Melatonin receptor activation regulates GnRH gene expression and secretion in GT1-7 GnRH neurons. Signal transduction mechanisms.

Melatonin plays a significant role in the control of the hypothalamic-pituitary-gonadal axis. Using the GT1-7 cell line, an in vitro model of GnRH-secreting neurons of the hypothalamus, we examined the potential signal transduction pathways activated by melatonin directly at the level of the GT1-7 neuron. We found that melatonin inhibits forskolin-stimulated adenosine 3'-, 5'-cyclic monophosphate accumulation in GT1-7 cells through an inhibitory G protein. Melatonin induced protein kinase C activity by 1.65-fold over basal levels, increased the phosphorylation of extracellular signal-regulated kinase 1 and 2 proteins, and activated c-fos and junB mRNA expression in GT1-7 cells. Using the protein kinase A inhibitor H-89, the protein kinase C inhibitor bisindolylmaleimide, and the mitogen-activated protein kinase kinase inhibitor PD98059, we found that the melatonin-mediated cyclical regulation of GnRH mRNA expression may involve the protein kinase C and the extracellular signal-regulated kinase 1 and 2 pathways, but not the protein kinase A pathway. We found that melatonin suppresses GnRH secretion by approximately 45% in the GT1-7 neurons. However, in the presence of the inhibitors H-89, bisindolylmaleimide, and PD98059 melatonin was unable to suppress GnRH secretion. These results provide insights into the potential signal transduction mechanisms involved in the control of GnRH gene expression and secretion by melatonin.     

Regulation of cyclic adenosine 3',5'- monophosphate signaling and pulsatile neurosecretion by Gi-coupled plasma membrane estrogen receptors in immortalized gonadotrophin-releasing hormone neurons

Immortalized GnRH neurons (GT1-7) express receptors for estrogen [estrogen receptor-alpha and-13(ERa and ERI3)] and progesterone (progesterone receptor A) and exhibit positive immunostaining for both intracellular and plasma membrane ERs. Exposure of GT1-7 cells to picomolar estradiol concentrations for 5-60 min caused rapid, sustained,and dose-dependent inhibition of cAMP production. In contrast, treatment with nanomolar estradiol concentrations for 60 min increased cAMP production. The inhibitory and stimulatory actions of estradiol on cAMP formation were abolished by the ER antagonist, ICI 182,780. The estradiol-induced inhibition of cAMP production was prevented by treatment with pertussis toxin, consistent with coupling of the plasma membrane ER to an inhibitory G protein. Coimmunoprecipitation studies demonstrated an estradiol-regulated stimulatory interaction between ERa and G,3 that was prevented by the ER antagonist, ICI 182,780. Exposure of perifused GT1-7 cells and hypothalamic neurons to picomolar estradiol levels increased the GnRH peak interval, shortened peak duration, and increased peak amplitude. These findings indicate that occupancy of the plasma membrane-associated ERs expressed in GT1-7 neurons by physio-logical estradiol levels causes activation of a G, protein and modulates cAMP signaling and neuropeptide secretion.     

Lowering cyclic adenosine-3',5'-monophosphate (cAMP) levels by expression of a cAMP-specific phosphodiesterase decreases intrinsic pulsatile gonadotropin-releasing hormone secretion from GT1 cells

Pulsatile GnRH secretion is an intrinsic property of GnRH neurons. Since increases in cAMP levels increase excitability and GnRH secretion in the GT1-1 GnRH cell line, we asked whether cAMP levels play a role in timing excitability and intrinsic pulsatile GnRH secretion. The expression of the cAMP-specific phosphodiesterase (PDE4D1) was used in a genetic approach to lower cAMP levels. Cells were infected with an adenovirus vector (Ad) expressing PDE4D1 (PDE-Ad), or for controls with an empty Ad (Null-Ad) or an Ad expressing green fluorescent protein (GFP-Ad). Infection with the PDE-Ad significantly inhibited forskolin-induced increases in cAMP production, GnRH secretion, and Ca2+ oscillations. Infection of GT1-1 cells with the PDE-Ad vs. GFP-Ad or Null-Ad controls significantly decreased spontaneous Ca2+ oscillations and inhibited the frequency of GnRH pulses. These data support the hypothesis that the level of cAMP in GT1 neurons is a component of the biological clock timing neuron excitability and pulsatile GnRH secretion. Genetically targeted expression of PDE4D1 represents a powerful approach to study the role of cAMP levels in specific populations of neurons in transgenic animals.     

Regulation of cyclic adenosine 3',5'-monophosphate signaling and pulsatile neurosecretion by Gi-coupled plasma membrane estrogen receptors in immortalized gonadotropin-releasing hormone neurons

Immortalized GnRH neurons (GT1-7) express receptors for estrogen [estrogen receptor-alpha and -beta(ERalpha and ERbeta)] and progesterone (progesterone receptor A) and exhibit positive immunostaining for both intracellular and plasma membrane ERs. Exposure of GT1-7 cells to picomolar estradiol concentrations for 5-60 min caused rapid, sustained, and dose-dependent inhibition of cAMP production. In contrast, treatment with nanomolar estradiol concentrations for 60 min increased cAMP production. The inhibitory and stimulatory actions of estradiol on cAMP formation were abolished by the ER antagonist, ICI 182,780. The estradiol-induced inhibition of cAMP production was prevented by treatment with pertussis toxin, consistent with coupling of the plasma membrane ER to an inhibitory G protein. Coimmunoprecipitation studies demonstrated an estradiol-regulated stimulatory interaction between ERalpha and Galphai3 that was prevented by the ER antagonist, ICI 182,780. Exposure of perifused GT1-7 cells and hypothalamic neurons to picomolar estradiol levels increased the GnRH peak interval, shortened peak duration, and increased peak amplitude. These findings indicate that occupancy of the plasma membrane-associated ERs expressed in GT1-7 neurons by physiological estradiol levels causes activation of a Gi protein and modulates cAMP signaling and neuropeptide secretion.     

The relationship between pulsatile GnRH secretion and cAMP production in immortalized GnRH neurons

In perifused immortalized GnRH neurons (GT1-7), simultaneous measurements of GnRH and cAMP revealed that the secretory profiles for both GnRH and cAMP are pulsatile. An analysis of GnRH and cAMP pulses in 16 independent experiments revealed that 25% of pulses coincide. Inversion of the peak and nadir levels was found in 33% and random relationship between GnRH and cAMP found in 42% of analyzed pulses. The random relation between GnRH and cAMP pulse resets to synchronous after an inverse relation between pulses occurred during the major GnRH release, indicating that GnRH acts as a switching mechanism to synchronize cAMP and GnRH release in perifused GT1-7 neurons. Activation of GnRH receptors with increasing agonist concentrations caused a biphasic change in cAMP levels. Low nanomolar concentrations increased cAMP production, but at high concentrations the initial increase was followed by a rapid decline to below the basal level. Blockade of the GnRH receptors by peptide and nonpeptide antagonists generated monotonic nonpulsatile increases in both GnRH and cAMP production. These findings indicate that cAMP positively regulates GnRH secretion but does not participate in the mechanism of pulsatile GnRH release.     

Testosterone processing by pituitary cells in culture: an examination of the role of 5 alpha-reduction in androgen action on the gonadotroph

Dispersed rat pituitary cells were exposed to [1,2,6,7-3H]testosterone ([3H]T, 10(-8) M) to assess the role of 5 alpha-reduction in T regulation of gonadotroph secretion. After 4 to 48 hours of exposure, [3H]T metabolites isolated by thin-layer chromatography were characterized in medium and cell homogenates as well as bound to androgen receptors salt-extracted from purified nuclear pellets. Receptor-bound 5 alpha-[3H]dihydrotestosterone ([3H]DHT)/total [3H]androgens rose progressively from 16% at 4 hours to more than 50% at 48 hours. Coincubation with 4-MA (10- to 1,000-fold molar excess) or testosterone-17 beta-carboxylic acid (TCA; 1,000-fold excess) reduced receptor-bound [3H]DHT/[3H]androgen to less than 10% and 20%, respectively, but elevated [3H]T-receptor levels. Despite inhibiting 5 alpha-reductase activity, TCA and 4-MA had no effect on T suppression of gonadotropin-releasing hormone-stimulated luteinizing hormone secretion or T enhancement of total (cell + secreted) follicle-stimulating hormone levels. The results suggest that 5 alpha-reduction to DHT is not essential for the expression of the direct influences of T on gonadotropin synthesis and secretion in rat gonadotrophs.     

Expression of a functional g protein-coupled receptor 54-kisspeptin autoregulatory system in hypothalamic gonadotropin-releasing hormone neurons

The G protein-coupled receptor 54 (GPR54) and its endogenous ligand, kisspeptin, are essential for activation and regulation of the hypothalamic-pituitary-gonadal axis. Analysis of RNA extracts from individually identified hypothalamic GnRH neurons with primers for GnRH, kisspeptin-1, and GPR54 revealed expression of all three gene products. Also, constitutive and GnRH agonist-induced bioluminescence resonance energy transfer between Renilla luciferase-tagged GnRH receptor and GPR54 tagged with green fluorescent protein, expressed in human embryonic kidney 293 cells, revealed heterooligomerization of the two receptors. Whole cell patch-clamp recordings from identified GnRH neurons showed initial depolarizing effects of kisspeptin on membrane potential, followed by increased action potential firing. In perifusion studies, treatment of GT1-7 neuronal cells with kisspeptin-10 increased GnRH peak amplitude and duration. The production and secretion of kisspeptin in cultured hypothalamic neurons and GT1-7 cells were detected by a specific RIA and was significantly reduced by treatment with GnRH. The expression of kisspeptin and GPR54 mRNAs in identified hypothalamic GnRH neurons, as well as kisspeptin secretion, indicate that kisspeptins may act as paracrine and/or autocrine regulators of the GnRH neuron. Stimulation of GnRH secretion by kisspeptin and the opposing effects of GnRH on kisspeptin secretion indicate that GnRH receptor/GnRH and GPR54/kisspeptin autoregulatory systems are integrated by negative feedback to regulate GnRH and kisspeptin secretion from GnRH neurons.     

Nuclear exclusion of the androgen receptor by melatonin

Androgen receptors (AR) play a crucial role in androgen-mediated processes and prostate cancer progression. The pineal hormone melatonin attenuates the androgen-dependent growth of benign and cancer prostate epithelial cells in vitro and may reverse clinical resistance to androgen ablation therapy in patients progressing on gonadotropin releasing hormone (GnRH) analogue. Where along the AR cascade does melatonin act remains to be determined. The effects of melatonin on AR localization, level and activity were assessed using androgen-insensitive prostate carcinoma PC3 cells stably transfected with a wild-type AR-expressing vector (PC3-AR).AR was localized to the PC3-AR cell nucleus in the absence of dihydrotestosterone (DHT). Melatonin caused a robust exclusion of the AR from the cell nucleus to the cytoplasm. The nuclear export inhibitor, leptomycin B prevented this process. The exclusion was selective since melatonin had no such effect on the nuclear localization of estrogen receptors alpha (ERalpha) in these cells.Melatonin also caused nuclear exclusion of the AR in the presence of DHT. In addition, it attenuated androgen induced reporter gene activity in PC3 cells co-transfected with the human AR and AR reporter plasmids. Elevated androgen concentrations counteracted melatonin's effects. Melatonin did not decrease AR level or androgen binding in the cells.The nuclear localization of the AR is a hallmark of its cellular activity. These data point to AR nuclear exclusion as a possible mechanism to attenuate androgen responses in target tissues.     

The role of transforming growth factor-β1, -β2, and -β3 in androgen-responsive growth of NRP-152 rat prostatic epithelial cells

We have investigated the role of autocrine/paracrine TGF-β secretion in the regulation of cell growth by androgens as demonstrated by its inhibition by two androgen response modifiers; the nonsteroidal antiandrogen hydroxyflutamide (OHF), believed to act by inhibiting androgen binding to androgen receptors, or finasteride, an inhibitor of 5α-reductase, the enzyme necessary for the conversion of testosterone to 5α-dihydrotestosterone (DHT), using the nontumorigenic rat prostatic epithelial cell line NRP-152. Growth of these cells was stimulated three- to sixfold over control by either testosterone or DHT under serum-free culture conditions. This was accompanied by a two- to threefold decrease in the secretion rate of TGF-β1, -β2, and -β3. Finasteride reversed the ability of testosterone but not DHT to stimulate growth and downregulate expression of TGF-β1, -β2, and -β3 in a dose-dependent fashion, suggesting that this activity of testosterone required its conversion to DHT. OHF antagonized the stimulatory effects of DHT on NRP-152 cell growth but could reverse the inhibitory effects of DHT only on TGF-β2 and TGF-β3 and not TGF-β1 secretion. This suggests that either TGF-β1 regulation by DHT or the androgen antagonism of OHF occurs independent of androgen receptor binding. Neutralizing antibodies to TGF-β (pantropic and isoform-specific) were able to block the ability of finasteride to antagonize the effects of testosterone nearly completely while only partially inhibiting the antiandrogenic effects of OHF. Thus, the ability of androgens to stimulate growth of NRP-152 cells involves the downregulation of the production of TGF-β1, -β2, and -β3 in addition to other growth-stimulatory mechanisms. J. Cell. Physiol. 175:184–192, 1998. Published 1998 Wiley-Liss, Inc.

1 This article is a US Government work and, as such, is in the public domain in the United States of America.

     

Androgens are not major down-regulators of androgen receptor levels during growth of the immature rat penis

This study was undertaken to investigate the prevalent hypothesis that androgens are responsible for the organ-specific down-regulation of penile androgen receptors (ARs) and decline of penile growth in the rat during sexual maturation. Sexually immature male rats (21 days old) were castrated and treated for 3 days (“short-term”), with high doses of: (a) testosterone and the -reductase inhibitor finasteride (T/F); (b) dihydrotestosterone (DHT); or (c) finasteride alone (F). Intact and castrate controls received vehicle only. PolyA + RNA was analysed by Northern blot hybridization and ARs were estimated in the penis and ventral prostates by (3-H)R-1881 binding in the cytosol. Short-term castration, with or without F, increased penile AR mRNA, whereas high doses of T/F and DHT reduced it considerably. Although penile cytosol AR concentration in the control castrates, with or without F, paralleled the AR mRNA rise, treatment with androgens left cytosol AR content per organ and AR concentration above those of the intact rat penis despite the drop in AR mRNA. A “long-term” treatment (10 days) on 19-day-old rats with either medium or high doses of T/F and DHT also failed to down-regulate penile cytosol ARs below the intact controls. Western blot analysis of penile cytosol AR levels confirmed these results. Block of pituitary FSH and LH release by a GnRH antagonist in castrates receiving T/F or DHT at high doses did not modify the response. In the case of intact rats, high doses of T/F or DHT actually increased penile cytosol AR content. No difference was observed between T/F and DHT effects. In contrast to what occurs during sexual maturation, the prostate ARs and growth rate responded to all treatments in a similar way to what was observed in the penis. Our results suggest that increases in serum T or DHT are not major factors in the physiological down-regulation of ARs and androgen-dependent growth in the rat corpora cavernosa.     

Dominant-negative androgen receptor inhibition of intracrine androgen-dependent growth of castration-recurrent prostate cancer

Background

Prostate cancer (CaP) is the second leading cause of cancer death in American men. Androgen deprivation therapy is initially effective in CaP treatment, but CaP recurs despite castrate levels of circulating androgen. Continued expression of the androgen receptor (AR) and its ligands has been linked to castration-recurrent CaP growth.

Principal Finding

In this report, the ligand-dependent dominant-negative ARΔ142–337 (ARΔTR) was expressed in castration-recurrent CWR-R1 cell and tumor models to elucidate the role of AR signaling. Expression of ARΔTR decreased CWR-R1 tumor growth in the presence and absence of exogenous testosterone (T) and improved survival in the presence of exogenous T. There was evidence for negative selection of ARΔTR transgene in T-treated mice. Mass spectrometry revealed castration-recurrent CaP dihydrotestosterone (DHT) levels sufficient to activate AR and ARΔTR. In the absence of exogenous testosterone, CWR-R1-ARΔTR and control cells exhibited altered androgen profiles that implicated epithelial CaP cells as a source of intratumoral AR ligands.

Conclusion

The study provides in vivo evidence that activation of AR signaling by intratumoral AR ligands is required for castration-recurrent CaP growth and that epithelial CaP cells produce sufficient active androgens for CaP recurrence during androgen deprivation therapy. Targeting intracrine T and DHT synthesis should provide a mechanism to inhibit AR and growth of castration-recurrent CaP.     

Testosterone-Dependent Interaction between Androgen Receptor and Aryl Hydrocarbon Receptor Induces Liver Receptor Homolog 1 Expression in Rat Granulosa Cells

Androgens play a major role in the regulation of normal ovarian function; however, they are also involved in the development of ovarian pathologies. These contrasting effects may involve a differential response of granulosa cells to the androgens testosterone (T) and dihydrotestosterone (DHT). To determine the molecular pathways that mediate the distinct effects of T and DHT, we studied the expression of the liver receptor homolog 1 (LRH-1) gene, which is differentially regulated by these steroids. We found that although both T and DHT stimulate androgen receptor (AR) binding to the LRH-1 promoter, DHT prevents T-mediated stimulation of LRH-1 expression. T stimulated the expression of aryl hydrocarbon receptor (AHR) and its interaction with the AR. T also promoted the recruitment of the AR/AHR complex to the LRH-1 promoter. These effects were not mimicked by DHT. We also observed that the activation of extracellular regulated kinases by T is required for AR and AHR interaction. In summary, T, but not DHT, stimulates AHR expression and the interaction between AHR and AR, leading to the stimulation of LRH-1 expression. These findings could explain the distinct response of granulosa cells to T and DHT and provide a molecular mechanism by which DHT negatively affects ovarian function.     

Molecular biology of androgen action: testosterone/dihydrotestosterone receptor and androgen 5 alpha-reductase in the human foreskin

Receptors for testosterone (T) and dihydrotestosterone (DHT) as well as the tissue specific androgen-5 alpha-reductase (A5R) were studied in the foreskin of 52 healthy boys (ages 1-14 years), in order to gain molecular endocrinological data and information about the ontogeny and cytogeny, respectively, of androgen specific target organs. Enzyme determinations were carried out in tissue homogenates by an enzyme kinetic method for the evaluation of Km- and Vmax-values. Reactions velocities were calculated from the turn over rates of T to DHT, 5 alpha-androstane-3 alpha,17 beta-diol and 5 alpha-androstane-3,beta,17 beta-diol. The precursor (T) was used in increasing concentrations, ranging from 8 to 208 nM. Separation of reaction products was done by thin-layer chromatography and verification of specific radioactivity of metabolites by means of radio gas chromatography on capillary columns. Results of the enzyme analyses: Km = 94.9 +/- 3.5 [nM], and Vmax = 15.8 +/- 1.9 [pmol/mg.h]. Receptors were examined in the cytosolic and nuclear fractions of the tissue specimens. Saturation analyses and calculation of binding data led to specific receptors for T and DHT in the cytosolic (T: Kd = 1.56 +/- 0.12 [nM], Nmax = 122.4 +/- 11.6 [fmol/mg]; DHT: Kd = 1.9 +/- 0.1 [nM], Nmax = 493.3 +/- 77.8 [fmol/mg]) and the nuclear fractions (T: Kd = 1.43 +/- 0.13 [nM], Nmax = 28.7 +/- 3.5 [fmol/mg]; DHT: Kd = 1.37 [nM], Nmax = 196.9 +/- 22.5 [fmol/mg]), Kd-values proved to be quite homogenous (coeff. var. = 0.15-0.21), whereas maximum specific receptor binding activities (Nmax) showed age dependent fluctuations (coeff. var. = 0.35-0.45). Binding capacities of both T- and DHT-receptor, respectively, in cytosolic and nuclear fractions showed peak values in the age group 10-11 years and additional "spikes" of binding rates at age 4-5 years. It is noteworthy that Vmax-values also reached maximum levels in the latter age group. Concerning the ontogeny of the androgen receptor a change of binding properties from the cytosol to the nuclear fraction was observed with the onset of puberty. A comparison of enzyme- and receptor data lead to the theory, that subcellular hormone actions depend on interrelational regulatory mechanisms between androgen receptors (T as well as DHT) and specific enzyme systems (A5R).     

Androgen synthesis during follicular development: evidence that rat granulosa cell 17-ketosteroid reductase is independent of hormonal regulation

Although androgens have been implicated in follicular atresia, ovarian follicular androgen synthesis is required for preovulatory follicular growth. To localize the site(s) of androgen biosynthesis and to obtain a better understanding of the regulation of the androgenic pathway(s) in rat ovarian follicles we examined the relative abilities of developing follicles to accumulate specific androgens [testosterone (T) and dihydrotestosterone (DHT)] using both radioimmunoassay (RIA) and 3H-substrate metabolism techniques. Small antral and preovulatory follicles were obtained from control or human chorionic gonadotropin (hCG)-primed immature rats, respectively (Richards and Bogovich, 1982). Small antral follicles, theca and granulosa cells produced little immunoassayable androgen (T + DHT) when incubated with or without 8-bromo-cAMP. In contrast, preovulatory follicles and theca produced more androgen than small antral tissues and in a manner acutely stimulable by cAMP. Granulosa cells produced little androgen under these conditions. Inclusion of [3H] androstenedione in the incubates yielded increased accumulation of [3H] T and [3H] DHT for all small antral and preovulatory tissues. Indeed, granulosa cells from both small antral and preovulatory follicles possessed a remarkable ability to accumulate [3H] T. This ability was not altered by hypophysectomy or subsequent treatment with estradiol and/or follicle-stimulating hormone (FSH). These results suggest that 17-ketosteroid reductase may be a constitutive enzyme in granulosa cells.     

Antiproliferative actions of the synthetic androgen, mibolerone, in breast cancer cells are mediated by both androgen and progesterone receptors

Androgen signaling, mediated by the androgen receptor (AR), is a critical factor influencing growth of normal and malignant breast cells. Given the increasing use of exogenous androgens in women, a better understanding of androgen action in the breast is essential. This study compared the effects of 5alpha-dihydrotestosterone (DHT) and a synthetic androgen, mibolerone, on estradiol (E(2))-induced proliferation of breast cancer cells. DHT modestly inhibited E(2)-induced proliferation and mibolerone significantly inhibited proliferation in T-47D cells. The effects of both androgens could be reversed by an AR antagonist, suggesting that their actions were mediated, in part, by AR. Whereas high physiological doses (10-100nM) of DHT reduced E(2)-mediated induction of the estrogen-regulated gene progesterone receptor (PR) to basal levels, mibolerone at lower doses (1nM) eliminated PR expression, suggesting that mibolerone may also act via the PR. In the AR positive, PR-negative MCF-7 cells, mibolerone had modest effects on E(2)-induced proliferation, but was a potent inhibitor of proliferation in the AR positive, PR positive MCF-7M11 PRA cells. The effects of mibolerone in breast cancer cells were similar to those of the progestin, medroxyprogesterone acetate. Our results demonstrate that mibolerone can have both androgenic and progestagenic actions in breast cancer cells.     

Regulation of thyroid hormones on the production of testosterone in rats

The effects of a thyroidectomy and thyroxine (T4) replacement on the spontaneous and human chorionic gonadotropin (hCG)-stimulated secretion of testosterone and the production of adenosine 3',5'-cyclic monophosphate (cAMP) in rat testes were studied. Thyroidectomy decreased the basal levels of plasma luteinizing hormone (LH) and testosterone, which delayed the maximal response of testosterone to gonadotropin-releasing hormone (GnRH) and hCG in male rats. T4 replacement in thyroparathyroidectomized (Tx) rats restored the concentrations of plasma LH and testosterone to euthyroid levels. Thyroidectomy decreased the basal release of hypothalamic GnRH, pituitary LH, and testicular testosterone as well as the LH response to GnRH and testosterone response to hCG in vitro. T4 replacement in Tx rats restored the in vitro release of GnRH, GnRH-stimulated LH release as well as hCG-stimulated testosterone release. Administration of T4 in vitro restored the release of testosterone by rat testicular interstitial cells (TICs). The increase of testosterone release in response to forskolin and androstenedione was less in TICs from Tx rats than in that from sham Tx rats. Administration of nifedipine in vitro resulted in a decrease of testosterone release by TICs from sham Tx but not from Tx rats. The basal level of cAMP in TICs was decreased by thyroidectomy. The increased accumulation of cAMP in TICs following administration of forskolin was eliminated in Tx rats. T4 replacement in Tx restored the testosterone response to forskolin. But the testosterone response to androstenedione and the cAMP response to forskolin in TICs was not restored by T4 in Tx rats. These results suggest that the inhibitory effect of a thyroidectomy on the production of testosterone in rat TICs is in part due to: 1) the decreased basal secretion of pituitary LH and its response to GnRH; 2) the decreased response of TICs to gonadotropin; and 3) the diminished production of cAMP, influx of calcium, and activity of 17beta-HSD. T4 may enhance testosterone production by acting directly at the testicular interstitial cells of Tx rats.