Impact of subdomain D1 of the short form S1b of the human prolactin receptor on its inhibitory action on the function of the long form of the receptor induced by prolactin.

BACKGROUND: Long-form (LF) homodimers of the human prolactin receptor (PRLR) mediate prolactin's diverse actions. Short form S1b inhibits the LF function through heterodimerization. Reduced S1b/LF-ratio in breast cancer could contribute to tumor development/progression. Current work defines the structural and functional relevance of the D1 domain of S1b on its inhibitory function on prolactin-induced LF function. METHODS: Studies were conducted using mutagenesis, promoter/signaling analyses, bioluminescence resonance energy transfer (BRET) and molecular modeling approaches. RESULTS: Mutation of E69 in D1 S1b or adjacent residues at the receptor surface near to the binding pocket (S) causes loss of its inhibitory effect while mutations away from this region (A) or in the D2 domain display inhibitory action as the wild-type. All S1b mutants preserved prolactin-induced Jak2 activation. BRET reveals an increased affinity in D1 mutated S1b (S) homodimers in transfected cells stably expressing LF. In contrast, affinity in S1b homodimers with either D1 (A) or D2 mutations remained unchanged. This favors LF mediated signaling induced by prolactin. Molecular dynamics simulations show that mutations (S) elicit major conformational changes that propagate downward to the D1/D2 interface and change their relative orientation in the dimers. CONCLUSIONS: These findings demonstrate the essential role of D1 on the S1b structure and its inhibitory action on prolactin-induced LF-mediated function. GENERAL SIGNIFICANCE: Major changes in receptor conformation and dimerization affinity are triggered by single mutations in critical regions of D1. Our structure-function/simulation studies provide a basis for modeling and design of small molecules to enhance inhibition of LF activation for potential use in breast cancer treatment.

Estrogen-dependent Leydig cell protein recognized by monoclonal antibody to MCF-7 cell line.

A protein (27,000 molecular weight) was previously found in rat Leydig cells after treatment with estradiol (E2) and human chorionic gonadotropin (hCG) in vitro. The effect of hCG occurred through increased E2 production. This hormone-regulated rat testicular protein was compared to an estrogen-regulated protein of similar physical characteristics isolated from a human mammary cancer cell line (MCF-7) and present in normal human estrogen target organs. The Leydig cells from rat and human tissue showed specific immunofluorescence and immunoperoxidase staining in the cytoplasm upon incubation with a monoclonal antibody (C11) to the estrogen-regulated protein from MCF-7 cells. Leydig cells after exposure to E2 or hCG showed the highest fluorescence intensity; this intensity was reduced by treatment with Tamoxifen. No reaction was associated with other testicular cells. The estrogen-regulated protein from human cell lines is therefore immunologically similar to that from the rat Leydig cell. The monoclonal antibody should be useful for further characterization of the Leydig cell protein.

Hormonal regulation of gonadotropin receptors and steroidogenesis in cultured fetal rat tests.

Gonadotropic activation of the adult rat testis in vitro and in vivo is followed by down-regulation of luteinizing hormone receptors and decreased androgen responses to subsequent hormonal stimulation. In contrast, treatment of cultured fetal testes with gonadotropins and dibutyryl adenosine 3',5'-monophosphate enhanced steroidogenic responsiveness and did not cause the luteinizing hormone-receptor loss and desensitization that is characteristic of the adult gonad. The analysis of gonadotropin receptors and action in cultured fetal testis cells facilitates developmental studies of gonadal function, and has revealed significant differences in the responses of fetal and adult Leydig cells to gonadotropic regulation.

Estradiol modulates the pulsatile secretion of biologically active luteinizing hormone in man.

We investigated the effects of estradiol on bioactive luteinizing hormone (LH) release in normal men using two complementary strategies: (i) steady state intravenous infusions of estradiol at its endogenous production rate, and (ii) oral administration of the antiestrogen, tamoxifen HCl. Immunoreactive and biologically active LH were monitored by radioimmunoassay and the rat interstitial cell testosterone bioassay, respectively. Estradiol infusions significantly suppressed mean plasma bioactive LH concentrations and decreased the bio/immuno LH ratio. Conversely, antiestrogen treatment enhanced spontaneous bioactive LH pulse frequency, increased bioactive LH pulse amplitude, and augmented plasma intrapulse and interpulse bio/immuno LH ratios. Low-dose pulsed injections of exogenous gonadotropin-releasing hormone (GnRH) also increased plasma bio/immuno LH ratios. However, tamoxifen attenuated the ability of exogenous GnRH to further enhance the bio/immuno LH ratio, which suggests that endogenous LH release was already maximally enriched in LH bioactivity during antiestrogen administration. We conclude that estradiol modulates the pulsatile secretion of LH molecules enriched in biological activity in man.

Attenuated release of biologically active luteinizing hormone in healthy aging men.

To examine the biological quality and physiologically pulsatile mode of endogenous luteinizing hormone release in active, healthy aging men, we used the rat interstitial-cell testosterone in vitro bioassay to probe LH bioactivity in response to (a) endogenous gonadotropin-releasing hormone (GnRH) action (basal pulsatile bioactive LH secretion); (b) exogenous GnRH stimulation (10 micrograms IV pulses); and (c) inhibition of endogenous estrogen negative feedback (treatment with a nonsteroidal antiestrogen, tamoxifen). Basally, some healthy older men exhibited evidence of neuroendocrine dysfunction, reflected by irregular bursts of bioactive LH release followed by transiently low plasma bio:immuno (B:I) LH ratios. However, mean basal plasma bioactive LH concentrations, B:I ratios, and spontaneous LH pulse properties (peak frequency, amplitude, duration, and enhanced B:I ratios within LH peaks) were not altered in older men. On the other hand, augmentation of bioactive LH secretion and enhancement of plasma B:I ratios by pulsed injections of exogenous GnRH were either significantly reduced or absent in older men. In addition, although tamoxifen increased bioactive LH pulse frequency in both age groups and facilitated exogenous GnRH action in some subjects, older men increased their 12-h mean bioactive LH concentrations, B:I ratios, and bioactive LH peak amplitudes to a significantly lesser degree than young men. In summary, young and older healthy men exhibit similar mean basal plasma bioactive LH concentrations and spontaneous LH pulse properties. However, pituitary bioactive LH reserve is markedly attenuated in older men challenged with either exogenous GnRH or antiestrogen. Accordingly, we conclude that healthy aging men manifest an impaired secretory reserve for biologically active LH release.

Metabolic clearance of biologically active luteinizing hormone in man.

The plasma metabolic clearance of biologically active luteinizing hormone (bioactive LH) was studied using the rat interstitial cell testosterone (RICT) bioassay in six hypogonadotropic men after single bolus injection of highly purified human LH and during continuous steady-state infusions of three graded doses of LH. The LH bolus disappearance curves provided estimates of metabolic clearance rates (MCR) of 24.1 +/- 4.7 (+/- SD) ml/min for bioactive LH vs. 56.2 +/- 12 ml/min for immunoactive LH in the same men (P = 0.03). A lower MCR of bioactive LH compared with immunoactive LH was also observed during continuous infusions of physiological doses of LH; for example, the mean steady-state MCRs for bioactive and immunoactive LH were, respectively, 26.1 +/- 3.1 and 34.2 +/- 3.2 ml/min (P = 0.02). Moreover, the stepped-dose infusion regimens permitted us to demonstrate that increasing doses of pure human LH resulted in progressive and parallel decreases in the apparent MCRs of both bioactive and immunoactive LH. Based on the respective steady-state MCRs calculated at physiological plasma concentrations of immunoactive and bioactive LH, we estimate a mean endogenous production rate for bioactive hormone of 1,937 IU/24 h, and for immunoactive LH of 589 IU/24 h in normal men. These results indicate that previous estimates of LH production rates from immunoassay data alone markedly underestimate the quantity of biologically active hormone secreted in man.

Regulation of primate testicular luteinizing hormone receptors and steroidogenesis.

The testicular luteinizing hormone (LH) receptors of the rhesus monkey and human have many features in common, including high equilibrium association constant, marked species specificity, and relatively low binding capacity. We have, therefore, used rhesus monkeys as models for human LH-receptor regulation in vivo during gonadotropin treatment. In four adult male monkeys, treated with 10,000 IU human chorionic gonadotropin (hCG), serum and testicular steroidogenic responses were monitored at 24-h intervals during the following 4 d, and LH-receptor concentrations were measured by (125)I-hCG binding to 15,000-g particles prepared from testis biopsy specimens. In treated animals, serum hCG was maximal on day 1 at 322+/-16 ng/ml and declined to 24.4+/-2.3 ng/ml by day 4. Serum testosterone was increased threefold during the subsequent 4 d (from 6.5+/-2.0 to 18.6+/-4.4 ng/ml) but serum progesterone remained unchanged. In contrast, serum 17alpha-hydroxyprogesterone increased 12-fold to 5.5+/-0.5 ng/ml at day 1 and was increased fourfold during the subsequent 3 d. The LH-receptor binding capacity of the primate testis was reduced by 18.3+/-6.0% on day 1, 51.7+/-7.4% on day 2, and 45.3+/-2.4% on day 4. Occupancy of the LH receptors by endogenously bound hCG was significant on day 1 but was negligible by day 4. These data demonstrate that gonadotropin-induced LH-receptor depletion occurs in the rhesus testis and indicate that primate gonadotropin receptors are susceptible to the regulatory processes recently described in the rat. In addition, the simultaneous and disproportionate accumulation of 17alpha-hydroxyprogesterone indicates that 17,20-desmolase was rate-limiting under these conditions in the primate testis Leydig cell.

Estrogen dependence of a gonadotropin-induced steroidogenic lesion in rat testicular Leydig cells.

Leydig cells isolated from the testes of rats treated with intravenous exogenous gonadotropin (hCG) or subcutaneous gonadotropin-releasing hormone (GnRH) show markedly decreased luteinizing hormone (LH) receptors and a partial block in testicular 17,20 desmolase activity. In contrast, Leydig cells from animals with equivalent degrees of LH receptor loss induced by subcutaneous hCG treatment show no change in 17,20 desmolase activity. These findings indicated that the acuteness of gonadotrophic stimulation, rather than the extent of LH receptor loss, was responsible for the steroidogenic lesion. A role of estradiol in the enzymatic block produced in vivo by acute elevation of circulating gonadotropin (intravenous hCG or GnRH-stimulated endogenous LH) was suggested by rapid elevations of testicular 17beta-estradiol within 30 min after intravenous hCG, whereas more gradual increases in estradiol occurred 4-8 h after subcutaneous hCG. The inhibitory effect of endogenous estrogen on testicular steroidogenesis was confirmed by the ability of an estrogen antagonist (Tamoxifen) to prevent the reduction of testosterone responses caused by intravenous hCG and subcutaneous GnRH. In addition, Tamoxifen significantly increased the number of LH receptors in Leydig cells from both control and gonadotropin-desensitized animals. These findings indicate that the acute elevations of intratesticular estrogen produced by treatment with hCG or GnRH are responsible for the steroidogenic lesion seen in gonadotropin-desensitized Leydig cells. These results also suggest that locally produced estrogens contribute to the regulation of testicular LH receptors and 17,20 desmolase activity.

Sequence of the 3'-noncoding region of the luteinizing hormone receptor gene and identification of two polyadenylation domains that generate the major mRNA forms.

We present 6.2 kb of the 3'-noncoding region sequence of the rat luteinizing hormone receptor (LHR) gene and identification of two functional polyadenylation (pA) domains, H1 (nt 2368-2491) and H2 (nt 5579-5768) responsible for 3'-end processing of the 2.6/2.3 kb and the 5.8 kb LHR mRNA, respectively. Two identical copies of pA elements AAUAUA in H1 and of AAUAAA in H2 account for micro-heterogeneous poly(A) addition at each of the two pA regions. Both LH holoreceptor and major splice variant form B (lacking the first 266 bp of exon 11) are identified in H1-terminated (2.6 kb and 2.3 kb) and H2-terminated (5.8 kb) mRNA transcripts. A rodent repetitive DNA LINE R domain 3' of H1 within the major 5.8 kb species and a B2 element downstream of H2 were identified. Alignment of the 3'-noncoding region of LHR with TSH, FSH and beta 2-adrenergic receptors indicate that H1 pA signal is unique to the LHR and may represent an insertion domain.

Prolactin receptor gene diversity: structure and regulation.

The diverse functionality of prolactin and the wide expression of the prolactin receptor suggest a complex system regulated by this polypeptide hormone. Different hormone and receptor forms, as well as differential signal transduction pathways, contribute to the functional diversity of prolactin's actions. The heterogeneity of rat prolactin receptor gene transcripts in their 5'-untranslated region has led to the recognition of multiple and tissue-specific utilization of prolactin receptor gene promoters in gonadal and non-gonadal tissues. These findings have provided insights into the molecular bases for the diversity of prolactin's actions. It is now clear that cellular responsiveness to prolactin can be regulated through differential promoter control of the expression of the surface receptors for prolactin in different target tissues.

EAR2 and EAR3/COUP-TFI regulate transcription of the rat LH receptor.

Our previous studies demonstrated regulation of the human LH receptor (hLHR) promoter by nuclear orphan receptors EAR2, EAR3/COUP-TFI (repression), and TR4 (activation) through a direct-repeat motif (hDR). The current studies investigated the differential binding of orphan receptors to rat (rLHR) and hLHR promoters, and their modulation of rLHR gene transcription in rat granulosa cells. The rLHR DR with one nucleotide difference from hDR at its core sequence mediated inhibition of the rLHR transcription, to which EAR2 and EAR3/COUP-TFI but not TR4 bound. The A/C mismatch was responsible for the lack of TR4 binding and function, but had no effect on EAR2 and EAR3/COUP-TFI. EAR2 and EAR3/COUP-TF bound to the rLHR DR with lower affinity than to the hDR, and exhibited lesser inhibitory capacity. This difference resulted from the lack of a guanine in the rDR, which is present 3' next to the hDR core. These studies have identified sequence-specific requirements for the binding of EAR2, EAR3/COUP-TFI, and TR4 to the DRs that explain their differential regulation of rat and human LHR genes. In addition, hCG treatment significantly reduced the inhibition of rLHR gene in granulosa cells and also decreased EAR2 and EAR3/COUP-TFI protein levels. These results indicate that hormonally regulated expression of EAR2 and EAR3/COUP-TFI contributes to gonadotropin-induced derepression of LHR promoter activity in granulosa cells.

Dissociation of hydroxylase and lyase activities by site-directed mutagenesis of the rat P45017 alpha.

Site-directed mutagenesis of the arginine-rich region within the putative active site of the rat testicular P45017 alpha (17 alpha-hydroxylase cytochrome P-450, the product of P450XVII gene) was performed to identify specific amino acids that contribute to either the hydroxylase or lyase activities catalyzed by this enzyme. The conversion of Arg346 to alanine differentially abolished lyase activity without affecting hydroxylase activity, resulting in an accumulation of the 17 alpha-hydroxylated intermediate, and partial lyase activity was recovered by conversion of this mutant to Lys346. Similar results were obtained with the conversion of Arg357 to alanine, although this mutant also diminished hydroxylase activity, and full lyase and hydroxylase activities were recovered with a lysine in this position. Major reductions in hydroxylase activity were apparent with the conversion of Arg363 to Ala, and this inhibition was reversed by a lysine at position 363. In contrast, differential effects were not observed with the mutants Arg361 Ala, Arg361 Lys, or Tyr334Phe. Both mutations at the Arg361 position resulted in a total loss of hydroxylase and lyase activities, and mutation at the Tyr334 position had no effect on either activity. The identification of specific amino acids that are essential for either the hydroxylase or lyase reaction indicates that the steroid substrate-protein interaction changes during the course of the two consecutive reactions and reveals the potential for separation of the two activities by chemical and biological modulators within the active site region of the P45017 alpha.

Modulation of prolactin-stimulated Nb2 lymphoma cell mitogenesis by cholera toxin and pertussis toxin.

We have investigated whether cholera toxin (CT)- or pertussis toxin (IAP)-sensitive G proteins are involved in ovine (o) PRL-stimulated mitogenesis in the lactogen-dependent rat Nb2 node lymphoma cell line. Addition of IAP to medium caused a biphasic effect on oPRL-stimulated cell number. Low doses (10(-3) ng/ml) enhanced (mean +/- SEM, 15 +/- 3%) whereas higher doses (greater than or equal to 10 ng/ml) inhibited (24 +/- 3%) mitogenesis stimulated by a submaximal dose of oPRL (0.1 ng/ml) compared to control values. The cAMP analog 8-bromo-cAMP also had a biphasic effect on cell division stimulated by submaximal doses of PRL. Low doses (10(-5) M) enhanced whereas higher doses (10(-3) M) inhibited Nb2 cell growth in response to PRL. Incubation with CT only inhibited oPRL-stimulated mitogenesis in a dose-dependent manner. Maximal inhibition (63 +/- 7%) occurred at a concentration of 10 ng/ml or more. Phorbol myristate acetate (PMA) enhanced mitogenesis stimulated by PRL alone and in the presence of either stimulatory or inhibitory doses of IAP, but PMA did not block IAP inhibition. In contrast, PMA had no effect on cells incubated with CT; the inhibition of PRL-stimulated cell division by CT remained unchanged. Lactogenic receptor-binding sites per cell and affinity were not significantly affected by PMA, IAP, or CT, suggesting a postreceptor mechanism of action. In summary, these data demonstrate that cAMP modifies PRL-stimulated Nb2 cell mitogenesis. The differences between IAP and CT (i.e. biphasic effect, degree of inhibition, and differential effect of PMA) suggest that these agents could also modulate PRL actions in the Nb2 cell through different mechanisms, including a cAMP-independent pathway.

Changes in ribonucleic acid polymerase activities in gonadotropin-treated Leydig cells: an estradiol-mediated process.

The effects of desensitizing doses of hCG on the activities of Leydig cell DNA-dependent RNA polymerases I, II, and III were investigated after optimization of the enzyme assay. Individual activities were obtained by taking advantage of their different sensitivities to alpha-amanitin. RNA polymerase III was a minor component of the alpha-amanitin-resistant activity at 0.10 micrograms/ml, and was therefore measured together with RNA polymerase I. In adult rats treated with 10 micrograms hCG, sc, RNA polymerases II and I plus III activities of Leydig cells rose within 45 min to 180 +/- 6% and 162 +/- 2% of the control value, and then decreased to control values at 60 min. The initial stimulation of polymerase activities was coincident with maximal increases in testosterone and estradiol levels in plasma. A second and more sustained increase in polymerases II and I plus III activities occurred between 12 and 24 h (212 +/- 12% and 180 +/- 10% of the control value) and was maintained until 48 h after hCG injection. The hCG-induced rise in polymerase activities was due to activation of the enzymes, since chromatin template capacity was unaltered. In animals treated with the antiestrogen tamoxifen, stimulation of RNA polymerase activity by hCG was completely inhibited. Also, hCG did not stimulate polymerase activity in animals treated with aminoglutethimide, which blocks steroid synthesis from cholesterol, or in those treated with adrostatriendione, which inhibits aromatization of testosterone leading to estradiol. Increases in RNA polymerase activities were also achieved by the administration of lower doses of hCG (0.1 and 1 micrograms) and the administration of estradiol (2 micrograms), resembling the pattern seen with 10 micrograms hCG. These studies have indicated that the hCG-induced RNA polymerase activation in the Leydig cell is mediated through nuclear actions of estradiol, since stimulation of the enzymes was prevented by tamoxifen and inhibition of steroid biosynthesis, and was induced by estradiol administration.

Rat chorionic gonadotropin: augmentation of bioactivity in the absence of the pituitary.

Significant gonadotropin-like bioactivity, as measured by rat interstitial cell testosterone assay, is present in the serum and placenta of pregnant rats. The patterns of activity for this putative rat chorionic gonadotropin (rCG) in serum and placenta are distinct from those observed for rat placental lactogen (rPL) radioreceptor activity. Placental and serum rCG (and rPL) activities are substantially increased, as are placental weights, following removal of the maternal pituitary on day 12 of pregnancy. These findings strengthen the evidence for the existence of a rat CG and suggest a novel role for the maternal pituitary in regulating placental function during normal gestation.

Functional and morphological studies on isolated Leydig cells: purification by centrifugal elutriation and metrizamide fractionation.

Rat interstitial cells were fractionated by centrifugal elutriation to facilitate the purification of Leydig cells for analysis of mechanisms of gonadotropin action in vitro. By this procedure, 10(9) collagenase-dispersed interstitial cells from adult rat testes were separated into 12 fractions in about 1 h. Fractions 1-7 (sedimentation velocities, 1.9-12 mm/h.g) comprised 80-85% of the total cells applied, including erythrocytes, lymphocytes, germinal cells, macrophages, endothelial cells, damaged Leydig cells, and contained less than 4% intact Leydig cells. Fractions 8-12 (sedimentation velocities, greater than 12 mm/h.g) comprised 15-20% of the original cells and contained 90-95% intact Leydig cells. Despite their different sedimentation velocities, the Leydig cell-rich fractions were similar in their LH receptor content (mean +/- SD, 36,115 +/- 4,815 sites/cell) and showed similar 5-fold increases above the original cell preparation in testosterone and cAMP responses to hCG. The pooled Leydig cell-rich fractions (8-12) were further resolved on 16-24% Metrizamide gradients into 5 bands. Bands II-V (density range, 1.075-1.110 g/ml) contained pure Leydig cells, and band I (1.048 g/ml) contained pachytene spermatocytes, the contaminating cell type present in the Leydig cell-rich fractions obtained by elutriation. Each of the 4 Leydig cell-rich bands showed similar morphology and functional activity. Essentially similar results were observed using 14-32% Metrizamide gradients. Leydig cells desensitized in vivo by hCG treatment and isolated by elutriation were also resolved by Metrizamide gradients into 4 bands, but showed a redistribution in the gradient, due to the shift of about 50% of the cells originally present in the heaviest bands to lower density fractions. However, in spite of their changes in density, the Leydig cell bands still showed similar degrees of receptor down-regulation and impairment of the steroid responses to hCG in vitro. This study has demonstrated that centrifugal elutriation is a rapid and effective method for obtaining large quantities of purified (greater than 90%) and active Leydig cells. Further resolution of the Leydig cell-rich fractions in Metrizamide gradients has allowed complete Leydig cell purification, which is not achieved by density gradient centrifugation alone. Since less responsive or inactive Leydig cells displayed various degrees of structural damage, such cells should not be considered as a population of physiological significance.(ABSTRACT TRUNCATED AT 400 WORDS)

Lactogen receptors in rat Leydig cells: analysis of their structure with bifunctional cross-linking reagents.

[125I]Iodohuman GH was found to bind to receptors with specificity for lactogenic hormones in a Triton X-100 extract from Leydig cell membranes displaying an affinity constant of 3.8 X 10(9) M-1 and a binding capacity of 167 fmol/mg protein. Cross-linking of solubilized [125I]iodohuman GH-receptor complexes with disuccinimidyl suberate followed by analysis by sodium dodecyl sulfate-gel electrophoresis in the presence of beta-mercaptoethanol and autoradiography resulted in the appearance of bands with apparent mol wt of 113,000, 103,000, 59,000, and 53,000. The appearance of these bands was prevented by incubation in the presence of lactogenic hormones. By using a two-dimensional electrophoresis technique (first dimension under nonreducing conditions; second dimension under reducing conditions), it was demonstrated that a fraction of the mol wt 59,000 species can be released from the mol wt 103,000 species upon cleavage of disulfide bonds. These results suggest the existence of lactogen receptor species with approximate mol wt of 91,000, 81,000, 37,000, and 31,000 in Triton X-100 extracts from Leydig cell membranes if the contribution of the free hormone (mol wt, 22,000) is subtracted. A fraction of the mol wt 37,000 subunits appears to be contained within the 81,000 species linked through disulfide bonds.

Water-soluble gonadotropin receptors of the rat ovary.

Soluble receptors for LH/hCG were characterized in aqueous and low ionic strength buffer extracts of the luteinized rat ovary. These receptors are proteins with high affinity (KA = 10(10) M-1) and specificity for hCG- and LH-like gonadotropins. The water-soluble receptors are stable and retained 60-80% of their binding capacity after lyophilization. Resolution of the water-soluble receptors by gel electrophoresis demonstrated five binding species, with molecular weights of 165,000, 81,000, 48,000, 24,000 and 12,000. These findings have demonstrated that a proportion (5-10%) of ovarian LH/hCG membrane receptors can be rendered water soluble with preservation of their binding properties and that the specific binding site for LH/hCG is present in proteins much smaller than the predominant 6.5S (mol wt, 194,000) form extracted by nonionic detergents. The stability of these small LH-binding sites is of value for further purification and analysis of structural determinants for gonadotropin binding.

Serotonergic inhibition of rat Leydig cell function by propranolol.

The beta-adrenergic antagonist propranolol binds to serotonin (5HT) receptors (5HT1B > 5HT1A > 5HT2) in brain membranes. We have recently demonstrated that 5HT acts through 5HT2 receptors in rat Leydig cells to release CRF, which, in turn, inhibits hCG-stimulated cAMP production and steroidogenesis. These observations prompted us to study the effects of propranolol on CRF secretion and cAMP and testosterone production in cultured rat Leydig cells. Treatment with (-)propranolol increased CRF release and inhibited basal and hCG-stimulated cAMP and steroidogenesis, with effects evident at 0.1 microM, an IC50 of 6 microM, and reduction of stimulated levels to near basal at 100 microM. These effects of the drug were prevented by pretreatment of cultures with the 5HT2 receptor antagonist ketanserin or a CRF antagonist or antiserum. Furthermore, propranolol increased the level of 5HT in the incubation medium of cultured Leydig cells. The (+)isomer of propranolol had minor effects on these parameters. Increasing concentrations of (-)propranolol displaced the binding of [125I] +/- 1-[2,5-dimethoxy-4-iodophyryl]-2-amino propane hydrochloride (DOI), a selective 5HT2 ligand, to Leydig cell membranes (IC50, 0.2 microM), and (+)propanolol showed weaker potency. The inhibitory actions of propranolol were exerted through its blockade of the Leydig cell 5HT2 low affinity receptor, which has functional properties of an autoreceptor, with consequent increases in 5HT and stimulation of CRF release through 5HT action at the high affinity site. The serotonergic actions of propranolol were prevented by DOI, an inhibitor of 5HT actions at the high affinity site. In addition, the propranolol-induced blockade of the low affinity site further increased the cAMP and steroidogenic responses to gonadotropin over those observed with DOI treatment alone. These studies demonstrate that propranolol acts as an antagonist at the Leydig cell low affinity 5HT2 receptor and stimulates CRF release via a serotonergic mechanism, with consequent inhibition of cAMP generation and steroidogenesis. This serotonergic action of the drug could contribute to the impairment of sexual function reported during propranolol treatment in man.

Regulation of corticotropin-releasing factor secretion from Leydig cells by serotonin.

CRF is produced in the Leydig cells and acts as a negative autocrine regulator of Leydig cell function. To clarify the hormonal control of CRF secretion by Leydig cells, we evaluated the participation of serotonin (5HT) and serotonin agonists in the release of CRF from Leydig cells and their effects on hCG-induced cAMP generation and steroidogenesis. Serotonin stimulated CRF secretion up to 4-fold above basal levels and inhibited basal and hCG-stimulated cAMP generation and testosterone production (ID50, 1 nM). The inhibitory action of 5HT was prevented by a CRF antibody and the alpha-helical CRF-(9-41) antagonist. The selective 5HT2 receptor agonist (+-)1-[2,5-dimethoxy-4-iodophyryl]2-amino propane hydrochloride (DOI) also stimulated CRF secretion and inhibited hCG-stimulated cAMP generation and testosterone production to control levels (ID50, 7 microM). Serotonergic 5HT1A, 5HT1B/1C, 5HT1D, and 5HT3/5HT2 agonists were less effective inhibitors of hCG-stimulated cAMP and testosterone production, while agonists for the 5HT3 receptor had no effect. [125I]DOI binding studies in Leydig cells demonstrated two sets of receptors with Kd values in the nanomolar and micromolar range, with low and high capacities, respectively. The low affinity site differed from that of brain receptors (Kd, 4.2 nM) and displayed higher binding capacity (50-fold). The selective 5HT2 receptor antagonist ketanserin prevented CRF stimulation and blocked the inhibitory actions of 5HT and DOI, while the alpha 1-adrenergic antagonist prazosin had no effect. Also, treatment of cells with ketanserin increased sensitivity to hCG and raised maximal cAMP and testosterone production. 5HT was a more effective stimulus than hCG in stimulating CRF secretion, and gonadotropin-induced CRF release was inhibited by ketanserin. Inhibitory effects of exogenous CRF were demonstrable after blockade of 5HT action by ketanserin. The inhibitory actions of 5HT were unaffected by pertussis and cholera toxins and were reversed by the addition of 8-bromo-cAMP. These results demonstrate that 5HT acts on 5HT2 receptors in Leydig cells that are distinct from those in the brain to stimulate CRF secretion through a pertussis toxin-insensitive G-protein. This action of 5HT is predominantly mediated by the low affinity 5HT2-binding site and requires full occupancy for maximal CRF stimulation, indicating the absence of spare receptors. 5HT-stimulated CRF inhibits basal and hCG-induced cAMP generation and steroidogenesis. Furthermore, 5HT mediates the stimulatory action of LH/hCG on CRF secretion from Leydig cells and, thus, participates in a negative autoregulatory loop to limit the testosterone response to the gonadotropic stimulus.