Increased perinatal remodelling of the pancreas in somatostatin-deficient mice: potential role of transforming growth factor-beta signalling in regulating beta cell growth in early life.

Early postnatal life is a critical period for development of the endocrine pancreas, involving remodelling of islet cells and maturation of secretory responses. Factors that regulate these processes are undefined. Somatostatin-secreting delta-cells are abundant in the developing pancreas and, because somatostatin inhibits growth, the hormone may regulate islet expansion in early life. The aim of this study was to investigate effects of somatostatin-deficiency on proliferation, apoptosis and pancreas expansion in the first 3 weeks of life in mice. Pancreases from control or somatostatin-knockout mice were analysed for beta cell, alpha cell and pancreatic volumes by morphometry, proliferation by BrdU incorporation and apoptosis by TUNEL labelling. Signalling pathways associated with proliferation and apoptosis were studied by immunohistochemistry and Western blotting. Knockout mice grew normally in the first 3 weeks of life, but had high circulating insulin that normalised by day 21. Beta cell, alpha cell and pancreatic volumes were decreased in knockout mice, accompanied by reduced proliferation and increased apoptosis in the pancreas. Decreased growth was not due to impaired Akt signalling, as Akt phosphorylation and nuclear cyclin-D2 increased in the knockout pancreas. Levels of TGF-ß1, a factor implicated in tissue remodelling, together with SMAD phosphorylation through which TGF-ß mediates its effects, were increased in the knockout pancreas. Beta cell expansion was impaired in knockout mice, potentially compensating for increased insulin secretion from islets lacking inhibitory effects of somatostatin, and was associated with increased TGF-ß1 levels. TGF-ß1 may represent an important regulator of beta cell mass in early life.

Abnormalities of the somatotrophic axis in the obese agouti mouse.

OBJECTIVE: Abnormalities of the melanocortin system produce obesity and increased linear growth. While the obesity phenotype is well characterised, the mechanism responsible for increased linear growth is unclear. The somatotrophic axis was studied in the obese agouti (A(y)/a) mouse as a model of a perturbed melanocortin system. DESIGN: Adult obese A(y)/a mice were compared to age- and sex-matched wild-type (WT) controls. Weight and body length (nose-anus) were recorded. Plasma growth hormone (GH), insulin-like growth factor-I (IGFI), insulin and leptin were measured using radioimmunoassay. Since ghrelin is a potent GH secretagogue, plasma ghrelin, stomach ghrelin peptide and stomach ghrelin mRNA expression were studied. Hypothalamic periventricular (PeVN) somatostatin neurones and arcuate (Arc) neuropeptide Y (NPY) neurones inhibit the growth axis, whereas Arc growth hormone-releasing hormone (GHRH) neurones are stimulatory. Therefore, specific hypothalamic expression of somatostatin, NPY and GHRH was measured using quantitative in situ hybridisation. RESULTS: Obese A(y)/a mice were significantly heavier and longer than WT controls. Plasma IGFI concentrations were 30% greater in obese A(y)/a mice. Obese A(y) /a mice were hyperinsulinaemic and hyperleptinaemic, yet plasma ghrelin, and stomach ghrelin peptide and mRNA were significantly reduced. In obese A(y)/a mice, PeVN somatostatin and Arc NPY mRNA expression were reduced by 50% compared to WT controls, whereas Arc GHRH mRNA expression was unchanged. CONCLUSION: Increased body length in adult obese A(y)/a mice may result from reduced Arc NPY and PeVN somatostatin mRNA expression, which in turn, may increase plasma IGFI concentrations and upregulate the somatotrophic axis.

A secreted fluorescent reporter targeted to pituitary growth hormone cells in transgenic mice.

In stable transfection experiments in the GH-producing GC cell line, a construct containing the entire signal peptide and the first 22 residues of human GH linked in frame with enhanced green fluorescent protein (eGFP), produced brightly fluorescent cells with a granular distribution of eGFP. This eGFP reporter was then inserted into a 40-kb cosmid transgene containing the locus control region for the hGH gene and used to generate transgenic mice. Anterior pituitaries from these GH-eGFP transgenic mice showed numerous clusters of strongly fluorescent cells, which were also immunopositive for GH, and which could be isolated and enriched by fluorescence-activated cell sorting. Confocal scanning microscopy of pituitary GH cells from GH-eGFP transgenic mice showed a markedly granular appearance of fluorescence. Immunogold electron microscopy and RIA confirmed that the eGFP product was packaged in the dense cored secretory vesicles of somatotrophs and was secreted in parallel with GH in response to stimulation by GRF. Using eGFP fluorescence, it was possible to identify clusters of GH cells in acute pituitary slices and to observe spontaneous transient rises in their intracellular Ca2+ concentrations after loading with Ca2+ sensitive dyes. This transgenic approach opens the way to direct visualization of spontaneous and secretagogue-induced secretory mechanisms in identified GH cells.

Glucocorticoid regulation of growth hormone (GH) secretagogue-induced growth responses and GH secretagogue receptor expression in the rat.

Synthetic GH-releasing peptides such as GHRP-6 are potent GH secretagogues (GHSs) in several species, but attempts to stimulate growth by continuous GHS exposure have had limited success. GHSs also release ACTH and adrenal steroids. Since glucocorticoid excess is associated with poor linear growth, stimulation of the hypothalamo-pituitary-adrenal (HPA) axis by continuous GHS administration may compromise their growth-promoting effects. We have now examined the effects of continuous GHRP-6 infusion (100 mg/day, s.c. for 14 days) in normal 150-day-old female rats, and in adrenalectomized (Adx) rats with or without dexamethasone (Dex) replacement. Infusion of GHRP-6 did not significantly affect body weight gain compared with excipient-treated controls in either intact rats (controls, 9.0 +/- 1.6 vs GHRP-6, 11.8 +/- 0.9 g) or Adx rats (4.4 +/- 1.5 vs 7.9 +/- 2.7 g). However, GHRP-6 significantly increased weight gain in Adx rats treated with Dex (controls, 3.5 +/- 1.4 vs GHRP-6, 15.4 +/- 1.6 g;P<0.01). Adrenalectomy decreased plasma triglycerides (P<0.01), and Dex treatment increased plasma cholesterol (P<0.001), GHRP-6 treatment did not affect these plasma lipids. Dex treatment also reduced plasma GH-binding protein levels and hepatic GH binding (P<0.05). Pituitary GH content was decreased in Adx rats (P<0.05) but not in Dex-treated Adx rats. Adrenalectomy markedly decreased GHS-receptor mRNA expression in the arcuate (P<0. 001) and ventromedial nuclei (P<0.01), whilst Dex treatment normalized GHS-receptor expression. These results suggest that adrenal steroids are necessary for normal GHS-receptor expression and GHRP-6-induced weight gain, but long-term stimulation of the HPA axis by continuous GHS exposure may be detrimental to the growth response.

A linear hexapeptide somatostatin antagonist blocks somatostatin activity in vitro and influences growth hormone release in rats.

Somatostatin (SRIF) is the main inhibitory peptide regulating growth hormone (GH) secretion. It has been difficult to establish the role of endogenous SRIF release in the absence of pure SRIF antagonists. Although several SRIF antagonists have recently been described, none have been shown to possess in vivo activity in the absence of added SRIF. Here, an SRIF antagonist with no detectable agonist activity has been identified from a synthetic combinatorial hexapeptide library containing 6.4 x 10(7) unique peptides. Each peptide in the library is amino-terminally acetylated and carboxyl-terminally amidated and consists entirely of D-amino acids. A SRIF-responsive yeast growth assay was used as a primary screening tool, and cAMP accumulation, competitive binding, and microphysiometry also were used to confirm and further characterize SRIF antagonist activity. The hexapeptide library was screened in stepwise iterative fashion to identify AC-178,335, a pure SRIF antagonist of the sequence Ac-hfirwf-NH2. This D-hexapeptide bound SRIF receptor type 2 with an affinity constant (Ki) of 172 +/- 12 nM, blocked SRIF inhibition of adenylate cyclase in vitro (IC50 = 5.1 +/- 1.4 microM), and induced GH release when given alone (50 micrograms intravenously) to anesthetized rats with or without pretreatment with a long-acting SRIF agonist.

Effects of growth hormone secretagogues on prolactin release in anesthetized dwarf (dw/dw) rats.

In addition to stimulating GH release, GH secretagogues such as GH-releasing peptide-6 (GHRP-6) stimulate small amounts of ACTH and PRL release. Although the effects on ACTH have recently been studied, there is little information about the effects of GHRP-6 on PRL. We have now studied GHRP-6-induced GH and PRL release and their regulation by estrogen (E2) in anesthetized male and female rats and in GH-deficient dwarf (dw/dw) rats that maintain high pituitary PRL stores and show elevated hypothalamic GH secretagogue receptor expression. Whereas GHRP-6 (0.1-2.5 microg, i.v.) did not induce PRL release in normal male or female rats, significant PRL responses were observed in dw/dw females. These responses were abolished by ovariectomy and could be strongly induced in male dw/dw rats by E2 treatment. These effects could be dissociated from GHRP-6-induced GH release in the same animals, but not from PRL release induced by TRH, which was also abolished by ovariectomy and induced in males by E2 treatment. However, the effects of GHRP-6 on PRL were unlikely to be mediated by TRH because in the same animals, TSH levels were unaffected by GHRP-6 whereas they were increased by TRH. The increased PRL response could reflect an increase in GH secretagogue receptor expression that was observed in the arcuate and ventromedial nuclei of E2-treated rats. Our results suggest that the minimal PRL-releasing activity of GHRP-6 in normal rats becomes prominent in GH-deficient female dw/dw rats and is probably exerted directly at the pituitary; these GHRP-6 actions may be modulated by E2 at both hypothalamic and pituitary sites.

Intrahypothalamic growth hormone feedback: from dwarfism to acromegaly in the rat.

Two different dwarf rat models with primary (dw/dw, DW) or secondary (transgenic growth retarded, WF/Tgr) GH deficiency and contrasting hypothalamic GH-releasing hormone (GHRH) and somatostatin (SRIH) expression were implanted sc with GC cells. These form encapsulated rat GH-secreting tumors that maintain high plasma rat GH levels for several weeks. In both strains, GC cell tumors stimulated growth and raised GHBP levels, without affecting pituitary GH content. In DW rats, GC cell implants increased SRIH expression in the periventricular nucleus (PeV), but not in the arcuate nucleus (ARC), whereas their high GHRH expression in ARC was decreased by GC cells. In contrast, GC cell implants in WF/Tgr rats had little effect on the already high SRIH expression in PeV or low GHRH expression in ARC, although they reduced SRIH expression in ARC. GC cell implants also reduced GH receptor expression in both ARC and PeV in the WF/Tgr dwarves. Thus, chronic GH overexposure stimulates rapid growth in both dwarf strains, but has differential hypothalamic effects in these models. This experimental approach now makes it possible to study the effects of pathophysiological concentrations of GH ranging from dwarfism to acromegaly in the same animal model.

Effects of growth hormone secretagogues in the transgenic growth-retarded (Tgr) rat.

Exogenous GH inhibits endogenous GH release by hypothalamic feedback. We have recently exploited this to generate transgenic growth-retarded (Tgr) rats, in which human GH is expressed in the hypothalamus, under the control of the rat GRF gene promoter. These rats show reduced pituitary size, GH deficiency, and dominant dwarfism, but are large enough for serial blood sampling studies to examine their spontaneous GH secretion and responses to GRF, somatostatin, and GH-releasing peptide-6 (GHRP-6). Like their normal wild-type littermates, Tgr rats show a sexually dimorphic pattern of GH secretion; males secrete GH in 3-h episodes, whereas females exhibit a more continuous irregular output, with higher baseline GH levels. In anesthetized male Tgr rats, the GH responses to GRF or GHRP-6 were markedly reduced compared with those of their nontransgenic littermates, but the differences were smaller in females. Despite the reduction in pituitary GH, peak plasma GH responses to serial GRF injections in conscious Tgr males or intermittent somatostatin infusions in conscious Tgr females were indistinguishable from the responses in their wild-type littermates. Furthermore, 7-day iv infusions of GRF (12.5-100 micrograms/day), given either continuously or as a pulsatile infusion stimulated growth in Tgr rats, as did pulsatile infusions of GHRP-6. Thus, despite their pituitary GH deficiency and dwarfism, Tgr rats maintain a sexually dimorphic pattern of GH release and can produce large GH secretory responses to exogenous secretagogues. They represent the first genetic model of GH deficiency in the rat in which dwarfism can be corrected by treatment with exogenous GH secretagogues.

Pituitary growth hormone-releasing factor receptor expression in normal and dwarf rats.

Growth hormone-releasing factor (GRF) regulates GH release and somatotrope proliferation via a specific G-protein-coupled receptor. Little is known about the endocrine factors that regulate the expression of the GRF receptor (GRF-R) in the pituitary gland. We have developed a sensitive solution hybridization/RNAse protection assay for GRF-R mRNA in rat pituitary extracts. GRF-R transcripts were readily detectable in similar amounts in normal male and female rats, but were markedly reduced in extracts from age-matched growth hormone (GH)-deficient dwarf (dw) rats of either sex. The reduced GRF-R expression would appear to reflect somatotrope hypoplasia rather than GH deficiency per se since a similar reduction in GRF-R expression was seen in a transgenic model of dominant dwarfism, whereas GRF-R expression was significantly elevated in rats with GH deficiency induced by hypothyroidism. We were unable to demonstrate significant effects on GRF-R expression with infusions of human GH (hGH) or insulin-like growth factor 1, which stimulate growth in dw rats, but dexamethasone treatment induced a significant, time-related increase in GRF-R mRNA levels. We conclude that this assay can usefully quantify pituitary GRF-R expression in normal rats, and its reduction in two different strains of mutant dwarf rats with somatotrope hypoplasia.

Differential regulation of the growth hormone receptor gene: effects of dexamethasone and estradiol.

GH receptor (GHR) expression differs during development between central and peripheral tissues. Peripheral GHR expression is known to be sensitive to gonadal and adrenal steroids, but little is known about their effects on GHR in the central nervous system. We have now studied the effects of estradiol (E2) or dexamethasone on GHR expression in rat arcuate nucleus (ARC) and hippocampus, using quantitative in situ hybridization. Dexamethasone, which strongly down-regulates hepatic GHR expression, had no effect on central GHR transcript abundance, whereas E2 treatment, which stimulates hepatic GHR expression, significantly reduced ARC GHR messenger RNA (mRNA) levels. E2 also increased somatostatin (SS) expression significantly in both ARC and periventricular nuclei but did not reduce ARC GH-releasing hormone (GHRH) mRNA levels. Ovariectomy stimulated GHR and GHRH mRNA levels in the ARC, whereas it lowered ARC SS expression. E2 replacement in ovariectomized animals restored GHRH and SS mRNA levels to control values. Hippocampal GHR mRNA transcripts showed the same response to these endocrine manipulations as seen in the ARC. The induction of hepatic GHR expression by E2 is known to involve the transcription of an alternate 5' untranslated first exon, GHR1. This was readily detectable in the liver using a specific GHR1 probe but could not be detected in any CNS area. Our results show that GHR expression in the CNS is sensitive to regulation by peripheral steroids but that CNS and hepatic expression of GHR is differentially regulated by the same treatments.

Dominant dwarfism in transgenic rats by targeting human growth hormone (GH) expression to hypothalamic GH-releasing factor neurons.

Expression of human growth hormone (hGH) was targeted to growth hormone-releasing (GRF) neurons in the hypothalamus of transgenic rats. This induced dominant dwarfism by local feedback inhibition of GRF. One line, bearing a single copy of a GRF-hGH transgene, has been characterized in detail, and has been termed Tgr (for Transgenic growth-retarded). hGH was detected by immunocytochemistry in the brain, restricted to the median eminence of the hypothalamus. Low levels were also detected in the anterior pituitary gland by radioimmunoassay. Transgene expression in these sites was confirmed by RT-PCR. Tgr rats had reduced hypothalamic GRF and mRNA, in contrast to the increased GRF expression which accompanies GH deficiency in other dwarf rats. Endogenous GH mRNA, GH content, pituitary size and somatotroph cell number were also reduced significantly in Tgr rats. Pituitary adrenocorticotrophic hormone (ACTH) and thyroid-stimulating hormone (TSH) levels were normal, but prolactin content, mRNA levels and lactotroph cell numbers were also slightly reduced, probably due to feedback inhibition of prolactin by the lactogenic properties of the hGH transgene. This is the first dominant dwarf rat strain to be reported and will provide a valuable model for evaluating the effects of transgene expression on endogenous GH secretion, as well as the use of GH secretagogues for the treatment of dwarfism.

The obese growth hormone (GH)-deficient dwarf rat: body fat responses to patterned delivery of GH and insulin-like growth factor-I.

We describe a new animal model of obesity and GH deficiency and report the effects on body fat of administering (GH) and insulin-like growth factor (IGF-I) in the model. Female GH-deficient dwarf rats fed a high-fat diet became obese and insulin-resistant compared with chow-fed controls. They were treated with recombinant human GH (rhGH 100-500 micrograms/day, s.c. for 14 days) by daily injection or minipump infusion with or without rhIGF-I (200 micrograms/day, sc infusion). Injections of rhGH increased body weight; infusions of rhGH caused weight loss. RhIGF-I by itself, or rhIGF-I plus GH injections had little effect, whereas rhGH infusions plus rhIGF-I caused a weight loss equivalent to the weight gained during the high-fat feeding and a decrease in fat pad weight. For some responses (serum IGF-1 and GHBP), the obese rats were GH resistant. Fat was lost from the internal fat pads when obese rats were returned to a chow diet, and injections of rhGH surprisingly attenuated this loss of fat. In obese dwarf rats, the lipolytic effects of rhGH are dose-regime dependent. By itself IGF-I is not insulin-like, but in the presence of GH it has antiinsulin actions causing a powerful net lipolysis. If GH plus IGF-I have similar effects in humans they may be useful for reducing body fat.

The human growth hormone (hGH) antagonist G120RhGH does not antagonize GH in the rat, but has paradoxical agonist activity, probably via the prolactin receptor.

Human GH (hGH) acts by dimerizing two hGH receptors that bind to different sites in hGH. G120RhGH, an analog mutated in the second binding site to prevent receptor dimerization, acts as an antagonist in vitro. We have now tested the activity of this analog in vivo in rats with low or absent endogenous GH secretion. Surprisingly, treatment with G120RhGH failed to antagonize the effects of infusions or injections of hGH in hypophysectomized (Hx) rats and had little effect on hepatic GH-sensitive CYP2C transcripts in GH-deficient dwarf (dw) rats. Paradoxically, G120RhGH stimulated skeletal growth when infused into Hx rats; a pulsatile iv infusion was more effective than a continuous pattern. Coinfusion of G120RhGH with hGH produced an additive effect on growth. In addition, continuous, but not pulsatile, G120RhGH infusion elevated hepatic 2C12 messenger RNA (mRNA) expression and reduced 2C11 mRNA expression in Hx rats. The direct effects of G120RhGH on hepatic CYP2C transcripts were confirmed in cultured hepatocytes in vitro, which also revealed a significant action of PRL in elevating 2C12 mRNA expression. Binding studies revealed that G120RhGH bound preferentially to hepatic PRL receptors, as [125I]G120hGH was completely displaced by ovine PRL but was unaffected by bGH, a specific GH receptor ligand. The weak growth-promoting effects of G120RhGH were similar to those induced by recombinant hPRL in Hx rats. Our results show that G120RhGH is a poor in vivo GH antagonist in the rat, but shows a paradoxical agonist effect, probably mediated by PRL receptors in this species.

Steroid regulation of growth hormone (GH) receptor and GH-binding protein messenger ribonucleic acids in the rat.

In the rat, the GH receptor (GHR) and the GH-binding protein (GHBP), which arise from alternative splicing of the same gene, show a sexually dimorphic and GH-dependent expression pattern. Multiple alternative 5'-untranslated regions (UTRs) are present in GHR and GHBP transcripts in the rat, one of which, GHR1, has recently been shown to be liver specific and found at higher levels in females. We have measured the hepatic GHR1, GHR, and GHBP transcript levels, by RNase protection and solution hybridization assay, in animals with differing hormonal status, in which hepatic GHR binding and plasma GHBP have been previously assayed. Estradiol (E2) induced GHR1 in males, whereas ovariectomy or the antiestrogen tamoxifen reduced GHR1 expression in females. The induction of GHR1 by E2 was GH dependent, being lower in GH-deficient dwarf rats and absent in hypophysectomized rats, paralleling previous measurements of plasma GHBP and hepatic GHR binding in these animals. Significant changes in GHR1 could explain the trends seen in the same extracts when coding region probes were used. Short-term adrenalectomy had no effect on GHR and GHBP expression, but dexamethasone markedly reduced both protein and messenger RNA (mRNA) levels. Corticosterone treatment had no effect alone but reduced the E2-induced increase in GHR1 levels, whereas methylprednisolone administered orally reduced hepatic GH binding, plasma GHBP, and GHR1 mRNA levels. Thus, 5'-UTRs, encoded by different first exons, are involved in the regulation of hepatic GHR and GHBP expression and need to be considered when comparing effects of hormonal manipulation on the mRNA transcripts and protein products of the GHR gene. Previous studies have found discrepancies between levels of protein expression and mRNA transcripts measured only with coding region probes. Our results suggest that posttranscriptional differences related to 5'-UTR heterogeneity in the GHR gene explain some of these discrepancies.

Growth responses to patterned GH delivery.

We have investigated the effects of different patterns of administration of recombinant human growth hormone (rhGH) on weight gain, organ growth, serum GH binding protein (GHBP) and insulin-like growth factor-l (IGF-1) levels in a series of studies using hypophysectomized (Hx) or GH-deficient dwarf (dw/dw) rats. Animals were given rhGH either by subcutaneous (s.c.) injections (1 or 2 per day) or s.c. infusions and rhlGF-1 (2 mg/kg/day) by s.c. infusion. In Hx rats, all rhGH regimes increased body weight, tibial epiphyseal plate width, and organ weights in a dose-related manner. Dwarf rats showed a smaller growth response to rhGH than Hx rats, whereas rhGH induced greater elevations in serum GHBP in drarf rats. Growth responses depended on the pattern of rhGH administration (twice daily injections > continuous infusions > daily injections). The shape of the body growth curves also differed; rhGH injections increased weight gain linearly, whereas infusions gave an initial rapid weight gain which slowed with time (a curvilinear response). For both regimens, tibial epiphyseal plate width increased linearly with rhGH dose but infusions were 5-fold more potent than daily injections. Spleen and thymus weights were markedly increased by rhGH and were also affected by the pattern of GH exposure. At 5 mg rhGH/kg/day, thymus weights were 390±35 mg for injectionsvs. 613 ± 34 mg for infusions (P<0.001) compared with 248 ± 16 mg in vehicle-treated Hx controls. Infusions of rhlGF-1 also stimulated specific organ growth but caused less weight gain. RhlGF-1 additively increased the weight gain caused by rhGH injections but not by rhGH infusions. Circulating IGF-1 and GHBP levels were increased in a dose-dependent manner by rhGH infusion, whereas daily injections were ineffective. Thus, differential organ growth could be related to the higher serum IGF-1 concentrations induced by continuous rhGH administration. These studies show that whole body growth is best maintained by intermittent rhGH exposure, whereas, paradoxically, differential organ growth is most pronounced with continuous rhGH administration.

Growth hormone binding protein in the rat: effects of gonadal steroids.

In normal rats, females have higher circulating GH-binding protein (GHBP) levels than males, whereas in the GH-deficient dwarf (Dw) rat, there is no sexual dimorphism in plasma GHBP, suggesting that GH secretion may be involved in this difference. In order to study the relationship between gonadal steroids and GH on GHBP and GH receptor regulation, the levels of plasma GHBP, hepatic bovine GH, and human GH (hGH) binding as well as GHBP and GH receptor messenger RNA (mRNA) have now been studied in normal, Dw, hypophysectomized (Hx), or ovariectomized (Ovx) rats, subjected to different GH and gonadal steroid exposure. In normal male rats, estradiol (E2, 12.5-25 micrograms/day for 1 or 2 weeks) markedly increased plasma GHBP and hepatic hGH, and bGH binding. These effects of E2 were diminished in Dw rats, absent in Hx rats, but restored in Hx rats given exogenous hGH. Plasma GHBP rose in female rats given E2, and fell in females given the anti-estrogen tamoxifen. Ovx animals had lower plasma GHBP and hepatic GH binding which was reversed by E2, but not testosterone treatment. Continuous hGH infusions in Ovx rats restored hepatic GH binding, and increased plasma GHBP. In Dw males, hGH increased plasma GHBP and hepatic GH binding, whereas testosterone had no effect on GHBP or GH receptors and did not affect their up-regulation by hGH. Hepatic levels of GHBP-, and GH receptor mRNA transcripts showed the same trends in response to steroid or GH treatment, but the differences were rarely significant, except in Ovx animals which had higher GHBP mRNA transcripts after GH or E2 treatment. Thus E2 and GH increase both plasma GHBP and hepatic GH receptor binding. GH up-regulates GHBP in the absence of E2, whereas E2 treatment does not raise GHBP in the absence of GH. Whereas some of the effects of estrogen could be mediated via alterations in GH secretion, estrogen may also directly influence GHBP production at the liver, but only in the presence of GH.

Growth hormone receptor regulation in growth hormone-deficient dwarf rats.

In the rat, many actions of GH depend upon the sexually dimorphic pattern of exposure to GH. Hepatic human GH (hGH) receptor binding differs between the sexes and is sensitive to GH deficiency, but this has mostly been studied in acutely hypophysectomized rats, which lack all pituitary hormones. We have used a strain of GH-deficient dwarf (Dw) rats to determine whether chronic GH deficiency alters the normal developmental pattern and sexually dimorphic expression of hepatic GH receptors. Adult female Dw rats had lower levels of 125I-labelled hGH binding (reflecting predominantly lactogenic receptors) than their normal counterparts whereas there was no difference between adult Dw and normal males; binding capacity increased from 25 days of age, becoming sexually dimorphic from 40 days to adulthood in both strains (% specific binding/mg protein: normal males 1.6 +/- 0.3, normal females 13.2 +/- 1.1, Dw males 2.1 +/- 0.4, Dw females 10.0 +/- 0.6). In contrast, hepatic 125I-labelled bovine GH (bGH) binding (somatogenic receptors) was much lower, and similar in both Dw and normal animals. A sex difference in 125I-labelled bGH binding was only seen in adult animals, and was considerably less marked in Dw rats compared with normal animals (normal males 1.3 +/- 0.1, normal females 2.5 +/- 0.2, Dw males 1.9 +/- 0.2, Dw females 2.4 +/- 0.2%/mg protein). Continuous hGH infusion stimulated growth in female Dw rats, and raised somatogenic and lactogenic GH binding (3.2 +/- 0.4 and 19.6 +/- 2.5%/mg protein) compared with sham-infused controls (2.4 +/- 0.2 and 7.9 +/- 0.6%/mg protein).(ABSTRACT TRUNCATED AT 250 WORDS)

Growth hormone (GH)-binding protein in normal and GH-deficient dwarf rats.

There are GH-binding proteins (GHBPs) present in the blood of many species, and these correspond to the extracellular GH-binding domain of the GH receptor. In the rat, GHBP arises by alternative splicing of the GH receptor mRNA, but little is known of the physiological role of circulating GHBP, or its relationship with episodic GH secretion. We have developed a sensitive radioimmunoassay based on recombinant GHBP, and have measured rat GHBP levels in small samples of plasma from normal and GH-deficient dwarf rats. In normal adult rats, GHBP levels were two- to threefold higher in females than in males (16.6 +/- 0.8 vs 6.4 +/- 0.4 micrograms/l, P < 0.001), but this sex difference was not seen in dwarf rats. A continuous infusion of human GH in dwarf males raised plasma GHBP to 23.5 +/- 3.5 micrograms/l compared with 6.7 +/- 0.5 micrograms/l in sham-infused animals, whereas suppression of GH by continuous infusion of a long-acting somatostatin analogue in female dwarf rats had no effect on GHBP. In anaesthetized rats, large changes in plasma GH caused by i.v. administration of rat GH, somatostatin or GH-releasing factor did not affect GHBP acutely. Both GH and GHBP were also measured in serial blood samples from conscious normal and dwarf rats. A sexually dimorphic GH secretory pattern was observed in both strains. Males showed peaks and troughs of GH every 3 h varying over a 100-fold range, whereas females exhibited more continuous GH secretion. Despite the large fluctuations in endogenous GH, GHBP levels remained relatively constant in individual normal or dwarf males, as well as in females of both strains, and there was no significant correlation between GH and GHBP either in individual rats or as a group. Our results suggest that GHBP is GH-dependent in the longer term, and that the higher GHBP levels in female rats require their continuous GH secretory pattern. However, plasma GHBP levels remain stable and are not affected by acute changes in endogenous or exogenous GH.

Growth hormone (GH) binding protein and GH interactions in vivo in the guinea pig.

GH binding proteins (GHBPs) are present in the blood of several species and by complexing with circulating GH may alter its clearance and distribution. The actions of human GHBP (hGHBP) have hitherto been studied indirectly via their effects on the clearance of hGH. In the present experiments, recombinant preparations of hGHBP and a shorter variant lacking exon 3 (delta 3hGHBP) were tested in vivo, either alone or after preincubation with recombinant hGH, recombinant 20K methionyl-hGH, or rat GH. Multiple serial blood sampling was performed in conscious chronically cannulated guinea pigs and both GHBP and GH variants monitored in the same samples by specific RIAs, unaffected by endogenous activities in this species. hGHBP, delta 3hGHBP, hGH, and 20K met-hGH (10-40 micrograms) were all cleared rapidly when injected alone (t1/2 = 11-20 min). However, when the same amounts of hGHBP and hGH were incubated together for 60 min and then injected, the clearance of both proteins was greatly prolonged (t1/2 hGHBP = 75-94 min, hGH = 80-137 min). Similar results were obtained for hGH complexed with delta 3hGHBP, and over a range of GH:GHBP ratios from 0.3:1 to 4:1. The effect was specific for 22K hGH; neither rat GH nor 20K met-hGH altered hGHBP clearance, nor were their clearances slowed by preincubation with hGHBP. Rapid complex formation also could be demonstrated in vivo. Injections of hGH 30 min after delta 3hGHBP, or vice versa, altered the established plasma disappearance curve of hGHBP, and hGH clearance was slowed in parallel. During a continuous infusion of hGH to steady state, injections of delta 3hGHBP increased plateau hGH concentrations by forming a delta 3hGHBP/hGH complex in the circulation. These experiments show that the conscious chronically cannulated guinea pig provides a useful model in which the in vivo interaction between hGHBP and hGH can be studied dynamically. The results imply that the location, extent, and rate of complex formation between GHBP and GH may be an important determinant of the passage of GH between the intra- and extravascular compartments, which could affect the pattern of tissue exposure to GH.