 1.02 ± 0.07 (16)c | 1.3d | 0.02 | |
| Nuclear | Regressed | 6.2 ± 1.4 (4) | 0.66 ± 0.09 (4)a | — | — |
| Cytosolic | 5.7 ± 0.74 (3) | 1.55 ± 0.45 (3)c | — | — | |
| Ovary | |||||
| Nuclear | Recrudesced | — | — | 4.5 ± 2.6 (2) | 2.3 ± 0.2 (2) |
| Cytosolic | ND | ND | 0.62 ± 0.1 (14) | 0.38 ± 0.06 (14) | |
| Nuclear | Regressed | — | — | — | — |
| Cytosolic | — | — | 0.17 ± 0.03 (4) | 0.47 ± 0.1 (4) | |
| Testesc | |||||
| Cytosolic | Recrudesced | ND | ND | 0.94 ± 0.4 (2) | 0.06 ± 0.01 (2) |
| Tissue | Reproductive stage | AR1 (binding [3H]T) | AR2 (binding [3H]DHT) | ||
|---|---|---|---|---|---|
| Kd (nm) | Bmax (pmol/g tissue) | Kd (nm) | Bmax (pmol/g tissue) | ||
| Brain | |||||
| Nuclear | Recrudesced | 1.1 ± 0.15 (16) | 1.4 ± 0.14 (16)a | —b | — |
| Cytosolic | 0.95 ± 0.15 (16) | 1.02 ± 0.07 (16)c | 1.3d | 0.02 | |
| Nuclear | Regressed | 6.2 ± 1.4 (4) | 0.66 ± 0.09 (4)a | — | — |
| Cytosolic | 5.7 ± 0.74 (3) | 1.55 ± 0.45 (3)c | — | — | |
| Ovary | |||||
| Nuclear | Recrudesced | — | — | 4.5 ± 2.6 (2) | 2.3 ± 0.2 (2) |
| Cytosolic | ND | ND | 0.62 ± 0.1 (14) | 0.38 ± 0.06 (14) | |
| Nuclear | Regressed | — | — | — | — |
| Cytosolic | — | — | 0.17 ± 0.03 (4) | 0.47 ± 0.1 (4) | |
| Testesc | |||||
| Cytosolic | Recrudesced | ND | ND | 0.94 ± 0.4 (2) | 0.06 ± 0.01 (2) |
The Kd and Bmax values were determined from Scatchard analyses of saturation binding curves.
ND, Nondetectable binding. The number of determinations is in parentheses.
Values significantly different from one another (P < 0.05).
Binding not tested.
Values significantly different from one another (P < 0.05).
The cytosolic fractions of testicular tissue were prepared using the standard protocols for AR1 and AR2 binding analyses.
Saturation analysis of [3H]DHT binding to AR2 in the cytosolic fraction of ovarian tissue homogenates demonstrated high affinity and saturable binding (Fig. 2). Saturation was reached at 1–2 nm [3H]DHT, and binding remained constant up to a concentration of 10 nm steroid (results not shown). The linear Scatchard plots (Fig. 2, inset) of the saturation data showed a single class of high affinity binding sites with a Kd of 0.62 ± 0.1 nm and a Bmax of 0.38 ± 0.06 pmol/g tissue (n = 14). Similar binding kinetics of [3H]DHT binding to AR2 in the nuclear fraction of ovarian tissue homogenates were found (Fig. 3) with a Kd of 4.5 ± 2.6 nm and a Bmax of 2.3 ± 0.2 pmol/g tissue (n = 2; Fig. 3, inset). There was no significant difference in the concentrations of the cytosolic receptor in the ovaries between different stages of the reproductive cycle when the values were expressed on a per g tissue basis (Table 1). The omission of 10 mm Na2MoO4 from the H-2 buffer reduced the Bmax by 2-fold, but the binding kinetics of the receptor, with a Kd of 0.65 nm, did not change. Saturation analysis of [3H]DHT binding to AR2 in the testes demonstrated the presence of high affinity binding that had one sixth the number of receptors that were present in ovaries (Table 1).
![A representative saturation curve of[ 3H]DHT binding to AR2 in a cytosolic fraction of ovarian homogenates. Specific binding (♦) was determined by subtracting nonspecific (▵; 100 nm DHT plus [3H]DHT) from total binding (□). Inset, Scatchard analysis of the specific binding.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631002.jpeg?Expires=1710102366&Signature=LvyGWc0mPuvOHC1mhnVk9sasjDW7p-2qv8asvAywEayWggN6NRyqx4vZzXmo2L~YuY1pe8IWMwOw6NXjz69uKA5nCgJsUNRDlkL6qziYZHqPc4wKZHBQ75OUGyTULaUmKeGzn9Ievlu2CNAJ1wVAM4Ht1rTGLZCDGPhb0ByZbvjj5p0MR53qu1cXUPGNn3jgO~m68tHy~TOPcqMOybZW50O3RzHXOmXBM3p5IVHKiUsQ4g3pVecpR6TvrvXVm4PJI73KS-HGrmFI-4PVs~MWzgBRVkOt~AJ4ijryaseZ-hu-RdM7l6PRoWlN9OUqgsQ9fSovdxCkIIjVc8QcyJ20PQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
A representative saturation curve of[ 3H]DHT binding to AR2 in a cytosolic fraction of ovarian homogenates. Specific binding (♦) was determined by subtracting nonspecific (▵; 100 nm DHT plus [3H]DHT) from total binding (□). Inset, Scatchard analysis of the specific binding.
![A saturation curve of [3H]DHT binding to AR2 in a nuclear fraction of ovarian homogenates using a sucrose pad/exchange assay (32 ). Specific binding (▾) was determined by subtracting nonspecific (▵; 500 nm DHT plus[ 3H]DHT) from total binding (□). Inset, Scatchard analysis of the specific binding.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631003.jpeg?Expires=1710102366&Signature=K9WaYfFa9JMyOgupmzLZnDwTLnncATc2WesAU94YuGbXGc4MtewjGD9tnNj75if89vnWg2eiwamF01CegOQuorScSvJBwAdg76nkbApe4DPhVGGtsH-Pj41HBD9NaUzrDtb2OstZAupepEdc72XBHlfF7Plwf2~yQnvrZU5opwznR5zneQNuZJ4hiCSqGBwP2P7gZ6MtzcA991HOLETRjQTXgYzjEBPya1STtEdPd~N4XzQkFMe5Vo-T87TIOvXZ77wUHLQ-uK9~Ff57BAnI729l-tNtG2Bih5ZLrA4xYNUWzkbju5H7962B~5Fui835-KGfOg4e71gYhKzcrRVJkQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
A saturation curve of [3H]DHT binding to AR2 in a nuclear fraction of ovarian homogenates using a sucrose pad/exchange assay (32 ). Specific binding (▾) was determined by subtracting nonspecific (▵; 500 nm DHT plus[ 3H]DHT) from total binding (□). Inset, Scatchard analysis of the specific binding.
AR1 and AR2 association and dissociation kinetics
Equilibrium studies of [3H]T binding to AR1 in the nuclear fraction of pooled whole male and female brain homogenates and[ 3H]DHT binding to AR2 in the cytosolic fraction of ovarian tissue homogenates demonstrated that both receptors have rapid association kinetics at 4 C. [3H]T binding to AR1 had a t1/2 of 15 min, reached equilibrium within 10 h, and remained stable for 24 h (Fig. 4).[ 3H]DHT binding to AR2 had a t1/2 of 44± 3 min (n = 3) at 4 C, reaching equilibrium within 8 h, with a small loss of binding by 24 h (Fig. 5). The association kinetics of[ 3H]T binding to AR1 were identical when examined using buffer H-2 rather than either H-1 or E-1 (data not shown). The rate of association of [3H]DHT binding to AR2 increased 2-fold at 20 C with a t1/2 of 20 ± 5 min (n = 3), and maximum binding was reached within 2 h, with no loss of binding through 4 h (data not shown).
![Association (•) and dissociation (○) kinetics at 4 C of [3H]T binding to AR1 in brain nuclear extracts. Association kinetics were calculated by adding 10 nm[ 3H]T with or without 1 μm T and determining the amount of specific binding at various time points. Dissociation kinetics were calculated by incubating 4 nm[ 3H]T, with or without 500 nm T for determining nonspecific binding, for 6 h before adding 500 nm T, and binding was subsequently measured at various time points. Each data point is the average of three replicate assays ± sem.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631004.jpeg?Expires=1710102366&Signature=oFy4tM08KVl36gOlYIANwolikkhOxav8L4bsOOaQp3PExgeOA7WNZU7quH71m1Ze-twyD9J8G~c6v~SanytfUSXlwFgpaY612IIzMse~UaqQfP9uzNjfuzvnLnYHhWdsokgSwlguF7XDWYFeWX0d06qx0tkljKAVXV4asryXC1ABvSGE3MrJ3Nfo3bkIkdbQANwdOHvlaiz62FeVV5TVjBOSVpQnMhBNuj5t3~v54OkqaY-Dz5VSozuFujGRcGAb3Eyrj6T5qbdNpN~DMGJfZzQHFPVvog-NfCdHC328fnNntJDd2NyAqCw4AiCQ0eR5rzYJRFFKr6~~al~v7-7BMw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Association (•) and dissociation (○) kinetics at 4 C of [3H]T binding to AR1 in brain nuclear extracts. Association kinetics were calculated by adding 10 nm[ 3H]T with or without 1 μm T and determining the amount of specific binding at various time points. Dissociation kinetics were calculated by incubating 4 nm[ 3H]T, with or without 500 nm T for determining nonspecific binding, for 6 h before adding 500 nm T, and binding was subsequently measured at various time points. Each data point is the average of three replicate assays ± sem.
![Association and dissociation (inset) kinetics at 4 C of [3H]DHT binding to AR2 in ovarian cytosolic extracts. Association kinetics were calculated by adding 0.6 nm [3H]DHT with or without 100 nm DHT. Each data point is the average of three replicate assays ± sem. Dissociation kinetics were calculated by incubating 1 nm[ 3H]DHT, with or without 100 nm DHT for determining nonspecific binding, for 8 h before adding 100 nm cold DHT, and binding was subsequently measured at various time points.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631005.jpeg?Expires=1710102366&Signature=GY59aOw9ELJCxQZpcQc-yGC50svkflQgykacNmld47a2KcqR3nLFzXmjx4CjT3zRgcGI5dnjZp0x5zmfwdtJ6I4lJCJ6AN2RGfHv0ew2z8X18IX5fMnqqW7B9SR5EkZkXRP7KOitN41Y3KExfMnIUafR0pV8cNaGeon3-mM6Mc7IXNfrIvzmZhxpNQMLwU~tOToxl-~RFkiMpbFN~SUHlWrcXtK0spjPQZnijGdNW46f3HGJ8r2ekOidhKA8NaSLbB98FrbxBeprmR5N-8YUoH1I4c8~bVjk7-UTUf2SAaXzCbyJn6SOm0pvUCecy8TUq-Mr-pANpstLGLMVYeJnNQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Association and dissociation (inset) kinetics at 4 C of [3H]DHT binding to AR2 in ovarian cytosolic extracts. Association kinetics were calculated by adding 0.6 nm [3H]DHT with or without 100 nm DHT. Each data point is the average of three replicate assays ± sem. Dissociation kinetics were calculated by incubating 1 nm[ 3H]DHT, with or without 100 nm DHT for determining nonspecific binding, for 8 h before adding 100 nm cold DHT, and binding was subsequently measured at various time points.
The complete dissociation of [3H]T bound to AR1 at 4 C occurred within 24 h, with a t1/2 of 2.6 ± 0.7 h (n = 3; Fig. 4), with no loss of receptor over the 30-h period (data not shown). The dissociation kinetics of[ 3H]T binding to AR1 were also identical when examined using buffer H-2 rather than either H-1 or E-1 (data not shown). This was in contrast to the dissociation of [3H]DHT from AR2, which was extremely slow, with a t1/2 of 45 ± 14 h (Fig. 5). The rate of dissociation increased dramatically at 20 C, with a t1/2 of 1.7 ± 0.4 h (n = 3); however, AR2 also degraded rapidly, with a greater than 50% loss of receptor binding within 12 h (data not shown).
The binding kinetics at 4 C of [3H]T to AR1 from the cytosolic fraction were similar to those from the nuclear fraction, with equilibrium binding occurring at 5 h with a t1/2 of 30 min and nearly complete dissociation occurring in 24 h with a t1/2 of 5.8 h. There was no loss in receptor binding over 24 h (data not shown).
AR1 and AR2 steroid specificity
The binding curves presented in Figs. 6 and 7 are parallel, indicating competitive binding between the various unlabeled competitors and [3H]T for AR1 and [3H]DHT for AR2, allowing the EC50 values to be determined and the RBAs to be calculated (Table 2, see also for key to steroid abbreviations). The steroid specificity studies demonstrated that AR1 was specific for androgens. High affinity binding (RBA, >10%) to AR1 was only observed with T, whereas DHT and the synthetic androgen MT had moderate affinity. AR1 demonstrated low affinity for 11-KT and mibolerone (Fig. 6A). AR1 had almost no affinity for the androgens androstenedione and MDHT (Fig. 6A) or the nonandrogenic steroids E2 and the C21 steroids (Fig. 6B). The steroid specificity of AR1 from the cytosolic fraction of brain homogenates was also characterized and demonstrated no differences from that of nuclear AR1 (data not shown). AR2 bound a much broader range of steroids than AR1 and bound with relatively high affinity all of the androgens examined except androstenedione (Fig. 7, A and B). DHT, with an EC50 of 0.8 nm, had the highest RBA of all of the nonsynthetic steroids tested (Table 2). E2 and certain C21 steroids bound to AR2 with low affinity, with EC50 values ranging from 20 nm for P4 and 21-P to 1300 nm for cortisol (Fig. 7B and Table 2).
![A and B, Competition curves of the binding of various natural and synthetic steroids to AR1 in brain nuclear extracts. Varying concentrations of unlabeled steroids were incubated with 4 nm [3H]T for 12–16 h. Each data point is the average of two to four assays with sems less than 10%. See Table 2 for key to steroid abbreviations.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631006.jpeg?Expires=1710102366&Signature=PqrH4lBeq3oyi8EWhGrOVZK8wfa9W7uO-~q1EjMwR0gztOttSvXaxXUAWSu8wROkYGK0r99A21963Tj-f0zpupHWlNmvPk6Vff6KC83CJgYa1TqeI8OzqViBIeVHN9kBN6X5heYtL0CwnKENin7cYz4dvmCtE9S6a472qCMALw7Nd0N2F~hrav30EknzFGS1W5hbgG6cJuObFGb71TiOtic~TLAZZV988zXQHEh6zsOLKU5XJWXoSHpgPH2hBREY91n5LIsIkX2JAZTgBPnrCWyl9GCYC4~3SNsFC6xzVEgny7E1ILO673~wadIKO~wuCgpFHwfDOi5ZQDJkMrnwhw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
A and B, Competition curves of the binding of various natural and synthetic steroids to AR1 in brain nuclear extracts. Varying concentrations of unlabeled steroids were incubated with 4 nm [3H]T for 12–16 h. Each data point is the average of two to four assays with sems less than 10%. See Table 2 for key to steroid abbreviations.
![A and B, Competition curves of the binding of various natural and synthetic steroids to AR2 in ovarian cytosolic extracts. Varying concentrations of unlabeled steroids were incubated with 1 nm [3H]DHT for 8–12 h. Each data point is the average of three to five assays with sems less than 5%. See Table 2 for key to steroid abbreviations.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631007.jpeg?Expires=1710102366&Signature=Ul-4ZN105g5YGL2PQPX5aTjZi4keAHypVZLbyCc-sSBZ8qLUizG4flWpseG7YBy5hb1LAYEtVWlSPl8ndFf-~bs1zExI3Sps-VhDjTBlaqqYzM1kgrr0zPqnZ41AaXiQpw2shXgYMdT8XlDcWbvJ~Ea37LZiBVQQieyJhFRLDwWTLLyX1l~eu~n~vkTS-pVNqonqj-LT4EUJY4p~QVssSdYMJj6-7n5GbrwGJKrlvytirfrn6B0M6BzFD2cWbAml-NOIuBGytZNvB1nSsEfKsSktN6ZHOMysw61gsBTKU-edKAhbVHNn~bVzvXiKVqcvxaxnNL4eS3SgCf79Vvp9zA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
A and B, Competition curves of the binding of various natural and synthetic steroids to AR2 in ovarian cytosolic extracts. Varying concentrations of unlabeled steroids were incubated with 1 nm [3H]DHT for 8–12 h. Each data point is the average of three to five assays with sems less than 5%. See Table 2 for key to steroid abbreviations.
Relative binding affinities of various steroids for brain nuclear AR1 and ovarian cytosolic AR2
| Steroid | EC50 (nm)a | RBA (%)b | ||
|---|---|---|---|---|
| AR1 | AR2 | AR1 | AR2 | |
| T (testosterone) | 2.5 | 1.2 | 100 | 67 |
| DHT (5α-dihydrotestosterone) | 40 | 0.8 | 6.3 | 100 |
| Androstenedione | ND | 50 | 0 | 1.6 |
| 11-KT (17β-hydroxy-4-androsten-3,11-dione) | 1800 | 5 | 0.14 | 16 |
| MT (17α-methyl-17β-hydroxy-4-androsten-3-one) | 90 | 2 | 2.8 | 40 |
| MDHT (17α-methyl-17β-hydroxy-5α-androstan-3-one) | ND | 0.8 | 0 | 100 |
| Mibolerone (7α,17α-dimethyl-17β-hydroxy-4-estren-3-one) | 2200 | 1.9 | 0.11 | 42 |
| E2 (17β-estradiol) | ND | 60 | 0 | 1.3 |
| P4 (progesterone) | 9000 | 20 | 0.028 | 4 |
| 20β-S (17,20β,21-trihydroxy-4-pregnen-3-one) | ND | 90 | 0 | 0.9 |
| 17,20β-P (17,20β-dihydroxy-4-pregnen-3-one) | ND | 40 | 0 | 2 |
| 21-P (deoxycorticosterone) | ND | 20 | 0 | 4 |
| Cortisol | ND | 1300 | 0 | 0.06 |
| Steroid | EC50 (nm)a | RBA (%)b | ||
|---|---|---|---|---|
| AR1 | AR2 | AR1 | AR2 | |
| T (testosterone) | 2.5 | 1.2 | 100 | 67 |
| DHT (5α-dihydrotestosterone) | 40 | 0.8 | 6.3 | 100 |
| Androstenedione | ND | 50 | 0 | 1.6 |
| 11-KT (17β-hydroxy-4-androsten-3,11-dione) | 1800 | 5 | 0.14 | 16 |
| MT (17α-methyl-17β-hydroxy-4-androsten-3-one) | 90 | 2 | 2.8 | 40 |
| MDHT (17α-methyl-17β-hydroxy-5α-androstan-3-one) | ND | 0.8 | 0 | 100 |
| Mibolerone (7α,17α-dimethyl-17β-hydroxy-4-estren-3-one) | 2200 | 1.9 | 0.11 | 42 |
| E2 (17β-estradiol) | ND | 60 | 0 | 1.3 |
| P4 (progesterone) | 9000 | 20 | 0.028 | 4 |
| 20β-S (17,20β,21-trihydroxy-4-pregnen-3-one) | ND | 90 | 0 | 0.9 |
| 17,20β-P (17,20β-dihydroxy-4-pregnen-3-one) | ND | 40 | 0 | 2 |
| 21-P (deoxycorticosterone) | ND | 20 | 0 | 4 |
| Cortisol | ND | 1300 | 0 | 0.06 |
ND, The EC50 could not be determined.
The EC50 of each compound is calculated from the competition curves (Figs. 6 and 7) and is equal to the concentration of competitor necessary to displace 50% of either 4 nm [3H]T from AR1 or 1 nm[ 3H]DHT from AR2.
RBA is calculated from the ratio of the EC50 values of either T or DHT and the various competitors. The RBA values of T and DHT were set at 100 for AR1 and AR2 binding, respectively.
Relative binding affinities of various steroids for brain nuclear AR1 and ovarian cytosolic AR2
| Steroid | EC50 (nm)a | RBA (%)b | ||
|---|---|---|---|---|
| AR1 | AR2 | AR1 | AR2 | |
| T (testosterone) | 2.5 | 1.2 | 100 | 67 |
| DHT (5α-dihydrotestosterone) | 40 | 0.8 | 6.3 | 100 |
| Androstenedione | ND | 50 | 0 | 1.6 |
| 11-KT (17β-hydroxy-4-androsten-3,11-dione) | 1800 | 5 | 0.14 | 16 |
| MT (17α-methyl-17β-hydroxy-4-androsten-3-one) | 90 | 2 | 2.8 | 40 |
| MDHT (17α-methyl-17β-hydroxy-5α-androstan-3-one) | ND | 0.8 | 0 | 100 |
| Mibolerone (7α,17α-dimethyl-17β-hydroxy-4-estren-3-one) | 2200 | 1.9 | 0.11 | 42 |
| E2 (17β-estradiol) | ND | 60 | 0 | 1.3 |
| P4 (progesterone) | 9000 | 20 | 0.028 | 4 |
| 20β-S (17,20β,21-trihydroxy-4-pregnen-3-one) | ND | 90 | 0 | 0.9 |
| 17,20β-P (17,20β-dihydroxy-4-pregnen-3-one) | ND | 40 | 0 | 2 |
| 21-P (deoxycorticosterone) | ND | 20 | 0 | 4 |
| Cortisol | ND | 1300 | 0 | 0.06 |
| Steroid | EC50 (nm)a | RBA (%)b | ||
|---|---|---|---|---|
| AR1 | AR2 | AR1 | AR2 | |
| T (testosterone) | 2.5 | 1.2 | 100 | 67 |
| DHT (5α-dihydrotestosterone) | 40 | 0.8 | 6.3 | 100 |
| Androstenedione | ND | 50 | 0 | 1.6 |
| 11-KT (17β-hydroxy-4-androsten-3,11-dione) | 1800 | 5 | 0.14 | 16 |
| MT (17α-methyl-17β-hydroxy-4-androsten-3-one) | 90 | 2 | 2.8 | 40 |
| MDHT (17α-methyl-17β-hydroxy-5α-androstan-3-one) | ND | 0.8 | 0 | 100 |
| Mibolerone (7α,17α-dimethyl-17β-hydroxy-4-estren-3-one) | 2200 | 1.9 | 0.11 | 42 |
| E2 (17β-estradiol) | ND | 60 | 0 | 1.3 |
| P4 (progesterone) | 9000 | 20 | 0.028 | 4 |
| 20β-S (17,20β,21-trihydroxy-4-pregnen-3-one) | ND | 90 | 0 | 0.9 |
| 17,20β-P (17,20β-dihydroxy-4-pregnen-3-one) | ND | 40 | 0 | 2 |
| 21-P (deoxycorticosterone) | ND | 20 | 0 | 4 |
| Cortisol | ND | 1300 | 0 | 0.06 |
ND, The EC50 could not be determined.
The EC50 of each compound is calculated from the competition curves (Figs. 6 and 7) and is equal to the concentration of competitor necessary to displace 50% of either 4 nm [3H]T from AR1 or 1 nm[ 3H]DHT from AR2.
RBA is calculated from the ratio of the EC50 values of either T or DHT and the various competitors. The RBA values of T and DHT were set at 100 for AR1 and AR2 binding, respectively.
AR1 and AR2 assay specificity
The presence of 100 nm MDHT, which binds with high affinity to AR2 and not AR1, did not alter the Kd or Bmax of [3H]T binding to either nuclear or cytosolic AR1 (results not shown), suggesting that the AR1 assay does not measure any specific binding of [3H]T to AR2. Similarly, no specific binding of [3H]DHT was detected in brain cytosolic fractions prepared in H-1 buffer (results not shown), further indicating that the assay designed to measure AR1 is specific for this receptor. A small amount of specific [3H]DHT binding was measured (Kd, 1.3 nm; Bmax, 0.02 nm) in brain tissue prepared in H-2 buffer, which contains Na2MoO4. However, this binding was not consistent and could only be estimated in one assay of three, suggesting that the levels of AR2 are too low to be accurately measured in the brain by this assay (Table 1).
Both [3H]DHT and [3H]T demonstrated saturable, high affinity binding to an ovarian cytosolic homogenate prepared with H-2 buffer, with Kd values of 0.5 and 0.8 nm and Bmax values of 25 and 27 pm, respectively. All of the specific binding of[ 3H]T could be blocked by MDHT, indicating that the specific binding of [3H]T was occurring via binding to AR2.
DNA-cellulose affinity chromatography
No peaks of binding were recovered in the 0.05-, 0.4-, or 2-m NaCl eluates of cytosolic brain tissue incubated with 6 nm [3H]T plus 500 nm T. This indicates that the [3H]T retained on the DNA-cellulose columns and eluted by 0.4 m NaCl represents specific binding of AR1 to the DNA (Fig. 8). Specific binding was recovered in the 0.4-m eluates after incubating samples for 1 h at 4 C on the DNA-cellulose columns, and binding was increased 3-fold upon heat activation (5 min at 30 C followed by 10 min at 4 C). No peaks of binding were recovered in the 0.05-, 0.4-, or 2-m NaCl eluates of cytosolic ovarian tissue, prepared with buffer H-2 without Na2MoO4, and incubated with 6 nm[ 3H]DHT plus 500 nm DHT. Therefore, the[ 3H]DHT retained on the DNA-cellulose columns and eluted by 0.4 m NaCl represents specific binding of AR2 to the DNA (Fig. 9). Specific binding was recovered in the 0.4-m eluates after incubating samples for 1 h at 4 C on the DNA-cellulose columns, whereas binding to the columns was eliminated when the samples were incubated and prepared with buffer H-2 with Na2MoO4. Binding was increased by one third upon heat activation (5 min at 30 C followed by 10 min at 4 C).
![DNA-cellulose chromatography of cytosolic AR1 binding to calf thymus DNA. Cytosolic AR1, prepared from whole brain tissue, was incubated with 6 nm [3H]T with or without 500 nm T for 12 h and charcoal stripped before loading onto DNA-cellulose columns (Bio-Rad minicolumns with a bed volume of 1.5 ml). Samples without cold T were incubated with DNA-cellulose at 30 C for 5 min followed by 10 min at 4 C (♦) or 1 h at 4 C (▴). Nonspecific binding to the DNA-cellulose columns was determined by adding 500 nm T and incubating the samples at 30 C for 5 min followed by 1 h at 4 C (○). Binding of AR1 to the DNA was determined by consecutively washing the columns under increasingly stringent conditions with H-2 buffer without Na2MoO4 and 0.05, 0.4, or 2 m NaCl. Each point represents the mean of two replicates.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631008.jpeg?Expires=1710102366&Signature=KbRM4hp-jjp9TF7ariYbQY-4ZRW-xCSxfAVhQnuA722chUIGSEKpfLJyFA-eoBhiLRfA8hrL3rqTjZHCby8-~WTGw1VV1GdebTTbsVACw5BGDrc537DJf7bI7gSA2hqP84QxVDVudv2Nc-KVZFI75RLKy7s9-e3pMLYOIgUOjGhD86~DOibVvkC-NQQSmd6mBE5rhA8w2PqPg4NJoQwid8XFpGsZWHGggbGDU7teJo8M9zbSUJ08kJiFhIgVm7B-rtPyNNxLk7NEEBWuKqBvC3WMahPl1NZFNAMdBz6gXY8InzQaht8KF~VmbAlxFu2A8GKE7RQxcem4~nBCBHmP4Q__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
DNA-cellulose chromatography of cytosolic AR1 binding to calf thymus DNA. Cytosolic AR1, prepared from whole brain tissue, was incubated with 6 nm [3H]T with or without 500 nm T for 12 h and charcoal stripped before loading onto DNA-cellulose columns (Bio-Rad minicolumns with a bed volume of 1.5 ml). Samples without cold T were incubated with DNA-cellulose at 30 C for 5 min followed by 10 min at 4 C (♦) or 1 h at 4 C (▴). Nonspecific binding to the DNA-cellulose columns was determined by adding 500 nm T and incubating the samples at 30 C for 5 min followed by 1 h at 4 C (○). Binding of AR1 to the DNA was determined by consecutively washing the columns under increasingly stringent conditions with H-2 buffer without Na2MoO4 and 0.05, 0.4, or 2 m NaCl. Each point represents the mean of two replicates.
![DNA-cellulose chromatography of cytosolic AR2 binding to calf thymus DNA. Cytosolic AR2 from ovarian tissue was prepared without Na2MoO4, incubated with 6 nm [3H]DHT and with or without 500 nm DHT for 12 h, and charcoal stripped before loading onto DNA-cellulose columns (Bio-Rad minicolumns with a bed volume of 3.0 ml). A, Samples were incubated with the DNA-cellulose for 1 h at 4 C without (♦) or with 500 nm DHT (○). B, Additionally, samples were incubated with DNA-cellulose at 30 C for 5 min followed by 10 min at 4 C (▴), and samples prepared with Na2MoO4 were incubated for 1 h at 4 C (□). Binding of AR2 to the DNA was determined by consecutively washing the columns under increasingly stringent conditions with H-2 buffer without Na2MoO4 and 0.05, 0.4, or 2 m NaCl.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631009.jpeg?Expires=1710102366&Signature=bEhL~XAcWwMrK3Wbes~tSqiTb74W0ajz6BTf5g0Z1aR4IXN4lh8NFWOleH1ah1nO1LCWSw5DDIFrjh-ZWFpuFyucyG0VR~T0ovny2fgnnYWtUB1LDEGWZzr5i5R1A~DM0xImmrFxOrfIJnWeZmWnUm7Z2OH9zBepqCruIzn3iDO9Bp3CK3OaBHBJ1dRijJDWb5Hwl15FG3KvH0Sr7I2gqlbShwlxfOr35tffbzqXHgSzd4DyLUgV5m3L98H5aC807w3U0-YLbD6ZRmrMMNQXva-33yWafC4tp2srQZ~hixgSWxuO9XAiTwEOsFpyPbtifIOnEt3IoJzoDO9dyrDPLw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
DNA-cellulose chromatography of cytosolic AR2 binding to calf thymus DNA. Cytosolic AR2 from ovarian tissue was prepared without Na2MoO4, incubated with 6 nm [3H]DHT and with or without 500 nm DHT for 12 h, and charcoal stripped before loading onto DNA-cellulose columns (Bio-Rad minicolumns with a bed volume of 3.0 ml). A, Samples were incubated with the DNA-cellulose for 1 h at 4 C without (♦) or with 500 nm DHT (○). B, Additionally, samples were incubated with DNA-cellulose at 30 C for 5 min followed by 10 min at 4 C (▴), and samples prepared with Na2MoO4 were incubated for 1 h at 4 C (□). Binding of AR2 to the DNA was determined by consecutively washing the columns under increasingly stringent conditions with H-2 buffer without Na2MoO4 and 0.05, 0.4, or 2 m NaCl.
Sucrose gradient centrifugation
The peak of specifically bound [3H]T (6 nm) from cytosolic extracts of brain tissue in the presence of 100 nm MDHT, indicating the presence of AR1, sedimented at 5.6–5.7S compared with the 4.6S [14C]BSA standard (Fig. 10A). The specifically bound[ 3H]DHT (6 nm) binding from cytosolic extracts of ovarian tissue sedimented as a single peak at 7.4–7.8S compared with the [14C]BSA standard of 4.6S (Fig. 10B). The initial peak of radioactivity represents free steroid that did not migrate through the gradient, as the first three fractions comprise the sample aliquot.
![Sedimentation profiles of cytosolic extracts of brain tissue containing specifically bound [3H]T (A) or cytosolic extracts of ovarian tissue containing specifically bound[ 3H]DHT (B) on a 5–20% sucrose gradient. The[ 14C]BSA standard migrates at 4.6S.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/endo/140/4/10.1210_endo.140.4.6631/1/m_ee0496631010.jpeg?Expires=1710102366&Signature=30J7n4lClcT244ejIL4X9Q8-yyF6tgVbd4mhf0LyR91c-7DEV61ueKSgPrrLmpt5eafgKin4mdXtq68a78vCieMjwOKDnN4YQZHKSCCbrzXMMrgvw0gX0aMWZ9fYAHMZBdaIogBKNY90V7-kaEOwPSTvkqxyj~qpWvEbHl0FJkowx0PwH1Fg0zr-2vkUMJF6VZ1vDIgW0jS579YSsWOyBkVkkGJmolTgwQj9Pt9ilnJ9THEkcTuWUi30gtSkOiTnMpr2363kKd6KnbfkNLdzqxblefnbmx8bcvPZjCNfJNIHhE6mwdVff7yeg7WXUueFhGwAiaEYUBkqcMOEEncUww__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Sedimentation profiles of cytosolic extracts of brain tissue containing specifically bound [3H]T (A) or cytosolic extracts of ovarian tissue containing specifically bound[ 3H]DHT (B) on a 5–20% sucrose gradient. The[ 14C]BSA standard migrates at 4.6S.
Discussion
The results of this study clearly demonstrate that Atlantic croaker possesses two distinct nuclear ARs. Both AR1 and AR2 demonstrate saturable, high affinity, and specific binding to androgens and are both nuclear receptors. AR1 was present in isolated nuclei from brain tissue, and AR2 was detected in crude chromatin pellets of ovarian tissue. Although AR1 has a rapid dissociation rate that facilitates the exchange of endogenous bound T with [3H]T in nuclear fractions, AR2 has an extremely slow dissociation rate, making it necessary to use an exchange assay to measure[ 3H]DHT binding. It was not possible to attain equilibrium conditions using heat shock to increase the rate of dissociation because of the rapid loss of AR2 binding at elevated temperatures. Therefore, the mercurial reagent mersalyl acid was used to reversibly disrupt the binding sites of nuclear steroid receptors as described by Traish et al. (33). AR1 and AR2 both bind to DNA-cellulose, further indicating that each is a nuclear steroid receptor. The increase in receptor binding to the DNA-cellulose upon incubating the columns at 30 C has been shown previously to be the result of receptor activation (38), a necessary step before DNA binding competence (1). The binding characteristics of AR1 and AR2 differ markedly from those of the steroid-binding protein in this teleost family (39). For example, binding was detected to both AR1 and AR2 in the presence of EDTA, which when present in conjunction with a thiol-reducing agent, such as monothioglycerol, eliminates binding to the steroid-binding proteins (39).
AR1 was found primarily in male and female brain tissue, and no specific binding was detected in ovarian tissue. In contrast, AR2, was measured in brain, ovarian, and testicular tissues. As a result of the differences in the binding affinities, physical characteristics, and distributions of the two receptors, we were able to measure specific binding to the receptors without cross-reactivity between the assays. In addition to the differences in distribution, the amounts of receptor in the tissues varied. AR1 binding in both nuclear and cytosolic fractions of brain tissue was nearly 2 orders of magnitude greater than that of AR2 in the brain cytosolic fractions. AR2 levels in ovarian tissue were 6 times higher than those in testicular tissues and nearly 20 times higher than those in brain tissues, but were lower than the amounts of AR1 in the brain. Both AR1 and AR2 were present during the reproductive and nonreproductive seasons of Atlantic croaker. The levels of cytosolic AR1 in the brain tissue of recrudesced fish were significantly higher (P < 0.05) than those of regressed fish, whereas the levels of nuclear AR1 were significantly (P < 0.05) lower, which suggests that there are fewer activated receptors in the brain of regressed fish. Changes in receptor concentration correlating with the seasonal reproductive cycle have also been demonstrated in goldfish brain AR (23). On the other hand, similar levels of cytosolic AR2 binding were detected in fully recrudesced and fully regressed croaker ovaries.
All of the nuclear steroid hormone receptors that have been characterized to date form large heterocomplexes with heat shock proteins (hsp) and other associated chaperone proteins (40). The in vivo functions of these receptor heterocomplexes are not fully understood, but they may protect the receptors from degradation by keeping them in a high affinity binding state and may also help to regulate nuclear trafficking (40). In vitro, the hsp can be prevented from dissociating from the receptors by the addition of the metal-oxoanion MoO42−. The stabilized heterocomplex prevents the steroid receptors from being rapidly degraded, an effect that was shown for AR2 as well as for other ARs (41) and other steroid receptors (42). Interestingly, the presence of MoO42− did not influence the receptor concentrations of AR1, suggesting that this receptor may interact with hsp in a manner different from that of AR2.
Nuclear steroid receptor hsp heterocomplexes sediment at 8–10S on a 5–20% sucrose gradient depending upon the receptor type (40). The 7.4–7.8S sedimentation coefficient for cytosolic AR2 is consistent with the presence of a receptor-hsp heterocomplex. On the other hand, the 5.5–5.6S sedimentation coefficient of AR1 corresponds more closely to the predicted 4–5S sedimentation coefficient of monomeric nuclear steroid receptors.
AR1 and AR2 show the most profound differences in their steroid specificities. The steroid specificity of AR1 indicates that this receptor is similar to those previously characterized in goldfish brain tissue (23) and in rainbow trout lymphocytes (24). T bound to each of these ARs with the highest affinity, followed by DHT with an order of magnitude less affinity. 11-KT demonstrated low affinity, and E2 and cortisol did not bind. Although both AR1 and the goldfish AR bound mibolerone with low affinity, the rainbow trout AR showed no affinity for this synthetic androgen.
Unlike AR1, which bound only T with high affinity, AR2 had high affinity for a broad spectrum of natural and synthetic androgens. This is similar to the androgen specificity of the AR described in coho salmon ovaries (25). Although the orders of RBAs are not identical between these ARs, there are important similarities that distinguish them from AR1-like receptors. AR2 and coho salmon AR both demonstrate a broad steroid specificity with high affinity for several androgens, including the androgens of physiological importance, T, DHT, and 11-KT. In addition, both receptors bind synthetic androgens with high affinity. Thus, within the salmonids, an AR1-like receptor has been described, the rainbow trout AR, and an AR2-like receptor has been described, the coho salmon AR. The presence of two ARs in Atlantic croaker, an advanced perciform fish belonging to the family Sciaenidae, and the likely occurrence of multiple receptors in salmonid fishes, which belong to a primitive teleost family Salmonidae, suggest that the presence of two ARs is a common feature among teleost fishes.
ARs have been characterized in all of the major classes of vertebrates from chondrichthyes to mammals. However, in most cases only a limited number of steroids were used to determine the steroid specificity of a receptor; as a result, it is difficult to resolve whether they are more like croaker AR1 or AR2. The AR characterized in shark testes demonstrated a high binding affinity for progesterone (43), a property that makes it unique among the vertebrate ARs. The AR found in salamander testes had high affinity for T and DHT and a low affinity for other steroids as well as the synthetic androgens mibolerone and R1881 (44). Likewise, in the Harderian gland of the frog, the AR showed high affinity for T and DHT, but none for mibolerone (45), making these receptors more similar to AR1. This leaves open the possibility that AR1-like receptors are also present in vertebrate groups other than teleosts.
The AR has been most extensively characterized in mammals (mAR), and it has been demonstrated both biochemically and physiologically that there is only a single mAR that mediates the effects of both T and DHT (46, 47). There are numerous similarities between AR2 and the mAR, suggesting that the teleostean AR2 is related to the mAR. Both ARs are similar in their biochemical properties, with high affinities for DHT, very slow dissociation rates (47), and the importance of hsp for increasing receptor stability (48). The steroid specificities of AR2 and the mAR are similar; both receptors bind DHT better than T, possess low affinity for estrogens and progestogens, and bind synthetic androgens such as mibolerone with high affinity (47). In addition, the antiandrogens, both steroidal and nonsteroidal, which are designed to specifically bind to mAR, bind only to AR2 and show no affinity for AR1.
The AR is distinguished by its ability to bind androgens and mediate androgen-induced male specific behavior, sexual development, and maturation (49, 50). The presence of ARs in the female reproductive tissues of many vertebrates, from fish to mammals, suggests that androgens also play an essential role in regulating female physiology (51–53). However, the presence of two ARs in teleosts adds complexity to the question of receptor function. The differential binding specificities and tissue distributions of AR1 and AR2 suggest that the two receptors may have different functions.
The presence of two ARs can explain the differences in physiological activity of T and 11-KT in male teleosts. The effects of 11-KT will probably be mediated through AR2, whereas T may act via both receptors. Steroidogenic enzymes, in particular 5α-reductase, may play a role in determining which AR is activated. However, the physiological importance of DHT in teleosts has not been thoroughly examined. In mammals, the steroidogenic enzyme 5α-reductase has an integral role in amplifying the androgenic signal in a tissue-specific manner by converting T to the more active androgen DHT (54). To begin to understand the physiological roles of AR1 and AR2, it will be necessary to examine cell- and tissue-specific distributions and activities of both the receptors and the steroidogenic enzymes.
In summary, the biochemical characterization of two nuclear ARs, AR1 and AR2, are described in this paper. These receptors demonstrate different tissue distributions, steroid specificities, and physical characteristics. The results provide an explanation for the differential physiological activities of the two teleost androgens, T and 11-KT. In the future it will be important to determine the DNA sequences of these two receptors to understand their evolutionary relatedness and to investigate their cellular and tissue distributions. Our understanding of the physiological functions of androgens needs to be reevaluated in view of the demonstration of multiple ARs in a single vertebrate species.
This work was supported by NIH Grant ESO-4214 (to P.T.) and in part by the E. J. Lund Fellowship (to T.S.). This work was submitted in partial fulfillment of the Ph.D. requirements of the Marine Science Department of the University of Texas (Austin, TX).
Acknowledgments
We thank Upjohn Laboratories for their donation of mibolerone. In addition, we thank W. R. and S. Lawson for their assistance with fish care, and the members of the P. Thomas laboratory for assistance with collecting fish.
