Separation of the complexes formed between the regulatory and catalytic subunits of cyclic adenosine monophosphate-dependent protein kinase and topoisomerase I activity in preovulatory follicle-enriched immature rat ovaries.

Our previous studies have shown that the regulatory subunits of the type II form of cAMP-dependent protein kinase (RII) present in soluble extract of immature rat ovaries elute from diethylaminoethyl-cellulose as three separate peaks of activity, based on their association with the catalytic subunit (C) of this enzyme, as R2IIC2, an apparent R2IIC, and R2II. Based upon the existence of ovarian RII in three different subunit arrangements, the large amount of C subunit-free R2II in this tissue, and a previous report which indicated that RII exhibited intrinsic topoisomerase I activity, we determined whether DNA topoisomerase I activity was associated with any of these molecular complexes of the ovarian RII subunits. Cyclic AMP-binding activities in soluble extracts of preovulatory follicle-enriched immature rat ovaries were separated by diethylaminoethyl-cellulose chromatography and sucrose density gradient centrifugation. Topoisomerase I activity cosedimented with the apparent R2IIC and R2II obtained from sucrose gradients but was not detected in fractions containing R2IIC2. Upon cAMP affinity purification of the RII derived from fractions containing R2IIC2, R2IIC, and R2II, respectively, no topoisomerase I activity could be detected in any fraction. Phosphorylation of the affinity purified RIIs by the C subunit of beef heart cAMP-dependent protein kinase did not alter this result. These data indicate that none of the RII subunits in soluble extracts of preovulatory follicle-enriched ovaries exhibit intrinsic topoisomerase I activity.

Opposing effects of ethanol on pig ovarian adenylyl cyclase desensitized by human choriogonadotropin or isoproterenol.

Pig ovarian follicular membranes contain a gonadotropin-responsive adenylyl cyclase, which becomes partially desensitized (approximately 40%) upon a 40-min incubation with a saturating concentration of human (h) CG. This in vitro desensitization is time and hormone dependent and also requires the presence of micromolar concentrations of GTP. In this report we show that 10% ethanol present during the desensitization phase of the incubation increases the extent of hCG-induced desensitization of adenylyl cyclase by 2-fold. Ethanol shortened the time necessary to reach maximal hCG-induced desensitization from 20 to 10 min, but had no effect on the dose dependency for GTP. In addition, ethanol had no effect on the affinity of the LH/hCG receptor for 125I-hCG but did cause an increase in the ED50 of hCG for inducing desensitization from 0.25 to 0.75 nM. Interestingly, ethanol decreased the apparent number of LH/hCG-receptor sites by 55%, yet the control hCG-sensitive adenylyl cyclase activity was not reduced. The "hyperdesensitized" state achieved in the presence of ethanol could not be reversed by washing the membranes and incubating them in ethanol-free medium. NaF-sensitive adenylyl cyclase was also not impaired in hCG-desensitized membranes from control or ethanol-treated samples. Thus, hCG-induced desensitization was not due to a defect in the functioning of the stimulatory guanine nucleotide-binding regulatory protein (G8) or catalytic subunits, but rather was caused by an impairment of the coupling of the lutropin (LH)/hCG receptor with G8, which was exacerbated further by ethanol. In spite of the effect of ethanol on hCG-induced desensitization, this agent had an inhibitory effect on isoproterenol-induced desensitization of isoproterenol-responsive luteal adenylyl cyclase. These results indicate that membrane fluidity is important in modulating the structure and functional interaction of the LH/hCG receptor with G8 because ethanol is a well known lipid-fluidizing agent. The resistance to ethanol potentiation of desensitization of the isoproterenol-sensitive adenylyl cyclase suggests that there are differences between the LH/hCG- and beta-adrenergic receptors in factors controlling their structures and or interactions with G proteins, and that there is a fundamental difference in their mechanisms of desensitization.

Comparison of the luteinizing hormone-sensitive adenylyl cyclase of the pig ovarian follicle and corpus luteum and its susceptibility to in vitro hormone-dependent desensitization.

The hormone responsiveness of the adenylyl cyclase of pig ovarian follicles or corpora lutea was examined. Adenylyl cyclase activity was assayed in 10,000 x g membrane fractions that had been prepared with or without (control) a urea extraction. In control luteal membranes there was little stimulation (less than 2-fold) or adenylyl cyclase by saturating ovine (o) LH, hCG, or (-)isoproterenol in the absence or presence of 10 microM GTP. However, in urea-treated luteal membranes, a 2- to 3-fold stimulation of adenylyl cyclase was caused by saturating oLH or hCG, and a 4- to 5-fold stimulation by (-)isoproterenol; the marked stimulation by the gonadotropins was only observed if 10 microM GTP was added. In follicular membranes, a 3- to 4-fold stimulation of adenylyl cyclase by gonadotropins was observed regardless of whether GTP was added or the membranes had been urea extracted. Stimulation of adenylyl cyclase by (-)isoproterenol was always less than 2-fold in follicular membranes. The binding affinity for [125I]hCG was similar in control follicular and luteal membranes, but there were approximately 10-fold more [125I]hCG-binding sites in follicular compared with luteal membranes. The binding affinities and number of receptor sites were not significantly changed by urea treatment. The ED50 values for hCG or (-)isoproterenol were the same in follicular and luteal membranes and were uneffected by the addition of 10 microM GTP, but the ED50 for oLH was 3-fold lower in follicular than in luteal membranes. GTP caused a dose-dependent increase in adenylyl cyclase activity in luteal and follicular membranes, and both tissues had the same ED50. A saturating hormone concentration resulted in an approximately 2-fold decrease in the ED50 for GTP. In vitro hCG-induced desensitization of the hCG-responsive adenylyl cyclase was 31% in follicular membranes, but only 11-15% in luteal membranes. Hormone-induced desensitization was not increased in incubations of luteal homogenate or membranes plus cytosol. These results establish the existence of a LH/hCG-sensitive adenylyl cyclase in the pig corpus luteum and indicate that the G-protein and catalytic moieties of the follicular and luteal adenylyl cyclase complex are functionally the same, but some difference exists in the way the LH/hCG-receptor in the two tissues interacts with the G-protein/catalytic complex.

Guanosine triphosphate fulfills a complete and specific nucleotide requirement for luteinizing hormone-induced desensitization of pig ovarian adenylyl cyclase.

Homologous desensitization of the ovarian LH-sensitive adenylyl cyclase in cell-free systems is known to be dependent on micromolar concentrations of GTP. In this study, we sought to determine whether the nucleotide requirement of GTP for desensitization is complete and specific. LH-dependent desensitization of the adenylyl cyclase of pig ovarian follicular membranes was examined without adding adenyl-5'-yl imidodiphosphate [AMP-P(NH)P] to prevent the degradation of nucleotide triphosphates. GTP at 0.1 mM or higher [in the absence of AMP-P(NH)P] was able to support the same amount of desensitization (30-35%) that occurs with 1 mM AMP-P(NH)P and micromolar concentrations of GTP. ATP, UTP, and CTP also supported maximal desensitization, but were 5-30 times less potent than GTP. The ED50 value for GTP (14 microM) was about 100 times higher than when the reaction was performed in the presence of 1 mM AMP-P(NH)P; the ED50 values for UTP (70 microM) and CTP and ATP (400 microM) were only 2-5 times higher than when AMP-P(NH)P was included. Incubation of [alpha-32P]GTP with pig follicular membranes demonstrated that 1 mM GTP was stable, whereas 10 microM or less was rapidly degraded unless 1 mM AMP-P(NH)P was included. The specificity of GTP for supporting desensitization was also demonstrated in experiments where guanyl-5'-yl imidodiphosphate, the nonhydrolyzable GTP analog, inhibited LH-induced desensitization which otherwise would have occurred in the presence of ATP, CTP, or UTP. These studies establish the complete and specific requirement of GTP for supporting desensitization of ovarian LH-responsive adenylyl cyclase.

Homologous desensitization of ovarian luteinizing hormone/human chorionic gonadotropin-responsive adenylyl cyclase is dependent upon GTP.

Homologous desensitization of ovarian LH/hCG-responsive adenylyl cyclase in cell-free systems has been reported to be dependent upon either ATP or GTP. We investigated LH/hCG-dependent desensitization of adenylyl cyclase in rabbit or pig ovarian follicular membranes and included adenyl-5'-yl imidodiphosphate to prevent nucleotide triphosphate degradation. It was found that GTP supported LH/hCG-induced desensitization with an apparent Km of approximately 0.1 microM in rabbit or pig ovarian membranes. Other nucleotide triphosphates were 100-1000 times less potent than GTP in supporting desensitization. Several nonhydrolyzable GTP analogs and the GDP analog guanosine-5'-O-(2-thiodiphosphate) would not support hCG-induced desensitization of ovarian adenylyl cyclase. Instead, these guanine nucleotide analogs were all inhibitors of GTP-supported hormone-dependent desensitization. Cholera toxin had no effect on LH-dependent desensitization. These results establish that GTP is the preferred nucleotide for homologous desensitization of the LH/hCG-sensitive adenylyl cyclase and that the GTP-dependent mechanism differs from that typically associated with guanine nucleotide-binding proteins. The GTP-dependent mechanism of desensitization of the LH receptor distinguishes it from the ATP-dependent desensitization of the beta-adrenergic receptor.

The effect of protein kinases on desensitization of the porcine follicular membrane luteinizing hormone/chorionic gonadotropin-sensitive adenylyl cyclase.

Because the acute homologous phase of desensitization of the LH/CG-sensitive adenylyl cyclase in porcine follicles is readily demonstrated in a cell-free membrane preparation, it follows that any enzyme(s) required to achieve desensitization must be present in the membranes and must be activated upon LH/CG receptor activation. The purpose of the following studies was to determine whether modulation of endogenous membrane protein kinases, with activators or inhibitors, or addition of exogenous protein kinases affected desensitization of the LH/CG-sensitive adenylyl cyclase. The effects of these potential modulators were evaluated in both the presence and absence of ligand (hCG)-stimulated receptor activation. To this end, membranes were incubated in the presence or absence of hCG (stage 1) and then assayed for adenylyl cyclase activity in the presence or absence of hCG (stage 2). The results showed that although porcine follicular membranes rich in LH/CG-sensitive adenylyl cyclase activity also exhibited cAMP-dependent [protein kinase-A (PKA)], cGMP-dependent (PKG), lipid-dependent (PKC), Ca2+/calmodulin, and casein kinase-I and -II activities, only full hCG-stimulated adenylyl cyclase activity (measured with BSA in stage 1 and hCG in stage 2) was reduced upon addition of exogenous PKC (to the stage 1 incubation). hCG-dependent desensitization of cAMP synthesis (measured with hCG in stages 1 and 2) was unaffected by activators or inhibitors of endogenous PKA, PKC, or PKG, by an inhibitor of casein kinases and kinases in the beta-adrenergic receptor kinase family, or by the addition of exogenous active PKA, PKC, or rhodopsin kinase to the stage 1 incubation. These results suggest that the acute homologous phase of hCG-dependent desensitization of adenylyl cyclase activity in follicular membranes is not regulated by PKA, PKC, PKG, or messenger-independent heparin-sensitive protein kinases.

Investigation of putative androgen-like activity of alpha 2u-globulin in castrated and estrogen-treated male rats.

alpha 2u-Globulin, a male sex-dependent protein of the rat, was tested for its ability to produce androgenic or anabolic responses. Castrated rats were injected with alpha 2u-globulin isolated from urine or kidney and the effects on urinary nitrogen excretion, body weight, and weight of ventral prostate, seminal vesicles, and levator ani muscle were studied. alpha 2u-Globulin had no effect but testosterone propionate gave the expected stimulation of nitrogen retention, body growth, and growth of androgen-responsive tissues. alpha 2u-Globulin injected into estrogen-treated male rats (animals with suppressed serum levels of alpha 2u-globulin, gonadotropins, and testosterone) was also found to be without effect on body weight and on the weight of ventral prostate, seminal vesicles, levator ani muscle, and testes. We conclude that the alpha 2u-globulin molecule, by itself, is incapable of producing, directly or indirectly, androgenlike responses.

Modulation of soluble ovarian adenosine 3',5'-monophosphate-dependent protein kinase activity during prepubertal development in the rat. II. Evaluation of the catalytic subunit inhibitor activity.

In a previous report on the ontogeny of the ovarian adenosine 3',5'-monophosphate (cAMP)-dependent protein kinase activity during prepubertal development of the rat, we concluded that the 4-fold decline in cAMP-dependent protein kinase activity observed in ovaries of 21- to 23-day-old rats was due to the presence of a heat-labile inhibitor in the ovarian extracts (Hunzicker-Dunn et al., 1984). We developed an assay for this ovarian kinase inhibitor activity that was based on the observation that ovarian cytosol added to an exogenous catalytic subunit of cAMP-dependent protein kinase caused a time-dependent and ovarian cytosol protein concentration-dependent inhibition of exogenous catalytic subunit phosphotransferase activity. The present studies were conducted to evaluate the basis for this catalytic subunit inhibitor present in soluble rat ovarian extracts of prepubertal-aged rats. This inhibitor activity was absent from cytosol extracts of rat corpora lutea, rat liver, rabbit follicles, and rabbit corpora lutea. Inhibitor activity present in rat ovarian cytosol was not attributable to insufficient levels of the phosphorylation substrate Kemptide. Inhibitor activity was also not related to the presence of the large amount of catalytic subunit-free regulatory subunit of the cAMP-dependent protein kinase present in ovarian extracts of late juvenile-aged rats. Inhibitor activity, however, did correlate with an endogenous adenosine triphosphatase (ATPase) activity that reduced assay ATP concentrations below levels needed to accurately measure phosphotransferase activity, despite the presence of sodium fluoride (an ATPase inhibitor) and ATP concentrations 5- to 15-fold greater than the Km of the kinase for ATP.(ABSTRACT TRUNCATED AT 250 WORDS)

Reversal of the desensitized state of pig ovarian follicular human choriogonadotropin-sensitive adenylylcyclase by guanosine 5'-O-(2-thiodiphosphate).

We investigated the stability of the desensitized state of the human choriogonadotropin (hCG)-sensitive adenylylcyclase of the pig ovarian follicle. A 20,000 x g membrane preparation of pig follicular membranes was incubated under conditions which resulted in the hormone-induced desensitization of the hCG-responsive adenylylcyclase. The desensitized state was maintained upon subsequent incubation of the membranes with GTP, GDP, GMP, ATP, ADP, AMP, CTP, UTP, adenyl-5'-yl imidodiphosphate (AMP-P(NH)P), and adenyl (beta, gamma-methylene)-diphosphonate (AMP-P(CH2)P); however, the desensitized state was reverted to a fully active state upon incubation with guanosine 5'-O-(2-thiodiphosphate) (GDP beta S) and guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S). The reversal effect of GDP beta S on hCG-responsive adenylylcyclase activity was time- and temperature-dependent, and showed a selectivity for GDP beta S over adenosine 5'-O-(2-thiodiphosphate) (ADP beta S) (half-maximal effective dose of 12 microM versus 260 microM, respectively). GDP beta S had no effect on the binding affinity or apparent number of luteinizing hormone (LH)/CG receptors or on the dissociation rate of 125I-hCG from the receptor. GDP beta S promoted an hCG- and time-dependent release of guanine nucleotides from the membranes. A model is proposed which accounts for the unique characteristics of LH/CG-sensitive adenylylcyclase desensitization and subsequent reactivation by GDP beta S.

Protein phosphorylation in the corpus luteum.

Soluble luteal extracts were incubated with putative second messengers or regulators of protein kinases in the presence of [gamma-32P]ATP, and proteins were separated by SDS-PAGE. There was a novel phosphorylation of a protein of Mr 80,000 which was stimulated by phospholipid and 1,2-diacylglycerol but, unlike classical C-kinase phosphorylating activity, was increased by EGTA and reduced by Ca2+. This phospholipid/diolein-stimulated phosphorylation of a Mr 80,000 protein was detectable in rat and pig luteal extracts and was enhanced 2-fold in rabbit CL by administration of oestradiol-17 beta in vivo. These results suggest that the luteotrophic functions of oestradiol in rabbit CL are mediated in part by regulating either the kinase and/or the Mr 80,000 substrate.