Glucocorticoid and thyroid hormones transcriptionally regulate growth hormone gene expression

The c-erb-A gene encodes a thyroid hormone receptor

The cDNA sequence of human c-erb-A, the cellular counterpart of the viral oncogene v-erb-A, indicates that the protein encoded by the gene is related to the steroid hormone receptors. Binding studies with the protein show it to be a receptor for thyroid hormones.

The 5'-flanking sequences of the human growth hormone gene contain a cell-specific control element

The 5'-flanking sequences of the human growth hormone (hGH) gene contain a cell-specific control element. Hybrid genes containing truncated 5'-flanking DNA fragments from the hGH gene fused to the chloramphenicol acetyl transferase (cat) gene were examined using transient transfection of rat anterior pituitary (GC) and nonpituitary cell lines (HeLa, Rat 2, and KB); preferential expression of these gene hybrids was only observed in GC cells. Deletions through the 5'-flanking sequences of the hGH gene revealed that the region containing nucleotides -230 to -180 is required for efficient cat gene expression in GC cells. This region of DNA is highly homologous to a region of the rGH gene that contains a tissue-specific control element. A hybrid gene containing the tissue-specific control element, but lacking the proximal promoter elements, of the hGH gene upstream from the promoter of the thymidine kinase gene (TKp) from herpes simplex virus ligated to the cat gene produced more CAT activity than the TKp.cat gene in GC cells but not in HeLa cells. These data suggest that the tissue-specific control element located in the 5'-region of the hGH gene can act in the presence of a heterologous promoter and is specific for expression in pituitary cells.

Thyroid hormone receptors bind to defined regions of the growth hormone and placental lactogen genes

The intracellular receptor for thyroid hormone is a protein found in chromatin. Since thyroid hormone stimulates transcription of the growth hormone gene through an unknown mechanism, the hypothesis that the thyroid hormone-receptor complex interacts with defined regions of this gene has been investigated in a cell-free system. Nuclear extracts from human lymphoblastoid IM-9 cells containing thyroid hormone receptors were incubated with L-3,5,3'-tri[125I]iodothyronine and calf thymus DNA-cellulose. Restriction fragments of the human growth hormone gene were added to determine their ability to inhibit labeled receptor binding to DNA-cellulose. These fragments encompassed nucleotide sequences from about three kilobase pairs upstream to about four kilobase pairs downstream from the transcription initiation site. The thyroid hormone-receptor complex bound preferentially to the 5'-flanking sequences of the growth hormone gene in a region between nucleotide coordinates -290 and -129. The receptor also bound to an analogous promoter region in the human placental lactogen gene, which has 92% nucleotide sequence homology with the growth hormone gene. These binding regions appear to be distinct from those that are recognized by the receptor for glucocorticoids, which stimulate growth hormone gene expression synergistically with thyroid hormone. The presence of thyroid hormone was required for binding of its receptor to the growth hormone gene promoter, suggesting that thyroid hormone renders the receptor capable of recognizing specific gene regions.

Insulin regulation of rat growth hormone gene transcription

We have previously shown that insulin suppresses growth hormone (GH) messenger (m) RNA levels in rat pituitary cells. To further delineate the molecular mechanism of insulin action, the effect of insulin treatment on GH gene transcription rates was examined in GH3 pituitary cells grown in serum-free defined medium. A transcriptional run-off assay was performed when intact isolated nuclei were allowed to continue RNA synthesis in an in vitro reaction. Specific incorporation of [32P]GTP into RNA was quantified by hybridization to rat GH complementary (c) DNA. Hybridization efficiency was measured with an internal [3H]cRNA standard and ranged from 30 to 48%. Alpha-amanitin (1 microgram/ml) inhibited total transcription, and excess unlabeled rat pituitary mRNA (250 ng) competitively inhibited GH mRNA hybridization by greater than 80%. Insulin (0.7 nM) inhibited new GH mRNA synthesis, and maximal inhibition (30% of control) was observed with 7 nM insulin after 4 h treatment. The inhibitory effects of insulin on new GH mRNA synthesis were abolished by both insulin-receptor-antiserum and by guinea-pig anti-insulin serum. The results show that insulin exerts a rapid suppression of new GH mRNA synthesis. These data provide evidence for the direct transcriptional regulation of the GH gene by insulin.

Stabilization of a specific nuclear mRNA precursor by thyroid hormone

The regulation of a thyroid hormone-responsive gene in rats, designated spot 14, was explored. The expression of this gene in liver is rapidly (less than 10 min) and markedly (greater than 10-fold) altered by the administration of 3,5,3'-triiodo-L-thyronine (T3) to hypothyroid rats (P. Narayan, C. W. Liaw, and H. C. Towle, Proc. Natl. Acad. Sci. USA 81:4687-4691, 1984). To investigate the cellular site at which T3 acts to induce this hepatic mRNA, we made parallel measurements of the relative levels of spot 14 mRNA and nuclear precursor RNA and of the rate of gene transcription after treatments designed to alter the thyroid status of rats. The relative levels of both the mRNA and nuclear precursor were elevated roughly 5- to 6-fold in euthyroid animals and 9- to 12-fold in hyperthyroid animals over those in hypothyroid controls. However, only a small difference of approximately 1.5-fold was detected in the rate of spot 14 gene transcription. After a single injection of T3 into hypothyroid animals, a small and transient rise in the transcription rate was detected at 30 min. However, the levels of spot 14 mRNA and nuclear precursor RNA increased much more dramatically throughout the first 4 h of treatment. In both cases, changes in the rate of gene transcription were not capable of accounting for the alterations observed in mRNA levels. Thus, the major site of spot 14 gene regulation by T3 is at a posttranscriptional level. The proportional changes observed in the nuclear precursor and mRNA levels suggest that the site of control is at the level of stability of the nuclear precursor RNA for spot 14 mRNA.

Mechanisms of glucocorticoid hormone action

This report summarizes our studies, in context with the results of other laboratories, of the molecular mechanisms of glucocorticoid hormone action. The receptors for these steroids are comprised of single polypeptide chains of about 90,000 molecular weight. Binding of agonist steroids to the receptor induces a conformational change to an active receptor form that is followed by a second change in the glucocorticoid-receptor complex, termed activation, that alters the charge of the complex and results in its binding to specific sites on the DNA termed glucocorticoid regulatory elements (GREs). The GRE on the human metallothionein-IIA gene is located in the 5'-flanking DNA. It can function independently of the gene's promoter, and when ligated upstream from the herpes simplex virus (HSV) thymidine kinase (TK) gene promoter, can activate it. The binding of the glucocorticoid-receptor complex to the GRE probably alters chromatin structure over a limited span to facilitate RNA polymerase action. The regulation by glucocorticoids of growth hormone gene expression is more complex. The steroid appears to elicit both transcriptional and posttranscriptional influences that are also affected by thyroid hormone. Also the glucocorticoid influences appear to be exerted in part through DNA structures located downstream from the transcriptional initiation site. A GRE has been defined in intron A of the hGH gene through gene transfer and DNA binding experiments. Finally, gene transfer experiments suggest that pituitary-specific factors influence the ability of glucocorticoids to affect GH gene expression.

Primary structure and expression of a functional human glucocorticoid receptor cDNA

Identification of complementary DNAs encoding the human glucocorticoid receptor predicts two protein forms, of 777 (alpha) and 742 (beta) amino acids, which differ at their carboxy termini. The proteins contain a cysteine/lysine/arginine-rich region which may define the DNA-binding domain. Pure radiolabelled glucocorticoid receptor, synthesized in vitro, is immunoreactive and possesses intrinsic steroid-binding activity characteristic of the native glucocorticoid receptor.

Cloning of the human glucocorticoid receptor cDNA

We show that the human glucocorticoid receptor (GR), isolated from the breast cancer cell line MCF-7, has an apparent molecular weight identical to that of rat liver GR (94 kDa) and reacts with antibodies raised against the latter. These antibodies were used to clone cDNA sequences corresponding to the human GR from a lambda gt11 expression library constructed using MCF-7 poly(A)+ RNA. Three non-homologous cDNA clones with inserts of 125, 220 and 350 bp, which express epitopes recognised by the rat liver GR antibodies, were isolated. Rat liver GR antibodies, immunopurified using the immobilised purified beta-galactosidase fusion proteins, detect partially purified rat liver and human GRs on Western blots. In addition, these antibodies immuno-adsorb rat liver and human GRs affinity-labelled with [3H] triamcinolone acetonide. Northern blot analysis, using all three cDNA probes, reveals the presence of a major MCF-7 poly(A)+ RNA species of approximately 7 kb.

Binding of the human glucocorticoid receptor to defined regions in the human growth hormone and placental lactogen genes

An in vitro competition assay was used to investigate whether binding sites for the human glucocorticoid receptor occur in the human genes for growth hormone (hGH) and placental lactogen (chorionic somatomammotropin, hCS). These genes display 95% sequence homology. Two receptor-binding regions were found in the hGH gene, one of which is located within 290 bp upstream, and one within 251 bp downstream from the transcription initiation site. Two binding regions homologous to those in the hGH gene were found in the hCS gene. The receptor-binding DNA fragment from the structural part of the genes, but not that from their promoter area, contained a sequence homologous to a 15-bp consensus sequence proposed earlier for the glucocorticoid receptor binding site. It is unlikely that the putative difference in glucocorticoid sensitivity between the hGH and hCS genes is accounted for by major differences in glucocorticoid receptor binding pattern.

Stabilization of a specific nuclear mRNA precursor by thyroid hormone

The regulation of a thyroid hormone-responsive gene in rats, designated spot 14, was explored. The expression of this gene in liver is rapidly (less than 10 min) and markedly (greater than 10-fold) altered by the administration of 3,5,3'-triiodo-L-thyronine (T3) to hypothyroid rats (P. Narayan, C. W. Liaw, and H. C. Towle, Proc. Natl. Acad. Sci. USA 81:4687-4691, 1984). To investigate the cellular site at which T3 acts to induce this hepatic mRNA, we made parallel measurements of the relative levels of spot 14 mRNA and nuclear precursor RNA and of the rate of gene transcription after treatments designed to alter the thyroid status of rats. The relative levels of both the mRNA and nuclear precursor were elevated roughly 5- to 6-fold in euthyroid animals and 9- to 12-fold in hyperthyroid animals over those in hypothyroid controls. However, only a small difference of approximately 1.5-fold was detected in the rate of spot 14 gene transcription. After a single injection of T3 into hypothyroid animals, a small and transient rise in the transcription rate was detected at 30 min. However, the levels of spot 14 mRNA and nuclear precursor RNA increased much more dramatically throughout the first 4 h of treatment. In both cases, changes in the rate of gene transcription were not capable of accounting for the alterations observed in mRNA levels. Thus, the major site of spot 14 gene regulation by T3 is at a posttranscriptional level. The proportional changes observed in the nuclear precursor and mRNA levels suggest that the site of control is at the level of stability of the nuclear precursor RNA for spot 14 mRNA.

Mechanism of action of thyroid hormones

Thyroid hormones have ubiquitous effects and influence the function of most organs. The influences that thyroid hormones have on these diverse functions are primarily mediated through binding of T3 and T4 to specific nuclear receptor sites. The nuclear action of T3 results in organ-specific increases and decreases of specific mRNAs, leading to alteration in the level of the corresponding proteins. In addition to the well established nuclear action of T3, effects of thyroid hormone on other sites including cell membranes and mitochondria have been documented.

Thyroid hormone regulation of thyrotropin alpha- and beta-subunit gene transcription

We studied the effect of thyroid hormone on the transcription of the genes for the alpha- and beta-subunits of thyrotropin (TSH) in thyrotropic tumors (IAK 109D and 109F) carried in hypothyroid mice. Gene transcription was measured in isolated nuclei by allowing completion of RNA chains initiated in vivo in the presence of [alpha-32P]UTP and by hybridization of labeled RNA transcripts to filter-bound plasmids containing alpha or TSH-beta cDNA sequences. Treatment of animals carrying tumor IAK 109D with 3,5,3'-triiodo-L-thyronine (T3) (5 micrograms/100 g body weight) for 2 hr reduced TSH-beta gene transcription to less than 10% of control levels, whereas alpha RNA synthesis was reduced to 59% of control. The inhibition of TSH-beta gene activity was maintained after 6 hr of T3 treatment, whereas alpha gene transcription rose slightly to 77% of control. The tumor content of alpha and TSH-beta mRNA, determined by dot blot hybridization with 32P-labeled plasmid probes containing alpha or TSH-beta cDNAs, was unchanged after 2 hr of T3 treatment, and each was reduced by approximately 25% at 6 hr. These untreated tumors contained approximately equal amounts of alpha and TSH-beta mRNA. However, the basal rate of TSH-beta gene transcription was threefold greater than that of alpha gene transcription. Treatment of animals bearing tumor IAK 109F with the same dose of T3 for 30 min did not significantly affect alpha or TSH-beta gene transcription, but at 2 hr alpha and TSH-beta RNA synthesis had decreased to 50% and 10% of control values, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Independent effects of growth hormone releasing factor on growth hormone release and gene transcription

The synthesis and secretion of growth hormone (GH) by the somatotropic cells of the anterior pituitary is under complex hormonal regulation. The hypothalamic peptides, growth hormone releasing factor (GHRF) and somatostatin, respectively stimulate and block GH release. We have shown that treatment of primary cultures of rat anterior pituitary cells with GHRF stimulates transcription of the GH gene, as well as GH release, and that this reflects the response of pituitary cells in vivo. Gick et al. have shown a stimulation of GH mRNA levels in normal pituitary cells in response to GHRF. Because normal pituitary cells are the physiological target for GHRF and because they exhibit both transcriptional and release responses to the hormone, they provide a valuable system with which to examine the important possibility of a link between hormone synthesis and release in secretory cells. Using this system, we demonstrate here that GHRF stimulates GH gene transcription independently of GH release and, conversely, that other agents can stimulate GH release without affecting transcription of the GH gene.

Binding sites for growth hormone releasing factor on rat anterior pituitary cells

Growth hormone releasing factors (GRFs) have been isolated from human pancreatic tumours (hGRF) and rat hypothalamus (rhGRF). The response to GRF at the pituitary level can be modulated by other factors, including glucocorticoids, thyroid hormones, somatostatin and other neuropeptides and somatomedins. Glucocorticoids enhance GRF-induced growth hormone (GH) secretion in primary cultures of rat anterior pituitary cells, and the synthetic glucocorticoid dexamethasone has recently been shown to increase the amounts of GH released in freely moving rats in response to submaximal doses of intravenous GRF. To investigate whether somatotroph sensitivity to GRF is modulated at its receptor level, we have developed a radioreceptor assay using an iodinated analogue of hGRF as radioligand. We report here that the relative binding affinities of rGRF, hGRF and the two analogues are correlated with their in vitro biological potencies. Further, the number of GRF binding sites is drastically decreased in cells deprived of glucocorticoids either in vivo or in vitro.

The first intron of the human growth hormone gene contains a binding site for glucocorticoid receptor

Glucocorticoid receptor (GCR) protein stimulates transcription from a variety of cellular genes. We show here that GCR partially purified from rat liver binds specifically to a site within the first intron of the human growth hormone (hGH) gene, approximately 100 base pairs downstream from the start of hGH transcription. GCR binding is selectively inhibited by methylation of two short, symmetrically arranged clusters of guanine residues within this site. A cloned synthetic 24-base-pair deoxyoligonucleotide containing the predicted GCR binding sequence interacts specifically with GCR. The hGH binding site shares sequence homology with a GCR binding site upstream from the human metallothionein II gene and a subset of GCR binding sites from mouse mammary tumor virus. All of these binding sites for this eukaryotic transcriptional regulatory protein show remarkable similarity in overall geometry to the binding sites for several prokaryotic transcriptional regulatory proteins.

Platelet-derived growth factor induces rapid but transient expression of the c-fos gene and protein

Exposure of quiescent mouse fibroblasts to platelet-derived growth factor induces mRNA from the c-fos proto-oncogene within 10 min followed by synthesis of nuclear c-fos proteins. These are amongst the earliest described nuclear events that follow a mitogenic stimulus.

Control of gene expression by glucocorticoid hormones

Glucocorticoids control the expression of a small number of transcriptionally active genes by increasing or decreasing mRNA concentration. Either effect can result from a transcriptional or a post-transcriptional mechanism. Induction of mouse mammary tumour virus RNA results from a stimulation of transcription initiation and depends on the presence of defined regions in proviral DNA. These regions bind the glucocorticoid receptor and behave functionally as proto-enhancers. Glucocorticoid-inducible genes can retain their sensitivity to the hormone after transfer to a heterologous cell by transfection techniques. Non-inducible genes can become inducible when linked to the promoter region of an inducible gene. The mechanisms by which the receptor-steroid complex stimulates or inhibits transcription or influences mRNA stability are unknown. Receptor binding to nucleic acids appears to be a necessary but not sufficient condition. It is likely that the receptor also interacts with chromatin proteins. This might lead to a catalytic modification of these proteins, resulting in a modulation of gene expression. Development of glucocorticoid-sensitive, biochemically defined, cell-free transcription systems should provide a tool to delineate the molecular determinants of this essential regulatory mechanism.

Infectious and selectable retrovirus containing an inducible rat growth hormone minigene

A growth hormone minigene carrying its natural promoter (237 nucleotides of chromosomal DNA) was stably propagated in a murine retrovirus containing hypoxanthine-guanine phosphoribosyltransferase as a selectable marker. Glucocorticoid and thyroid hormone inducibility was transferred with the growth hormone gene. Recombinant virus with titers of 10(6) per milliliter was recovered. This demonstration that retroviruses can be used to transfer a nonselectable gene under its own regulatory control enlarges the scope of retroviral vectors as potent tools for gene transfer.

Thyroid hormone stimulation of phosphatidylcholine synthesis in cultured fetal rabbit lung

To investigate the mechanism of thyroid hormone action on pulmonary surfactant synthesis, we characterized the effect of triiodothyronine on phosphatidylcholine synthesis in cultured fetal rabbit lung. Since glucocorticoids stimulate surfactant synthesis and reduce the incidence of Respiratory Distress Syndrome in premature infants, we also examined the interaction of triiodothyronine and dexamethasone. The rate of choline incorporation into phosphatidylcholine was determined in organ cultures of rabbit lung maintained in serum-free Waymouth's medium. In 23-d lung cultured for 72 h, the increase in choline incorporation with triiodothyronine alone, dexamethasone alone, and triiodothyronine plus dexamethasone was 50, 62, and 161%, respectively. Both triiodothyronine and dexamethasone also increased incorporation rates with glucose, glycerol, and acetate as precursors, and stimulation with triiodothyronine plus dexamethasone was at least additive. Dexamethasone, but not triiodothyronine, affected distribution of radioactivity from [3H] acetate among phospholipids. Stimulation was first detected 8-12 h after addition of triiodothyronine, and then increased in a linear fashion. With triiodothyronine plus dexamethasone, stimulation was maximal at 48-72 h, and was supra-additive at all times. Exposure of cultured lung to dexamethasone enhanced the subsequent response to triiodothyronine, but not vice versa. When triiodothyronine was removed from cultures, there was no further stimulation and the triiodothyronine effect was partially reversed within 24 h. Half-maximal stimulation of choline incorporation occurred at a triiodothyronine concentration (0.10 nM) very similar to the dissociation constant for triiodothyronine binding to nuclear receptor (0.11 nM). The relative potencies of thyroid hormone analogs for nuclear binding and stimulation of phosphatidylcholine synthesis were also similar: triiodothyroacetic acid greater than triiodothyronine-proprionic acid greater than L-triiodothyronine approximately D-triiodothyronine much greater than thyroxine much greater than 3,5-diethyl-3'-isopropyl-DL-thyronine approximately 3,5-dimethyl-3'-isopropyl-L-thyronine approximately reverse triiodothyronine. The effect of triiodothyronine was blocked by the presence of either actinomycin D or cycloheximide, inhibitors of ribonucleic acid and protein synthesis, respectively. We conclude that triiodothyronine stimulates phosphatidylcholine synthesis by a process involving nuclear receptors and de novo ribonucleic acid and protein synthesis. These findings support the concept that endogenous triiodothyronine has a physiologic role in lung maturation and suggest that a combined antenatal therapy with thyroid hormone and glucocorticoid may be useful for prevention of Respiratory Distress Syndrome in the premature infant.

Dexamethasone is a potent stimulator of growth hormone-releasing factor-induced cyclic AMP accumulation in the adenohypophysis

The stimulatory effect of maximal concentrations of synthetic human pancreatic growth hormone (GH)-releasing factor (GRF)(1-40)NH on cyclic AMP accumulation in rat anterior pituitary cells in culture is 4.5-fold increased following a 48-h preincubation with the potent glucocorticoid dexamethasone while the sensitivity of GRF action is increased by approximately 4-fold. Dexamethasone pretreatment, on the other hand, has no effect on basal cyclic AMP levels but approximately doubles both basal and GRF-induced GH release. The present data suggest that the potent stimulatory effect of glucocorticoids on GH secretion is exerted on the adenylate cyclase system at a step preceding cyclic AMP formation.

Rapid induction of a specific nuclear mRNA precursor by thyroid hormone

Administration of thyroid hormone to the thyroidectomized rat results in a rapid and dramatic increase in the relative amount of hepatic mRNA coding for spot 14--a translational product with an approximate Mr of 17,500 and isoelectric point of 4.9. We have now isolated a cDNA clone containing sequences homologous to this thyroid hormone-responsive mRNA. Two distinct mRNA species that differed by 200 nucleotides in length were found to be capable of hybridizing to the cDNA probe. Both mRNA species were proportionally elevated in relative concentration in rats with increasing plasma levels of thyroid hormone. The earliest change in the levels of mature mRNA occurred at 20 minutes following thyroid hormone treatment of the thyroidectomized rat. Analysis of nuclear RNA revealed a single higher molecular weight species that was homologous to spot 14 mRNA. An increase in the nuclear level of this putative precursor occurred by 10 minutes following thyroid hormone administration, a time preceding the earliest change in mature mRNA. Thus, thyroid hormone appears to act at least in part at a nuclear level in altering the cellular concentration of this mRNA species. The rapidity of this change suggests that it may reflect a direct response to the binding of thyroid hormone to its nuclear receptor.

Specificity of avian leukosis virus-induced hyperlipidemia

Rous-associated virus 7 (RAV-7) is a subgroup C avian leukosis virus which does not transform cells in vitro or carry an oncogene. When injected into 1-day-old hatched chicks, RAV-7 causes a low incidence of lymphoid leukosis after a latent period of several months. In contrast, infection of 10-day-old chicken embryos with RAV-7 leads to a disease syndrome characterized by stunting, obesity, atrophy of the bursa and the thymus, high triglyceride and cholesterol levels, reduced thyroxine levels, and increased insulin levels (Carter et al., Infect. Immun. 39:410-422, 1983; J.K. Carter and R.E. Smith, Infect. Immun. 40:795-805, 1983). Histopathological examination of tissues from affected chicks revealed an accumulation of lipid in the liver and an extensive infiltration of the thyroid and pancreas by lymphoblastoid cells. In the present investigation, the subgroup specificity of this syndrome was investigated. Other subgroup C avian leukosis viruses (transformation-defective B77, transformation-defective Prague C strain of Rous sarcoma virus, and RAV-49) caused stunting, infiltration of the thyroid and pancreas, increased liver weights, decreased thyroxine levels, and increased insulin levels, but they did not cause a uniform, profound increase in triglyceride and cholesterol levels. Avian leukosis viruses of subgroup A [myeloblastosis-associated virus 1 causing osteopetrosis [MAV-1(O)] and RAV-1], subgroup B [MAV-2(O), MAV-2 causing nephroblastoma [MAV-2(N)], and RAV-2], subgroup D (RAV-50), and subgroup F (ring-necked pheasant virus and RAV-61) did not cause a syndrome identical to that induced by RAV-7. All of the viruses examined induced some stunting and a reduction in thyroxine levels which correlated with the stunting. The two subgroup F viruses caused an infiltration of the thyroid which may have been secondary to severe lung involvement. We conclude that the RAV-7 syndrome is unique, particularly in the induction of a hyperlipidemia.

Insulin suppresses growth hormone secretion by rat pituitary cells

The effects of insulin on basal and hydrocortisone-induced growth hormone (GH) secretion were studied in rat pituitary tumor cells (GH3). Cells were grown in monolayer culture and exposed to exogenously added insulin for up to 8 d. Basal GH secretion was inhibited by insulin (0.7 nM) after a 48-h lag period by approximately 50% (P less than 0.01, vs. untreated control cells). The suppression of GH secretion was reversible, as removal of added insulin resulted in return of GH secretion to normal levels after 24 h. Maximal suppression of basal GH secretion was achieved by 0.7 nM insulin, and these effects were prevented by simultaneous exposure of the cells to guinea pig anti-insulin serum (1:2,000). No effects of insulin on cell replication were evident, and glucose concentration in the medium did not differ in control or insulin-treated wells. Insulin (7 nM) significantly suppressed the fivefold hydrocortisone-induced GH stimulation during 5 d of incubation with up to 1,000 nM of the steroid (P less than 0.001). These inhibitory effects were similarly observed in glucose- and pyruvate-free medium, and in the presence of 2-deoxyglucose. Insulin also reversed the suppression of prolactin (PRL) secretion induced by hydrocortisone (1 uM), and actually stimulated basal PRL secretion by over 50%. Insulin did not alter the inhibitory effect of hydrocortisone on GH3 cell proliferation. Although higher doses (13 nM) of insulin-like growth factor (IGF-I) also suppressed basal GH secretion, IGF-I did not alter the GH and PRL secretory changes induced by hydrocortisone. The results show that insulin exerts a direct, specific inhibitory effect on basal and hydrocortisone-induced GH secretion by GH3 cells unrelated to glucose utilization by the cells.

Expression and hormonal regulation of the rat growth hormone gene in transfected mouse L cells

Expression of the rat growth hormone (rGH) gene in the pituitary and in cultured pituitary tumor cells is regulated by glucocorticoid hormones. After co-transfer of cloned DNA containing the rGH gene with the herpes simplex virus (HSV) thymidine kinase (tk) gene into mouse Ltk- cells, rGH gene transcripts were detected in eight of fifteen tk+ cell lines. However, in all eight clones, the predominant rGH gene transcript was only about 0.75 kb, 0.3 kb shorter than pituitary rGH mRNA. The 0.75-kb transcripts, examined from one clone, L-rGH-4, lacked sequences derived from exons 1 and 2 of the rGH gene. Although transcripts larger than 0.75 kb were detected, the normal 2.2-kb rGH gene primary transcript was present only at very low levels. Nuclease mapping studies also failed to reveal transcripts initiated at the normal rGH gene promoter, but instead revealed transcripts with 5' termini arising within intron B of the gene. These data suggest either that transcripts arise from internal promoters within the rGH gene or that a transcript initiated upstream from the normal promoter was processed abnormally. Dexamethasone increased the levels of the 0.75-kb rGH gene transcripts about fourfold in all eight clones expressing rGH mRNA. These data suggest that structural elements important for glucocorticoid-mediated influences on regulation of GH gene expression are contained within the transferred rGH gene fragment and can function even when the normal rGH gene promoter is not used and the pattern of expression is grossly abnormal.

Transcriptional regulation of growth hormone gene expression by growth hormone-releasing factor

Production and release of hormones by the pituitary is known to be under complex hormonal control. Prolactin, for example, is both positively and negatively regulated by steroid and thyroid hormones as well as by peptide hormones. Some hypothalamic releasing factors have been shown to regulate both hormone biosynthesis and hormone release. In the case of growth hormone (GH), glucocorticoids and thyroid hormone stimulate its production, at least in part by stimulating transcription of the GH gene, and somatostatin inhibits and growth hormone-releasing factor (GRF) stimulates its release. So far, however, polypeptide hormone regulation of GH biosynthesis has not been demonstrated. Recently a peptide with GH-releasing activity has been characterized from human pancreatic islet tumours. We report here that pure human pancreatic GRF (hpGRF) stimulates transcription of the GH gene, as well as stimulating GH release.

Glucocorticoids regulate proopiomelanocortin gene expression in vivo at the levels of transcription and secretion

After adrenalectomy, the plasma levels of adrenocorticotropic hormone (corticotropin, ACTH)/endorphin peptides in rats rise dramatically in the first 4 hr while pituitary peptide levels fall sharply. Eight hours after adrenalectomy, plasma levels are near control values again but they then increase continuously over the next 8 days. Proopiomelanocortin (POMC) mRNA levels in the anterior pituitary (quantitated by hybridization with cloned POMC cDNA) increase 2-fold in the first 24 hours, reaching 15- to 20-fold the control level 18 days after adrenalectomy. When dexamethasone is administered to rats 8 days after adrenalectomy, the above events are reversed. Plasma ACTH falls to control levels within 2 hr whereas anterior pituitary POMC mRNA requires 5 days of treatment for return to control levels. The levels of POMC mRNA in the neurointermediate lobe and the hypothalamus are not altered by either treatment. Adrenalectomy increases transcription of the POMC gene in the anterior pituitary approximately 20-fold and halves transcription of the growth hormone gene within 1 hr of operation. Administration of dexamethasone immediately after adrenalectomy suppresses the increase in transcription of the POMC gene and increases the transcription of the growth hormone gene. Transcription of the POMC gene(s) in the neurointermediate lobe is not altered by either of these treatments.

Production of a novel neuropeptide encoded by the calcitonin gene via tissue-specific RNA processing

Alternative processing of the RNA transcribed from the calcitonin gene appears to result in the production of a messenger RNA in neural tissue distinct from that in thyroidal 'C' cells. The thyroid mRNA encodes a precursor to the hormone calcitonin whereas that in neural tissues generates a novel neuropeptide, referred to as calcitonin gene-related peptide (CGRP). The distribution of CGRP-producing cells and pathways in the brain and other tissues suggests functions for the peptide in nociception, ingestive behaviour and modulation of the autonomic and endocrine systems. The approach described here permits the application of recombinant DNA technology to analyses of complex neurobiological systems in the absence of prior structural or biological information.