ABSTRACT

The gonadotrophic function of the European eel (Anguilla anguilla L.) at the silver stage is very weak: gonadotrophin-releasing hormone (GnRH) secretion is deficient and, moreover, dopamine inhibition overrides GnRH action. At the silver stage, œstradiol stimulates the biosynthesis of the type-II gonadotrophin (GTH-II). To study the molecular mechanism of this activation further, we examined the effect of testosterone and œstradiol administration on pituitary levels of mRNA encoding GTH-II α and β subunits. Corresponding eel cDNA probes and Northern blot analysis were used. After 2 weeks, testosterone and œstradiol implantation resulted in a strong increase in mRNA encoding the GTH-II β subunit (7-fold and 25-fold, respectively) and in a slight, but non-significant, rise in the a subunit mRNA level (1.8-fold and 1.5-fold, respectively). Co-implantation of these two steroids suggested a potentiation of their effects on the β subunit (104-fold) while an additive effect was indicated on the α mRNA level (2.9-fold). Effects were detectable within 4 days and were maximal 4 weeks after implantation. These results indicate that in the European eel at the silver stage, gonadal steroids stimulate differentially the expression of GTH-II subunit genes at a pretranslational level.

Understanding of the principal pathways of steroid hormone biosynthesis was established over two decades ago through advances in steroid radioisotopic and chromatographic techniques. When the enzymes of individual pathways could be examined in more detail, the dissection of the complex pattern of enzyme activities began. At many points, separate pathways employ precisely the same enzyme for equivalent catalytic steps, e.g. for 21-hydroxylase, 11 β-hydroxylase, aromatase and several dehydrogenases (Orth et al. 1992). A further economy was found for 17α-hydroxylase and 17,20-lyase activities, which co-purify with the same P450c17 polypeptide. This enzyme was later cloned and expressed in tissue culture cells, revealing that, contrary to the enzyme in rat, human and cattle, 17α-hydroxylase cannot convert 17α-hydroxyprogesterone to androstenedione (Bradshaw et al. 1987, Fevold et al. 1989). Further complexity emerged with the existence of multiple tissue-specific forms of 5α-reductase (Wilson et al. 1993), and 3β-, 11β- and 17β-hydroxysteroid dehydrogenases, most of which

Upregulation of the steroidogenic acute regulatory protein (StAR) is implicated in the rapid synthesis and secretion of steroidogenic cells to produce steroids in response to stimulation by trophic hormones of the gonadal and stress axes. In the present study, we have assessed the kinetics of both StAR gene transcription and protein biosynthesis in primary cell cultures of bovine adrenocortical and ovarian theca cells, under conditions of acute stimulation by corticotrophin (ACTH) and luteinizing hormone (LH), respectively. In both cell systems, detectable upregulation of StAR gene transcription occurred within 1-2 h, reaching maxima at 4 h (theca cells) or 6 h (adrenocortical cells). mRNA levels returned rapidly to baseline, by 12 h or 24 h, respectively. Specific StAR protein levels were assessed by western blotting using a novel antibody raised against a bovine StAR peptide, and showed a similar fast upregulation, albeit delayed by 1-2 h compared with the mRNA. The response of the cultured theca cells was more acute than that of the adrenocortical cells, possibly reflecting the propensity of the LH receptor to desensitize rapidly, unlike the ACTH receptor. The primary bovine theca cell cultures were also used for fully homologous transfection studies using various deletion promoter-reporter constructs of the bovine StAR gene. Kinetic analysis of the results indicated that the acute transcriptional response resides within the proximal (-315 bp) promoter region, which includes two putative responsive elements for the steroidogenic factor-1. More distal promoter regions may be involved in modulating the specificity of expression by combining enhancer and inhibitory functions.

Steroid hormone biosynthesis in the adrenal cortex is controlled by the peptide hormone adrenocorticotropin (ACTH), which acts to increase intracellular cAMP and results in the activation of cAMP-dependent protein kinase A (PKA) and subsequent increase in steroidogenic gene transcription. Protein phosphorylation by PKA activates transcription of genes encoding steroidogenic enzymes; however the precise proteins which are phosphorylated remain to be determined. We have recently shown that phosphoprotein phosphatase (PP) activity is essential for cAMP-dependent transcription of the human CYP17 (hCYP17) gene in H295R adrenocortical cells. The aim of our current studies was to determine if inhibition of PP activity attenuates cAMP-dependent mRNA expression of other steroidogenic genes in H295R cells. Using various inhibitors of serine/threonine and tyrosine PPs, we examined the role of phosphatase activity on cAMP-dependent transcription of steroidogenic genes in the adrenal cortex. CYP11A, CYP11B1/2, CYP21, and adrenodoxin also require PP activity for cAMP-stimulated gene expression. Inhibition of both serine/threonine and tyrosine PP activities suppresses the cAMP-dependent mRNA expression of several steroidogenic genes, suggesting that a dual-specificity PP is essential for conveying ACTH/cAMP-stimulated transcription. We propose that PKA phosphorylates and activates a dual-specificity phosphatase, which mediates steroidogenic gene transcription in response to ACTH/cAMP.

ABSTRACT

To elucidate the structure and control of expression of the porcine FSH-β subunit gene, two genomic clones were isolated and the entire gene structure was determined to the extent of 10 kb, consisting of 6 kb of the 5′-flanking region and 4 kb of the transcriptional unit. The porcine FSH-β gene consisted of three exons the same as the human and bovine genes, but the positions of both splicing sites of porcine intron-1 were unique. It is known that the synthesis of FSH is regulated by gonadal steroids, gonadotrophin-releasing hormone (GnRH) and inhibin. However, the consensus steroid-responsive element was unexpectedly absent in the 5′-flanking region of 6 kb. On the other hand, the potential binding sites for activator protein-1 (AP1) and AP2, which might be stimulated by the GnRH—protein kinase C cascade, were present at seven and five positions respectively. An imperfect cyclic AMP-responsive element was also present. Southern blot analyses, using the cDNA and genomic fragments as probes, gave smear patterns suggesting the presence of repetitive sequences in the porcine FSH-β gene. A survey of homology with the repetitive sequences revealed that short interspersed repeated sequences (SINES)-type non-viral retroposons were present with about 250 bp length repeats twice in the 5′-flanking region and once each in intron-1 and the 3′-flanking region. Other SINES-like sequences were also found in intron-1, exon-2 and exon-3. In comparison with the 5′-flanking sequences of the porcine α and LH-β genes, there were no significantly conserved regions, implying a lack of common modulation of the three subunit genes.

Liver receptor homologue-1 (LRH-1, designated NR5A2) is a mammalian homologue of Drosophila fushi tarazu factor (dFTZ-F1) and structurally belongs to the orphan nuclear receptor superfamily. LRH-1 can recognize the DNA sequence 5'-AAGGTCA-3', the canonical recognition motif for steroidogenic factor 1 (SF-1). Herein, we hypothesized that LRH-1 might play a role in the regulation of human adrenal expression of steroidogenic enzymes. To test this hypothesis, LRH-1 expression in human adult and fetal adrenal glands was examined by RT-PCR analysis. The fetal and adult adrenal glands, as well as liver and pancreas, were observed to express LRH-1 mRNA using RT-PCR. The ability of LRH-1 to enhance transcription of the gene encoding human 11 beta- hydroxylase (hCYP11B1) was then examined using the H295R adrenal cell line. LRH-1 co-transfection with hCYP11B1 luciferase promoter constructs caused a 25-fold induction of luciferase activity. Furthermore, co-transfection of a hCYP11B1 reporter construct containing a mutation in the SF-1 binding cis-element abolished the stimulatory effect of both SF-1 and LRH-1. Electrophoretic mobility shift assay (EMSA) demonstrated that LRH-1 could bind to the SF-1 response element. Taken together, our data suggested that LRH-1 is expressed in the adrenal, and can substitute for SF-1 to enhance transcription of genes encoding certain of the steroid-metabolizing enzymes. A role for LRH-1 in the regulation of adrenal or gonadal steroid hormone production should be further studied.

Human 3beta-hydroxysteroid dehydrogenase/isomerase (3beta-HSD) is a key steroidogenic enzyme that catalyzes the first step in the conversion of circulating dehydroepiandrosterone (DHEA), pregnenolone or 17alpha-hydroxypregenolone to produce the appropriate, active steroid hormone(s): estradiol, testosterone, progesterone, aldosterone or cortisol respectively. Our mutagenesis studies have identified Tyr154 and Lys158 as catalytic residues for the 3beta-HSD reaction. Our three-dimensional homology model of 3beta-HSD shows that Tyr154 and Lys158 are oriented near the 3beta-hydroxyl group of the bound substrate steroid, and predicts that Ser123 or Ser124 completes a Tyr-Lys-Ser catalytic triad that operates in many other dehydrogenases. The S123A and S124A mutants of human type 1 3beta-hydroxysteroid dehydrogenase/isomerase (3beta-HSD1) were created by PCR-based mutagenesis, expressed in insect cells using baculovirus and purified to homogeneity. The S124A mutant exhibits no 3beta-HSD activity and has a K(m) value (83.6 microM) for the isomerase substrate that is threefold greater than that of wild-type 1 isomerase. In contrast, S123A has substantial 3beta-HSD activity (DHEA K(m)=11.2 microM; k(cat)=0.8 min(-1)) and utilizes isomerase substrate, 5-androstene-3,17-dione, with a K(m) value (27.6 microM) that is almost identical to wild-type. The K(m) value (4.3 microM) of S124A for NADH as an allosteric activator of isomerase is similar to that of the wild-type 1 enzyme, indicating that Ser124 is not involved in cofactor binding. S123A utilizes NAD as a cofactor for 3beta-HSD and NADH as the activator for isomerase with K(m) values that are similar to wild-type. The 3beta-HSD activities of S123A and wild-type 3beta-HSD increase by 2.7-fold when the pH is raised from 7.4 to the optimal pH 9.7, but S124A exhibits very low residual 3beta-HSD activity that is pH-independent.These kinetic analyses strongly suggest that the Ser124 residue completes the catalytic triad for the 3beta-HSD activity. Since there are 29 Ser residues in the primary structure of human 3beta-HSD1, our homology model of the catalytic domain has been validated by this accurate prediction. A role for Ser124 in the binding of the isomerase substrate, which is the 3beta-HSD product-steroid of the bifunctional enzyme protein, is also suggested. These observations further characterize the structure/function relationships of human 3beta-HSD and bring us closer to the goal of selectively inhibiting the type 1 enzyme in placenta to control the timing of labor or in hormone-sensitive breast tumors to slow their growth.

ABSTRACT

There are significant differences between rats and mice in the gonadal regulation of several aspects of gonadotroph function. To investigate whether these extend to the pretranslational regulation of FSH synthesis by gonadal steroids, we have measured FSH-β mRNA levels following gonadectomy and sex-steroid replacement and have related these to serum and pituitary FSH as a reflection of overall hormone synthesis.

In ovariectomized rats, FSH-β mRNA levels increased by 8 h, decreased, and then rose progressively over the next 28 days. A similar pattern of response was observed in orchidectomized rats. In mice, there were progressive increases in FSH-β mRNA levels in both males and females following gonadectomy, without evidence of the early peaks observed in rats. In both species, the change in FSH-β mRNA levels after gonadectomy was greater in females than in males. These changes in FSH-β mRNA following gonadectomy were paralleled by changes in the serum FSH concentration. In ovariectomized female rats and mice, pituitary FSH stores increased by 8 h and 3 days respectively, whereas in male rats, pituitary FSH content did not rise until 10 days after orchidectomy. The most striking species difference was the marked and prolonged reduction of pituitary FSH after orchidectomy of mice.

Treatment of rats and mice from the time of ovariectomy, with a dose of oestradiol that prevents increases in serum LH, only partially attenuated the rises in FSH-β mRNA and serum FSH and did not prevent the increase in pituitary FSH content. Treatment of intact or orchidectomized rats with testosterone suppressed FSH-β mRNA levels to 50% below intact control values without affecting pituitary FSH content. In mice, testosterone treatment for 10 days reduced the post-castration increase in FSH-β mRNA by only 26%, and prevented the fall in pituitary FSH content, although the increased serum concentration of FSH was unaffected.

In conclusion: (1) there is a good correlation between FSH-β mRNA levels and overall FSH biosynthesis in male and female rats and female mice, but this relationship is less obvious in male mice where pituitary FSH stores are not increased; (2) the inability of oestradiol to prevent completely the post-ovariectomy increase in FSH-β mRNA and FSH synthesis in female rats and mice indicates either that other gonadal products are necessary or that higher doses of oestradiol are required than for complete suppression of LH synthesis; (3) whilst the post-gonadectomy increases in FSH-β mRNA are larger in the female of both species, there are no major differences between rats and mice in the regulation of FSH-β gene expression by sex steroids.

ABSTRACT

The gene for the common α subunit of the porcine anterior pituitary glycoprotein hormones was cloned from a genomic library constructed in EMBL3. The nucleotide sequence of the entire coding sequence of the porcine common α-subunit gene was determined in addition to one intron and 1059 and 160 bp of the 5′-and 3′-flanking regions respectively. Southern blot analysis of the porcine genomic DNA indicated that the common α-subunit gene is present as a single copy. The transcriptional unit of the porcine common α subunit spanned about 14kb and contained four exons interrupted by three introns of about 11.5, 1.2 and 0.4kb. The short untranslated sequence in the first exon and the location of the exon/intron junctions at amino acid residues +9/+10 and +71/+72 were highly conserved among the rat, human and bovine common α-subunit genes. In the proximal portion of the 5′-flanking region, one TATA box and one CCAAT box were present. A steroid-responsive element was not found up to 1059 bases upstream from the transcription start site. The potential AP-1 and AP-2 factor-responsive elements were present at three and one positions respectively in the 5′-flanking region. This feature suggests that hypothalamic gonadotrophin-releasing hormone stimulates the expression of the common α-subunit gene predominantly by a signal-transduction system, with the protein kinase C cascade and factors AP-1 and AP-2 as mediators. The cyclic AMP-responsive element was also present at two positions, but a single base substitution was found in each sequence compared with the consensus sequence. The porcine common α-subunit gene has a structure distinct from its counterparts, the porcine FSH-β and LH-β genes, reflecting differential control of their synthesis during gametogenesis.

ABSTRACT

A radioimmunoassay was developed and validated for the major glycoprotein (gp58) of the mouse mammary tumour virus (MMTV). Using this assay, the expression of gp58 during pregnancy and lactation was found to parallel that for MMTV RNA. In particular, there was a very rapid induction in late pregnancy and a decline in late lactation, although some residual expression persisted well into involution. In cultures of normal mouse mammary tissue, induction of gp58 occurred after a 24-h lag period and began to reach a plateau after 3 days. Both the insulin and prolactin dose—response curves for gp58 resembled those for MMTV RNA; in contrast, the effects of steroid hormones on gp58 and MMTV RNA were disparate. Although progesterone stimulated the RNA, it only slightly increased gp58 levels; however, the presence of cortisol greatly augmented this stimulation, despite the inability of cortisol to induce RNA at physiological concentrations. These results suggest that insulin, prolactin and progesterone act primarily at the level of RNA accumulation in normal mammary epithelium, while cortisol affects some more distal event.