Antigonadotropic effects of the bovine ovarian gonadotropin-releasing hormone-binding inhibitor/histone H2A in rat luteal and granulosal cells

A gonadotropin-releasing hormone (GnRH)-binding inhibitor (GnRH-BI) was purified from bovine ovaries and identified as histone H2A. In the present studies, the biological effects of partially purified and purified ovarian GnRH-BI, as well as calf thymus histone H2A, were examined in rat ovarian cells. Since GnRH has direct antigonadotropic actions on these cells, the effects on luteinizing hormone-stimulated cAMP accumulation in luteal cells and follicle stimulating hormone-induced cAMP and progesterone production in granulosal cells were evaluated. Antigonadotropic activity in both luteal and granulosal cells coeluted directly with GnRH-BI activity during purification from bovine ovaries, and the antigonadotropic effects were dose dependent and reversible. In contrast to GnRH, GnRH-BI maximally inhibited gonadotropin responses and the effects of GnRH-BI were not blocked by a GnRH antagonist. The purified ovarian GnRH-BI and calf thymus histone H2A had identical antigonadotropic properties, and the half-maximal concentrations for inhibiting the gonadotropin responses of granulosal and luteal cells was 2 and 5 microM, respectively. In conclusion, the ovarian GnRH-binding inhibitor, identified as histone H2A, not only inhibits the high affinity binding of GnRH to rat ovarian membranes but also evokes GnRH-like antigonadotropic responses in rat ovarian cells that do not appear to be mediated by binding to GnRH receptors.     

A gonadotropin-releasing hormone-binding inhibitor from bovine ovaries. Purification and identification as histone H2A

Bovine, ovine, rat, and human ovaries contain a protein defined as gonadotropin-releasing hormone (GnRH)-like because it reversibly inhibits the high affinity binding of GnRH to rat ovarian membranes, but these same tissues contain little, if any, detectable GnRH. In the present study this GnRH-binding inhibitor (GnRH-BI) was purified from bovine ovaries by a combination of reversed-phase high pressure liquid chromatography, cation-exchange, and gel filtration chromatography. Purification was monitored by analysis of GnRH-like activity in the highly specific rat ovarian membrane receptor assay. Amino acid composition and partial sequence analysis indicated that the purified ovarian GnRH-BI is histone H2A. Calf thymus histone H2A and the purified ovarian protein showed identical dose-dependent (ID50 = 2 microM), competitive, and reversible inhibitory effects on GnRH binding to rat ovarian membranes. The inhibitory effects could not be explained by charge interactions alone since spermine and spermidine at 10-fold higher concentrations did not inhibit GnRH binding. Furthermore, the effects showed specificity since EGF binding to rat ovarian membranes was not inhibited. It is possible that the inhibition of GnRH binding by the purified ovarian fraction was due to low level contamination by a highly active binding inhibitor. However, based on the variety of different purification procedures, the identical effects seen with histone H2A, and the absence of other proteins on gel electrophoresis, we conclude that the ovarian GnRH-BI is probably histone H2A.     

Mechanisms for the antigonadotropic action of the ovarian gonadotropin-releasing hormone-binding inhibitor protein/histone H2A on ovarian cells.

A GnRH-binding inhibitor (GnRH-BI) was recently purified from bovine ovaries. On the basis of amino acid composition and partial sequence analysis this antigonadotropic GnRH-BI was identified as histone H2A. In the present study the mechanism for the antigonadotropic action of histone H2A was examined and compared to that of GnRH and poly-L-lysine. The potential sites examined were the receptor-coupled pathway of second message synthesis including receptor binding of hormone, G protein activation, and adenylyl cyclase activation. Histone H2A inhibited (ID50 = 2 microM) the binding of hCG by membrane receptors from luteinized rat ovaries in a noncompetitive and dose-dependent manner. The binding of FSH by membrane receptors from immature rat ovaries was not inhibited by histone H2A. Binding of GnRH by pituitary membrane receptors was inhibited by histone H2A, and the ID50 of 8 microM was similar to that previously observed for GnRH binding sites in rat ovarian membranes. No high-affinity binding of histone H2A by rat ovarian membranes was detected. Near-maximal doses of histone H2A (7 microM), poly-L-lysine (10 microM), and GnRH (1 microM) inhibited LH-stimulated cAMP production in isolated rat luteal cells. Inhibition by H2A and poly-L-lysine was larger than by GnRH. Furthermore, histone H2A and poly-L-lysine inhibited cholera toxin (CT)-stimulated cAMP production, but GnRH did not. Like GnRH, neither histone H2A nor poly-L-lysine inhibited forskolin (FK)-stimulated cAMP production. In isolated rat granulosa cells, histone H2A and poly-L-lysine inhibited FSH-, CT-, and FK-stimulated cAMP production.(ABSTRACT TRUNCATED AT 250 WORDS)     

Binding of gonadotropin releasing hormone agonist to rat ovarian granulosa cells

We have previously shown that gonadotropin releasing hormone (GnRH) and its agonists inhibit ovarian functions by a direct action on ovarian granulosa cells $\text{in vitro}$. A labeled GnRH agonist, [des-Gly¹⁰, D-Ser (TBu)⁶, Pro⁹-NHEt]GnRH, was used here to examine the possibility that these inhibitory actions of GnRH were mediated through specific receptors which recognize GnRH. Ovarian membrane fractions obtained from immature, hypophysectomized diethylstilbesterol-treated rats were incubated with the ¹²⁵I-GnRH agonist and specific binding was determined by a filtration assay. Stereospecific, high affinity binding was detected in the ovarian membranes; the dissociation constant for the labeled GnRH agonist was determined to be 0.84 ± 0.33 × 10^?10 M and the binding capacity was calculated to be 12.9 fmol/mg protein, or 0.142 fmol/μg DNA. The binding affinity for the GnRH decapeptide was 3.3 times lower than that of the GnRH agonist whereas two GnRH partial peptides did not compete for the ¹²⁵I-agonist binding. After sequential treatment with FSH, LH and prolactin to the hypophysectomized female rats, the ovarian GnRH binding capacity increased per ovary, but decreased per mg ovarian protein.Furthermore, ovarian granulosa cells were isolated and their binding capacity was determined to be 25.2 fmol/mg protein, or 0.133 fmol/μg DNA, suggesting that the granulosa cells contain GnRH binding sites. Thus, this report demonstrates the presence of stereospecific, high affinity GnRH binding sites in the rat ovarian granulosa cells.     

Effect of histone H2A on progesterone production by bovine luteal cells

Histone H2A competitively inhibits binding of GnRH to high affinity rat ovarian receptor sites and blocks gonadotropin-stimulated steroid and cAMP accumulation during culture of rat granulosal or luteal cells. The objective of our study was to examine the progesterone suppressive effects of histone H2A on bovine luteal cells. In the first study, luteal cells were treated at Time = 0 h with a partially purified preparation of bovine ovarian histone H2A (3 ng GnRH equivalents, 800 micrograms protein), equivalent amounts of GnRH (3 ng), or BSA (800 micrograms) and incubated for a total of 4 h. At Time = 2 h, cells were treated with 5 ng bovine LH (bLH) or with medium. Histone H2A completely blocked both basal and LH-induced accumulation of progesterone compared with untreated cultures or cultures treated with bLH. Neither BSA nor GnRH suppressed LH-induced progesterone accumulation. In the second study, histone H2A was added to cultures at Time = 0 h and bovine luteal cells were cultured for 8 h. After 2 h of treatment, histone H2A (3 ng GnRH equivalents) was removed from selected cultures and replaced with fresh medium. Four hours later cultures were treated with 5 ng bLH or medium. LH treatment of cultures from which histone H2A had been removed resulted in an increase in accumulation of progesterone compared with control cultures treated throughout the treatment period with histone H2A. The third study examined the effect of 9-181 pg GnRH equivalents (1.7-34 micrograms protein) of a highly purified preparation of bovine ovarian histone H2A on basal and LH-induced progesterone production during 2 or 3 h of culture.(ABSTRACT TRUNCATED AT 400 WORDS)     

Action of gonadotropin-releasing hormone II on the baboon ovary

The content, binding affinity, and bioactivity of chicken II GnRH (GnRH II) and a stable analogue of GnRH II (GnRH II analogue) in the baboon ovary were studied. Although mammalian GnRH is rapidly degraded by baboon ovarian extracts, we designed a GnRH II analogue that is stable to ovarian enzymatic degradation. This analogue binds to the ovarian membranes with high affinity (41 +/- 3 nM), having 20-fold the affinity of a potent mammalian GnRH analogue. The bioactivity of GnRH II and this GnRH II analogue on the regulation of ovarian progesterone release was compared with that for a potent mammalian GnRH analogue using a baboon granulosa cell culture system. Both GnRH II and GnRH II analogue produced significant inhibition of progesterone release from the granulosa cells (P < 0.03 and P < 0.005, respectively), with a greater reduction observed using the GnRH II analogue. After 24 h in culture, this GnRH II analogue produced a 59% +/- 5% inhibition of progesterone with a concentration as low as 1 nM. Maximal inhibition of 75% +/- 1% was attained with 10 nM GnRH II analogue. The endogenous GnRH II content in the baboon ovary was 5-14 pmoles/g protein. The release of endogenous GnRH II from granulosa cells was observed throughout the 48 h in culture. These studies demonstrated the presence of high enzymatic activity for the degradation of mammalian GnRH in the ovary, whereas this GnRH II analogue was stable. High-affinity binding sites for this GnRH II analogue were also found. GnRH II and this GnRH II analogue can regulate progesterone production from baboon granulosa cells, suggesting that GnRH II is a potent regulator of ovarian function.     

Identification of gamma-aminobutyric acid and its binding sites in the rat ovary

Gamma-aminobutyric acid (GABA), GABA synthesizing enzyme and GABA binding sites were measured in rat ovaries. The concentration of GABA in the ovary (0.56 μg/mg protein) was less than that in the brain (1.2–3.4 μg/mg protein), but was six-fold higher than any other non-neuronal tissue examined. Glutamate decarboxylase, the GABA synthesizing enzyme was also found in high concentrations in whole ovarian homogenate but not in enriched ovarian granulosa cells, testis, anterior pituitary or muscles. Furthermore, high affinity (Kd = 15–21 nM), specific GABA binding sites were identified in the ovaries by specific [³H]muscimol binding and the majority of GABA binding sites were associated with the granulosa cells. These data suggest a possible role of GABA in the regulation of ovarian functions.     

GnRH-like proteins in cows: concentrations during corpora lutea development and selective localization in granulosal cells

Gonadotropin-releasing hormone (GnRH)-like proteins with anti-gonadotropic properties were recently discovered in the ovaries of several species, including humans. Since neither GnRH receptors nor GnRH are in bovine ovarian tissue, we examined, in the present studies, whether concentrations of GnRH-like proteins varied during development of the corpus luteum (CL) and whether GnRH-like proteins were selectively localized in ovarian cells of cows. For these studies, GnRH-like proteins were extracted from various ovarian and nonovarian tissues and fluids and fractionated for hydrophobic interaction chromatography. A highly specific and sensitive radioreceptor assay (RRA) was used to quantify concentrations of GnRH-like proteins. The major findings of these studies demonstrated that 1) the amount of GnRH-like proteins in the corpus luteum (CL) was proportional to the weight of the CL; 2) the concentration of GnRH-like proteins in luteal tissue decreased during development of the CL; 3) GnRH-like proteins were in ovarian and numerous nonovarian tissues, but were not in the heart, plasma, or follicular fluid; 4) the retention time for GnRH-like proteins following high-pressure liquid chromatography (HPLC) varied with the tissue source; and 5) compared with all other tissues, the greatest concentration of GnRH-like proteins was in granulosal cells. We concluded that the concentration of GnRH-like proteins in luteal cells decreased during development of the CL, and that a specific GnRH-like protein was selectively localized in bovine granulosal cells.     

Production of activin-binding protein by rat granulosa cells in vitro

We have developed an assay method for activin-binding protein, which exploits its high affinity for sulfated polysaccharides. We used this method to investigate the production of activin-binding protein by rat ovarian granulosa cells, in vitro. The production of activin-binding protein by granulosa cells was dependent on the cell density; the maximum was observed at 6 x 10(5) cells/ml. Follicle-stimulating hormone (FSH), but not luteinizing hormone (LH), enhanced production significantly. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and ligand blotting analyses of the activin-binding protein secreted by rat granulosa cells demonstrated it was the same protein molecule as that purified from rat ovaries. It is inferred from these results that the granulosa cell is a source of ovarian activin-binding protein and that its secretion is regulated by FSH.     

Gonadotropin-releasing hormone receptor binding in bovine anterior pituitary

Synthetic gonadotropin-releasing hormone (GnRH) was monoiodinated at a high specific radioactivity with 125I. The iodinated hormone retained full biological activity as assessed by the release of luteinizing hormone in vitro from bovine anterior pituitary tissue slices. Specific binding of 125I-labeled gonadotropin-releasing hormone of high affinity and low capacity was obtained using dispersed bovine anterior pituitary cells. The binding had sigmoid characteristics, compatible with the presence of more than one binding site. The subcellular fraction responsible for binding was identified with the plasma membranes. However, significant binding also occurred in the secretory granules fraction. The plasma membranes were solubilized with sodium dodecyl sulfate. Using gonadotropin-releasing hormone covalently coupled to a solid phase, a protein was purified by an affinity technique from the solubilized plasma membrane preparation which possessed similar binding propperties as plasma membranes, both intact and solubilized. The protein migrated as a single component on polyacrylamide gel in sodium dodecyl sulfate and the estimated molecular weight was 60 000. The character of the gonadotropin-releasing hormone concentration dependence binding as well as association kinetics were multiphasic and suggested the presence of more than one binding site. When analyzed by the Hill plot, the Hill coefficient of all binding curves was always greater than one which is compatible with positive cooperativity. This was further supported by the dissociation studies where the dissociation rate was inversely proportionate to both the gonadotropin-releasing hormone concentration and the time interval during which the gonadotropin-releasing hormone-gonadotropin-releasing hormone receptor protein complex was formed. Using difference chromatography, aggregation of the purified gonadotropin-releasing hormone receptor protein was demonstrated to occur upon its exposure to gonadotropin-releasing hormone. The formed macromolecular complexes bound preferentially 125I-labeled gonadotropin-releasing hormone. It is concluded that a single receptor protein is responsible for gonadotropin-releasing hormone binding in the bovine anterior pituitary. It is a part of the plasma membranes. Its interaction with gonadotropin-releasing hormone provokes transitions of the protein into different allosteric forms and this may be related to the biological effect of gonadotropin-releasing hormone on gonadotropin secretion.     

Down-regulation of mt1 melatonin receptors in rat ovary following estrogen exposure.

In this study, we have demonstrated that 2-[125I]-iodomelatonin binds specifically to rat ovarian granulosa cell (GC) membranes with high affinity (KD=83 pM; Bmax=3.28 fmol/mg protein). Using immunoblot analysis and an anti-mt1 melatonin receptor antibody, we have also detected mt1 melatonin receptors in rat ovary. Because melatonin has been reported to alter the steroidogenic responses of ovarian tissues to gonadotropins, a physiological role for intra-ovarian melatonin may exist. Thus, in order to investigate a possible intra-ovarian role for melatonin, we have used both an in vivo and in vitro model of follicular development. Treatment of immature (day 21) female rats with estradiol (E; 0.2 mg/d x 3 d; subcutaneous) was used to induce follicular growth. Membranes from both untreated (U) and E-treated animals' ovaries contained high-affinity 2-[125I]-iodomelatonin (I-MEL) binding sites (Kd=83 and 23 pM, respectively). Estradiol treatment in vivo caused a significant decrease (P<0.05) in binding of 2-[125I]-iodomelatonin to ovarian membranes with untreated animals' ovaries having a Bmax=3.28 fmol/mg protein vs. estradiol-treated animals' ovaries having a Bmax=0.92 fmol/mg protein. In addition, following Estradiol treatment, mt1 melatonin receptors in rat ovary were down-regulated (approximately 95%) using immunoblot analysis. Granulosa cells isolated from E-treated rats were further matured in vitro with testosterone (T) and the pituitary gonadotropin follicle-stimulating hormone (FSH). Granulosa cells were cultured with either T (10 ng/ml) or FSH (5.71 ng ovine FSH-20/ml) alone, or both FSH and T for 48 h. There was no statistically significant specific binding of 2-[125I]-iodomelatonin to GC membranes cultured with T or FSH alone. However, following a 48-h exposure to FSH and T in vitro specific 2-[125I]-iodomelatonin binding occurred with total 2-[125I]-iodomelatonin binding =3.15 [corrected] fmol/mg protein. Therefore, the existence of hormonally-regulated expression of high-affinity melatonin binding sites suggests that melatonin may have an important intra-ovarian physiological role.     

Gonadotropin releasing hormone treatment of dairy cows with ovarian cysts: II. Histology of ovarian cyst walls

Dairy cows known to have ovarian cysts were assigned to receive either sterile water or 100 μg GnRH (5 cows/group). The ovaries were removed 9 to 13 days post-treatment and prepared for histologic study. The cyst walls of ovaries removed from cows following GnRH treatment were 2 to 15 times thicker than those from cows treated with water due to apparent luteinization of cells of theca interna. Following GnRH treatment, the cells in the outermost portion of the theca interna exhibited the greatest morphologic evidence of secretory activity whereas the innermost zone was hyalinized due to poor blood supply. The luteal cells in ovarian cystic cows treated with GnRH were similar to the small luteal cells in normal corpora lutea. In summary, GnRH treatment of cows with ovarian cysts appeared to induce luteinization of the cells in the outermost portion of the theca interna.     

Localization of putative gonadotrophin releasing hormone receptor protein in the anterior pituitary

Specific binding of a fully biologically active 125I-gonadotrophin releasing hormone (GnRH) to isolated anterior pituitary cells is time dependent, saturable and the concentration dependent binding curves exhibit positive cooperativity. Binding to intact or solubilized plasma membranes and an affinity purified GnRH receptor protein reveals in all instances multiple high affinity binding sites. Thus, GnRH receptor protein appears to be an intrinsic constituent of the cell membrane, and perhaps, other membranous organelles. To investigate the latter, the binding of 125I-GnRH to various subcellular fractions was studied and its affinity and time requirements determined. GnRH binding to plasma membranes and secretory granules was to multiple high affinity sites, while that to nuclei and microsomes was to a single high affinity site. Binding was 1.83 +/- 0.07, 0.78 +/- 0.04, 0.31 +/- 0.03 and 0.27 +/- 0.03 fmol micrograms-1 protein for isolated plasma membranes, secretory granules, microsomes and nuclei, respectively, after 30 min incubation with 10(-9) M GnRH. The magnitude of binding to microsomes did not change during the incubation period. It did not show any decrease (p greater than 0.05) in isolated nuclei and plasma membranes, except for the 24 h time period, when a significant drop (p less than 0.001) was seen. Binding to the secretory granule fraction culminated at 15 min and then decreased (p less than 0.001) steadily to a non-detectable level at 24 h. Thus GnRH receptor protein or its portion may be an integral part of some membranous particles in the anterior pituitary cells.(ABSTRACT TRUNCATED AT 250 WORDS)     

Gonadotropin-releasing hormone binding sites in ovaries of normal cycling and persistent-estrus rats

The gonadotropin-releasing hormone (GnRH) binding capacity in ovaries and pituitaries of normal cycling rats at different stages of the estrous cycle and in ovaries of persistent-estrus rats was measured using radioligand-receptor assay (RRA). Persistent estrus was induced either by neonatal administration of testosterone propionate (1.25 mg s.c.) on the second day of life or by a hypothalamic suprachiasmatic frontal cut made with Halász' knife. All animals were killed during the critical period (1400-1600 h), and GnRH receptor was assayed. GnRH receptor levels in both ovaries and pituitaries changed during the estrous cycle. The total number of ovarian GnRH binding sites was significantly higher in proestrus than in diestrus 1, the stage in which the lowest level was found. When binding sites were expressed in fmol/mg ovary, the highest level was observed in diestrus 2; however, no changes were observed during the estrous cycle when GnRH binding sites were expressed as fmol/mg protein. Changes noted were very similar to those demonstrated in pituitary GnRH receptors in our present and previous experiments. Higher levels of pituitary binding sites were found in diestrus 2 and proestrus than in estrus and diestrus 1. The changes in the GnRH receptor levels were more striking in the pituitary than in the ovaries. It appears that the total number of ovarian GnRH binding sites was not altered in either of the two persistent-estrus groups, but that their concentration was significantly higher (expressed in fmol/mg ovary or fmol/mg protein) than on any day during the estrous cycle.(ABSTRACT TRUNCATED AT 250 WORDS)     

The discovery of novel small molecule non-peptide gonadotropin releasing hormone (GnRH) receptor antagonists

A novel series of non-peptide derivatives 1, 14, and 15 that bind with high affinity to the human GnRH receptors is discussed. The discovery was made from screening our in-house libraries that contained the active structure 2 along with a trace amount of a second active structure 1 that was derived from an acid-induced rearrangement. From this structure type 1, a series of guanidine and non-guanidine containing analogues were prepared and tested as GnRH receptor antagonists. Compounds derived from this series bind to both human and rat GnRH receptors and antagonize GnRH-mediated increases in inositol phosphate production in cells containing recombinant human receptors. These compounds or their analogues may be useful as therapeutic agents for the treatment of hormone-dependent pathologies including prostate, breast and ovarian cancers.     

Characterization of a Mr 25,000 basic fibroblast growth factor form in adult, regenerating, and fetal rat liver

A heparin-binding Mr 25,000 immunoreactive bFGF-like protein (ir-bFGF) is recognized in adult rat liver extract by affinity-purified polyclonal anti-human placental bFGF antibodies. Hepatic levels of this protein increase 4-fold in regenerating rat liver during the first 48 h after partial hepatectomy. Also, they appear to be higher in embryonic than in newborn or in adult rat liver. Mr 25,000 ir-bFGF from regenerating rat liver, partially purified by heparin-affinity chromatography, induces plasminogen activator activity and cell proliferation in transformed fetal bovine aortic endothelial GM 7373 cells and competes with Mr 18,000 [125I]bFGF for the binding to high affinity bFGF receptors. The data indicate the presence in rat liver of a high molecular weight form of bFGF whose expression is modulated during embryonic development and liver regeneration.     

Suppression of inhibin A biological activity by alterations in the binding site for betaglycan

Inhibins A and B negatively regulate the production and secretion of follicle-stimulating hormone from the anterior pituitary, control ovarian follicle development and steroidogenesis, and act as tumor suppressors in the gonads. Inhibins regulate these reproductive events by forming high affinity complexes with betaglycan and activin or bone morphogenetic protein type II receptors. In this study, the binding site of inhibin A for betaglycan was characterized using inhibin A mutant proteins. An epitope for high affinity betaglycan binding was detected spanning the outer convex surface of the inhibin alpha-subunit. Homology modeling indicates that key alpha-subunit residues (Tyr(50), Val(108), Thr(111), Ser(112), Phe(118), Lys(119), and Tyr(120)) form a contiguous epitope in this region of the molecule. Disruption of betaglycan binding by the simultaneous substitution of Thr(111), Ser(112), and Tyr(120) to alanine yielded an inhibin A variant that was unable to suppress activin-induced follicle-stimulating hormone release by rat pituitary cells in culture. Together these results indicate that a high affinity interaction between betaglycan and residues Val(108)-Tyr(120) of the inhibin alpha-subunit mediate inhibin A biological activity.     

High affinity GnRH binding to testicular membrane homogenates

The binding of GnRH to membrane homogenates from collagenase-dissociated normal rat interstitial cells is shown to be of high affinity and specific. The dissociation constant for the radioligand used, [¹²⁵I-Tyr⁵, D-Ala⁶, N^αMeLeu⁷, Pro⁹-NEt]-GnRH is 0.26 ± 0.02 nM and the number of binding sites is 17 fmoles/mg membrane homogenate. There is one GnRH antagonist whose affinity is greater for the pituitary than for the testis. This antagonist has the same affinity for testis homogenates as for ovarian homogenates.     

A quantitative and interspecific test for biological activity of anti-müllerian hormone: the fetal ovary aromatase assay.

Anti-Müllerian hormone (AMH), also known as Müllerian-inhibiting substance or factor, has previously been shown to sex-reverse the steroidogenic pattern of fetal mammalian ovaries through repression of aromatase biosynthesis. Study of the ontogeny of the response of cyclic AMP-stimulated aromatase activity of rat fetal ovaries to AMH has allowed us to develop a quantitative bioassay for the hormone. Linear responses as a function of the logarithm of AMH concentration were observed over ranges of 0.2-7.5 micrograms/ml for the bovine protein and 0.15-2 micrograms/ml for the human protein, with a maximal decrease in aromatase activity of 90% for both proteins. Under the same in vitro conditions, AMH treatment did not affect cyclic AMP-stimulated fetal rat testicular aromatase activity. Partially purified chick AMH also decreased rat ovarian aromatase activity, allowing us to use this test to study AMH ontogeny in chick gonads. Analysis of the species specificity of AMH repression of ovarian aromatase activity indicated that turtle and rat fetal ovaries responded to AMH of other vertebrate classes, whereas aromatase activity of chick embryo ovaries could be repressed only by the homospecific hormone.