Membrane androgen receptors have been biochemically characterized in only a few vertebrate species to date. Therefore, the purpose of the current study was to comprehensively investigate the binding characteristics of a putative membrane androgen receptor in the ovary of the teleost, Atlantic croaker (Micropogonias undulatus). Specific androgen binding to an ovarian plasma membrane fraction was demonstrated using a radioreceptor assay protocol consisting of a short-term incubation with [(3)H]testosterone (T) and subsequent filtration of bound steroid from free steroid. Saturation and Scatchard analyses of T binding to an ovarian plasma membrane fraction indicated the presence of a single, high-affinity (K(d) = 15.32 +/- 2.68 nM [mean +/- SEM]), low-capacity (B(max) = 2.81 +/- 0.31 pmol/mg protein), androgen-binding site. Specific androgen binding to the receptor was readily displaceable, and the association and dissociation kinetics were rapid (half-time = 3.7 +/- 1.7 and 4.7 +/- 0.2 min, respectively). Competitive binding assays showed that 5alpha-dihydrotestosterone, T, and 11-ketotestosterone had relative binding affinities (RBAs) of 193%, 100%, and 13%, respectively, whereas none of the C(18) or C(21) steroids tested bound with high affinity except for progesterone (RBA = 191%). This androgen-binding moiety with high affinity for progesterone is unlikely to mediate the physiological actions of progestins in croaker, because it has low binding affinity for fish progestin hormones. Androgen-binding sites were also detected in membrane fractions of the brain, liver, kidney, and drumming muscle, whereas little or no binding was detected in the trunk muscle, heart, gills, or intestine. Receptor levels increased 10-fold during ovarian recrudescence, reaching maximum levels in fully mature ovaries, which suggests a likely physiological role for this receptor during the reproductive cycle of female croaker. It is concluded that the androgen-binding moiety identified in the plasma membrane fraction of Atlantic croaker ovarian tissue fulfils all the criteria for its designation as a steroid receptor.

The presence of androgen receptors in the ovaries of several vertebrate species, including Atlantic croaker, suggests that androgens may have important roles in ovarian function. In the current study the effects of androgens on ovarian steroidogenesis in Atlantic croaker were investigated. Addition of 17beta-hydroxy-5alpha-androstan-3-one (DHT), 11-ketotestosterone (11-KT), or Mibolerone to ovarian incubations caused dose-dependent decreases in gonadotropin-stimulated in vitro estradiol production, which was not reversed by cotreatment with the antiandrogens, cyproterone acetate or 1,1-dichloro-2,2-bis(p-chlorophenyl) ethylene. Androgen treatment also caused significant decreases in estradiol production in the presence of 17-hydroxyprogesterone, which suggests that the site of androgen action is downstream of this steroid in the steroidogenic pathway. The mechanism of androgen action on ovarian steroidogenesis was also investigated. Coincubation with actinomycin D did not reverse the inhibitory effect of the androgens, which suggests that the mechanism of androgen action is nongenomic. An androgen conjugated to bovine serum albumin (DHT-BSA), which does not enter the cell, also caused inhibition of estradiol production in vitro, indicating that the androgen is acting at the cell surface. In addition, time course experiments revealed that the androgen action is rapid; 5-min exposure to DHT was sufficient to cause a significant reduction in estradiol production. Finally, preliminary evidence was obtained for the existence of a high-affinity, low-capacity androgen binding site in croaker ovarian plasma membranes. These studies suggest that androgens can down-regulate estrogen production in croaker ovaries via a rapid, cell surface-mediated, nongenomic mechanism.

The present study was designed to test whether phospholipase-A2 stimulatory protein (PLSP) has any role during androgen-induced masculine differentiation. Thus, an investigation was made to identify such a protein in the fetal genital tract and to test whether this protein can produce masculinization in vitro. Fetal tracts (15/batch) containing genital ducts and urogenital sinus from male, female, and testosterone-exposed (40 mg/kg.day, from days 13-17 of gestation) female embryonic mice on day 18 of gestation were fractionated using Bio-Rad P-100 gel filtration, DEAE-cellulose, and carboxymethyl-Sephadex chromatography. Phospholipase-A2 (PLA2) stimulatory activity was identified at every step of purification. The final preparation stimulated both bee venom and mouse genital PLA2; however, it had no effect on PLC. The preparation lost its PLA2 stimulatory activity after pronase treatment. The partially purified PLSP fraction produced two bands (63K and 55K), as determined by sodium dodecyl sulfate-gel electrophoresis, and its PLA2 stimulatory activity appeared at the region of 55K on a P-100 gel filtration column. PLSP was also identified in female and testosterone-exposed female genital tracts. However, the specific activity of the female PLSP was much lower than that of the male or testosterone-exposed females. Sodium dodecyl sulfate-gel analysis of 2-3 micrograms partially purified PLSP revealed the presence of a faint 55K band in the females compared to the presence of a darker 55K band in the male and testosterone-exposed female. The intensity of the 63K band was similar in both sexes. PLSP from the male and testosterone-exposed females maintained and stimulated the Wolffian duct, whereas PLSP from the female tract had no masculinizing effect. Thus, the masculinizing activity of the PLSP preparation appears to correlate with its PLA2 stimulatory activity and 55K band intensity, suggesting the role of PLA2 stimulatory protein in masculine differentiation.