Prolactin (PRL)-releasing peptide stimulates PRL secretion from human fetal pituitary cultures and growth hormone release from cultured pituitary adenomas

The hypothalamic peptide PRL-releasing peptide (PrRP) has recently been cloned and identified as a ligand of an orphan pituitary receptor that stimulates in vitro PRL secretion. PrRP also induces PRL release in rats in vivo, especially in normal cycling females. However, no information on the effects of PrRP in the human is available. To elucidate the role of PrRP in regulating human anterior pituitary hormones, we used human PrRP-31 in primary cultures of human pituitary tissues, including fetal (20--27 weeks gestation) and normal adult pituitaries, as well as PRL- and GH-secreting adenomas. PrRP increased PRL secretion from human fetal pituitary cultures in a dose-dependent manner by up to 35% (maximal effect achieved with 10 nM), whereas TRH was slightly more potent for PRL release. Coincubation with estradiol resulted in enhanced fetal PRL response to PrRP, and GH release was only increased in the presence of estradiol. Although PRL secretion from PRL-cell adenomas was not affected by PrRP, PrRP induced PRL release from cultures of a GH-cell adenoma that cosecreted PRL. PrRP enhanced GH release in several GH-secreting adenomas studied by 25--27%, including GH stimulation in a mixed PRL-GH-cell tumor. These results show for the first time direct in vitro effects of PrRP-31 on human pituitary cells. PrRP is less potent than TRH in releasing PRL from human fetal lactotrophs and is unable to release PRL from PRL-cell adenomas in culture, but stimulated GH from several somatotroph adenomas. Thus, PrRP may participate in regulating GH, in addition to PRL, in the human pituitary.     

Caffeine stimulates growth hormone secretion by cultured rat pituitary cells

To study the effect of caffeine on growth hormone secretion a culture system of dispersed rat anterior pituitary cells was employed. The cells were incubated overnight in medium 199 containing 10(-5) to 10(-1) M caffeine. The medium was then collected and assayed for rat growth hormone content. A dose dependent stimulatory effect of caffeine on growth hormone secretion into the culture medium was observed. It is concluded that caffeine, like other xanthine phosphodiesterase inhibitors stimulates growth hormone secretion by a direct effect on pituitary cells.     

Cortistatin inhibits growth hormone release from human fetal and adenoma pituitary cells and prolactin secretion from cultured prolactinomas

CONTEXT: Cortistatin (CST) is a neuropeptide that shares high homology with somatostatin and binds with high affinity to all somatostatin receptor (SSTR) subtypes. Many of its endocrine and biological activities overlap with those of somatostatin. OBJECTIVE/DESIGN: The objective of the study was to assess the direct in vitro effects of CST on human pituitary hormone secretion. SETTING: This study was performed in the endocrine laboratory of a tertiary academic medical center. MATERIALS: Primary cell cultures of human fetal (21-25 wk gestation) pituitary tissues and cultured hormone-secreting adenoma cells were used in this study. INTERVENTIONS: Cell cultures were incubated with CST-14 or CST-17, somatostatin, GHRH, SSTR analogs, and ghrelin analogs, and hormone secretion was analyzed. OUTCOME MEASURES: GH and prolactin (PRL) medium concentrations were tested by hormone assay, and SSTR mRNA was tested by RT-PCR. RESULTS: CST-14 (10 nm) inhibited GH secretion by up to 65% in all fetal pituitary specimens after 4-h incubation (P < 0.05). CST-14 or CST-17 (10 nm) inhibited basal GH secretion in six of the 13 GH-cell adenomas and two of the three GH-PRL mixed adenomas. CST-17 (100 nm) suppressed the GH response to GHRH and ghrelin analog (10 nm each) by 30-50% in adenomas (P < 0.05). Three PRL-adenomas treated with CST-17 (10 nm) showed a 20-40% inhibition of PRL release (P < 0.05), whereas in three others no suppression or mild response was achieved at this concentration. A comparable inhibition of PRL secretion was obtained with SSTR5-selective analog but significantly less with SSTR2-preferential compounds. RT-PCR revealed the expression of both SSTR2 and SSTR5 in all GH-cell and mixed adenomas studied and all PRL-secreting adenomas studied, except for two of the CST-resistant prolactinomas, in which SSTR5 was absent. CONCLUSIONS: This is the first report of in vitro CST suppression of human GH and PRL in cultured pituitary tissues. The regulation of PRL release from cultured adenomas appears to be primarily mediated by SSTR5.     

Regulation of growth hormone release from cultured human pituitary adenomas by somatomedins and insulin

GH secretion is stimulated by hypothalamic GH-releasing factor (GHRH) and inhibited by somatostatin. Since GH induces the production of insulin-like growth factors (IGF) in liver and other tissues, it is of interest to learn whether IGF alters GH release through long loop feedback inhibition. Pituitary adenomas which had been removed from six acromegalic patients were processed for dispersed cell cultures and/or cell membrane preparations. Binding studies using 125I-labeled IGF-I, IGF-II, and insulin revealed specific hormone binding for each ligand to cell membranes derived from four somatotropinomas. A partially purified somatomedin preparation inhibited basal and/or GHRH-stimulated GH release from cultured pituitary cells derived from three of four adenomas; there was no effect of somatomedin in one tumor. In a single tumor, insulin also partially inhibited GHRH-stimulated GH release. Additionally, in one nonadenomatous pituitary removed from a patient with diabetes mellitus, insulin and somatomedin inhibited GHRH-stimulated GH release, and insulin inhibited basal GH secretion. These results indicate that specific cell membrane receptors for somatomedin peptides and insulin may be found on cell membranes from GH-secreting tumors, and that somatomedins and insulin can inhibit GH release in cultured human somatotropinoma cells. Thus, these data suggest that somatomedins may exert feedback inhibition of GH secretion in some patients with acromegaly.     

Ectopic growth hormone releasing factor stimulates growth hormone release from human somatotroph adenomas in vitro

The effects of an ectopic growth hormone releasing factor (E-GHRF) from a carcinoid tumor were studied in three human GH-secreting pituitary adenomas in monolayer culture. GH release by the adenomas was stimulated by 55 +/- 10%, 57 +/- 8%, and 59 +/- 17% during a 4 h exposure to E-GHRF at concentrations of 3, 0.8, and 24 Units/ml, respectively. Two of the tumors also secreted prolactin (PRL) and in one, E-GHRF (0.8 Units/ml) stimulated PRL release by 48 +/- 8%. GH and PRL release were both suppressed by somatostatin (10(-8)M) in two of the adenomas. These results 1) demonstrate for the first time E-GHRF stimulation of GH release from human pituitary tissue in vitro, 2) support an etiologic role for E-GHRF in the development of acromegaly, and 3) indicate that GH-secreting adenomas retain the capability of responding to GHRF.     

Interferon-alpha-2a is a potent inhibitor of hormone secretion by cultured human pituitary adenomas.

Interferon-alpha (IFN alpha) may exert direct inhibitory effects on cell proliferation and on the production of different peptide hormones. We investigated the effect of IFN alpha on hormone production by 15 GH-secreting pituitary adenomas, 4 clinically nonfunctioning or gonadotroph pituitary adenomas, and 4 prolactinomas in vitro. In the GH-secreting pituitary adenoma cultures, a short term (72-h) incubation with IFN alpha (50-100 U/mL) significantly inhibited GH secretion in 3 of 7 cases and PRL secretion in 6 of 7 cultures. During prolonged incubation (14 days) with IFN alpha, GH and/or PRL secretion was significantly inhibited in 7 of 8 cultures (GH, 17-78% inhibition; PRL, 39-88% inhibition). In the clinically nonfunctioning or gonadotroph cultures, incubation with IFN alpha resulted in inhibition of the secretion of gonadotropins and/or alpha-subunit in all cases (27-62%), whereas in the prolactinoma cultures PRL secretion was inhibited by IFN alpha in all cases (37-76%). The effect of IFN alpha was additive to the inhibitory effects of the dopamine agonist bromocriptine (10 nmol/L) or the somatostatin analog octreotide (10 nmol/L). The inhibition of hormone secretion by IFN alpha was accompanied by inhibition of the intracellular hormone concentrations. The effect of IFN alpha was dose dependent, with an IC50 for inhibition of hormone secretion of 2.3 +/- 0.3 U/mL (n = 5), which is relatively low compared with the concentrations that are reached in patients treated with IFN alpha for various malignancies. In conclusion, the potent antihormonal effect of IFN alpha on cultured pituitary adenomas suggests that this drug might be of benefit in the treatment of selected patients with secreting pituitary adenomas. As treatment with IFN alpha is associated with considerable adverse reactions, studies with this drug should only be considered in inoperable, invasive aggressive, and dopamine agonist- and/or somatostatin analog-resistant functioning pituitary macroadenomas.     

Somatostatin receptors in human growth hormone and prolactin-secreting pituitary adenomas

[125I-Tyr]Somatostatin [( 125I-Tyr]SRIH) binding was found in 11 GH-secreting pituitary adenomas [Kd = 0.46 +/- 0.15 (+/- SE) nM; maximum binding, 165 +/- 35 fmol/mg protein). This binding was specific, since it was displaced by somatostatin-14 (SRIH-14), N-Tyr-SRIH-14, and SRIH-28. In contrast, a number of peptides and drugs not structurally related to SRIH, such as bombesin, dopamine, LHRH, met-enkephalin, naloxone, neurotensin, secretin, substance P, TRH, or vasoactive intestinal peptide, did not affect [125I-Tyr]SRIH binding. [125I-Tyr]SRIH specific binding also was found in PRL-secreting pituitary adenomas. The kinetic characteristics of the specific binding were similar to those of GH-secreting adenomas. However, maximal binding was one quarter that of GH-secreting adenomas (37 +/- 9 fmol/mg protein). In contrast, nonsecreting (chromophobe) tumors were devoid of any specific binding. Finally, in acromegaly, the density of [125I-Tyr]SRIH-binding sites in the adenomas was negatively correlated with plasma GH levels before surgery (r = -0.80). This suggests that somatostatinergic control is involved in GH secretion in acromegalic patients.     

Tumor necrosis factor-alpha inhibits growth hormone secretion from cultured anterior pituitary cells

Anabolic proteins such as pituitary GH enhance the function of several immune cell types. The converse could also exist, that is communication from the immune cells to GH-producing somatotrophs. To test this hypothesis, tumor necrosis factor-alpha (TNF-alpha), a product of activated macrophages, was exposed to cultured rat pituitary cells, and GH release was monitored. TNF alpha inhibited basal and GH-releasing hormone-stimulated GH accumulation, with IC50 values of 170 U/ml (5.2 ng/ml) and 50 U/ml (1.5 ng/ml), respectively. This inhibition was first measured after 6 h of TNF alpha treatment, continued for at least 3 days, and was reversible. A number of measurements (e.g. trypan blue exclusion, chromium release, and GH cell content) yielded no signs of cytotoxicity to explain the inhibition. We conclude that TNF alpha can reduce basal and stimulated pituitary GH release in vitro.     

DNA and S-phase representation in human growth hormone producing pituitary adenomas

The DNA contents of 33 pituitary adenomas from patients with acromegaly were analysed with flow-cytofluorometry. Degrees of ploidy and the proliferation rate, expressed as percentage of cells in S-phase, were determined. The aim was to compare these morphological and functional tumour properties with clinical and laboratory parameters to establish a possible relation and to further elucidate the characteristics of these tumours. In 15 tumours (45%) diploid DNA pattern were found, while 18 (55%) showed varying degrees of aneuploidy. The frequency of cells in S-phase showed wide variations and were equally distributed in diploid and aneuploid tumours. Duration of symptoms, age at diagnosis, preoperative growth hormone (GH)- and prolactin (Prl)-levels, tumour size and grade of invasive tumour growth as determined by radiological estimations, did not correlate to ploidy or grade of proliferation. The lack of correlation between DNA pattern and proliferation rate in relation to clinical, laboratory and radiological parameters in tumours causing acromegaly contrasts to the documented relation between the degree of ploidy, cells in proliferation and grade of malignancy reported in tumours of other sites. The 55% aneuploid GH producing tumours indicate a certain malignant transformation. The high frequency of cells in S-phase in several GH secreting tumours completes the malignant morphological and functional cell properties. The benign character of tumours causing acromegaly is therefore in contrast to these findings. The lack of clinical significance of the DNA pattern and the frequency of cells in proliferation in GH producing tumours and the benign character despite malignant cell properties in most of these tumours are difficult to explain.(ABSTRACT TRUNCATED AT 250 WORDS)     

Epidermal growth factor stimulates secretion of rat pituitary luteinizing hormone in vitro

The effects of epidermal growth factor (EGF) on pituitary luteinizing hormone (LH) release and on the releases induced by oestradiol (E2) and LH-releasing hormone (LRH) were examined in a sequential double chamber perifusion system. In this system the mediobasal hypothalami (MBH) and/or pituitaries excised from normally cycling female rats in dioestrus were perifused with test media. Perifusion with EGF at 1 ng/ml for 30 min induced significant release (80-100% increase, P less than 0.05) of LH from hypothalamo-pituitary pairs, but not from the pituitary alone. Perifusion of the pituitary alone with medium containing 1 ng/ml EGF, resulted in significant release of LH (70-140% increase, P less than 0.05) after administration of 10(-7) M E2, but did not significantly influence LH release in response to 20 ng/ml LRH. These findings suggest that EGF may be involved in the regulation of pituitary gonadotrophin secretion by a direct effect on the hypothalamus and indirectly by increasing the pituitary responsiveness to E2.     

Epidermal growth factor-binding sites, present in normal human and rat pituitaries, are absent in human pituitary adenomas

To investigate the involvement of epidermal growth factor (EGF) and its receptor in the pathogenesis of human pituitary adenomas, we examined the presence of EGF-binding sites in normal rat and human pituitaries and in human PRL- and GH-secreting and nonsecreting pituitary adenomas. Using crude membrane preparations, specific binding for [125I] EGF was found in normal rat and human pituitaries. Equilibrium was reached at 25 C in 40 min. There was no change in the Kd values between male rats [Kd, 0.65 +/- 0.35 nM (mean +/- SD)] and female rats (Kd, 0.51 +/- 0.15 nM), while the maximum capacity was significantly higher (P less than 0.05) in male rats (21 +/- 8 fmol/mg protein) than in female rats (10 +/- 2 fmol/mg protein). Scatchard analysis of the data suggested the presence of a single class of binding sites. In the three normal human pituitaries tested, specific [125I]EGF binding was also demonstrated. However, both the Kd and the maximum capacity varied widely. Twenty-two human pituitary adenomas were tested, but no specific binding was detected in any of them. In addition to the binding experiments, a radioreceptor assay using rat liver membranes was developed to detect EGF or EGF-like material in extracts of six human pituitary adenomas (two of each type). No EGF activity was detected in any of the extracts. From these results, we conclude that EGF-binding sites are present in normal pituitary tissue, suggesting a physiological role for EGF in this tissue. Consequently, the reason(s) for the lack of EGF binding in pituitary adenoma membranes is not known.     

Growth hormone stimulates insulin gene expression in cultured human fetal pancreatic islets

The total body mass of the human fetus increases about 100-fold from 10-20 weeks of gestation, and peak serum GH concentrations occur at 20 weeks. Since insulin has an essential growth-promoting influence in the fetus, these experiments were designed to determine whether GH can function as a growth factor with insulin-releasing activity by stimulating insulin gene expression during embryogenesis. Pancreatic islets were isolated from human fetuses (n = 36) of 18-22 weeks gestational age. Insulin gene expression was quantified by measuring insulin mRNA by blot hybridization analysis using a [32P] cDNA probe encoding human insulin. We found that by 48 h of culture insulin gene expression was stimulated by 1.25 micrograms recombinant human GH/mL medium to 216% of the control value (n = 2). Insulin secretory capacity was expressed as a fractional stimulation ratio (FSR = F2/F1) of insulin release rates during two successive 1-h static incubations. After 48 h of culture 1.25 micrograms/mL GH stimulated the FSR value to 273% of the control value (n = 5). We conclude that recombinant human GH significantly enhances the steady state level of insulin mRNA concurrent with an increase in insulin secretory capacity, hence providing evidence for a regulatory function of GH on insulin gene expression during fetal development.     

Luteinizing hormone synthesis in cultured fetal human pituitary cells exposed to gonadotropin-releasing hormone.

To investigate whether GnRH regulates LH synthesis during human development, pituitary cells from second trimester fetuses were incubated with [35S]methionine ([35S]met) and [3H]glucosamine ([3H]gln) for 48 h with 0, 10(-9), and 10(-7) mol/L GnRH. Immunoassayable (i) LH was measured in media and cellular lysates, and dual label scintillation analysis was used to quantitate incorporation of radiolabeled precursors into cells, trichloroacetic acid-precipitable proteins, and immunoprecipitated LH subjected to electrophoresis. Exposure of cells to GnRH did not affect cellular uptake or incorporation of precursors into proteins, but specifically increased total (secreted plus cellular) LH synthesis. Both GnRH concentrations significantly increased iLH release and enhanced secreted and cellular [3H]gln-LH. The secretion of [35S] met-LH was stimulated only by 10(-7) mol/L GnRH. The proportion of newly synthesized LH that was secreted and the 3H/35S ratio of secreted and cellular LH were uninfluenced by GnRH. Although basal LH synthesis was not sex dependent, total iLH content and GnRH-stimulated LH translation were greater in cells from females than in those from males. Therefore, GnRH regulates LH synthesis by second trimester fetal human gonadotrophs without influencing the proportion of total radiolabeled LH that is secreted. The existence of a sex difference in total iLH content and GnRH-stimulated LH translation is consistent with the sexual dimorphism in pituitary LH content occurring during human development.     

Management of aggressive growth hormone secreting pituitary adenomas

Aggressive GH-secreting pituitary adenomas (GHPAs) represent an important clinical problem in patients with acromegaly. Surgical therapy, although often the mainstay of treatment for GHPAs, is less effective in aggressive GHPAs due to their invasive and destructive growth patterns, and their proclivity for infrasellar invasion. Medical therapies for GHPAs, including somatostatin analogues and GH receptor antagonists, are becoming increasingly important adjuncts to surgical intervention. Stereotactic radiosurgery serves as an important fallback therapy for tumors that cannot be cured with surgery and medications. Data suggests that patients with aggressive and refractory GHPAs are best treated at dedicated tertiary pituitary centers with multidisciplinary teams of neuroendocrinologists, neurosurgeons, radiation oncologists and other specialists who routinely provide advanced care to GHPA patients. Future research will help clarify the defining features of “aggressive” and “atypical” PAs, likely based on tumor behavior, preoperative imaging characteristics, histopathological characteristics, and molecular markers.

     

Expression of 5'-deiodinase enzymes in normal pituitaries and in various human pituitary adenomas

OBJECTIVE: Local 5'-deiodination of l-thyroxine (T(4)) to active thyroid hormone 3,3',5-tri-iodothyronine (T(3)) catalyzed by the two 5'-deiodinase enzymes (D1 and D2) regulates various T(3)-dependent functions in the anterior pituitary and has been well studied in rodents. Only limited information about deiodinase expression and its cellular distribution in human anterior pituitaries is available. DESIGN: We examined 5'-deiodinase enzyme activities in pituitary adenomas (18 non-functioning, seven TSH-producing, one GH- and TSH-producing, five GH-producing, eight prolactin (PRL)-producing, two adenomas each from patients with Cushing's disease and Nelson's syndrome) and three normal anterior pituitaries. METHODS: Activities were measured as release of (125)I(-) from tyrosyl-ring labeled reverse T(3) with or without propylthiouracil, a potent inhibitor of D1 which does not influence D2 activities. RESULTS: Most of the adenomas and normal tissues expressed both isoenzymes, with D2 activity higher than D1. In a few tissues D1 activity was higher than D2 and some tissues did not express D1 activity at all. Highest activities of both enzymes were found in TSH- and PRL-producing adenomas but absolute activities and the D1/D2 ratio were variable in the same kind of tumor in different patients. CONCLUSION: The finding that all examined tissues expressed 5'-deiodinase activity, most of them expressing both isoenzymes, implies that both enzymes are still active in tumors and that local deiodination is important for the function and feedback regulation of human anterior pituitary.