Human Anterior Pituitary Research Articles

The ontogeny of pit-1 expression in the human fetal pituitary gland.

Pit-1 expression has been reported to be cell-type-specific in the adult human pituitary and in human pituitary adenomas. In contrast, studies of rodent fetal adenohypophysial development as well as mature rodent glands have indicated that the pit-1 mRNA is ubiquitously expressed and the protein is under translational control. To determine the ontogeny and localization of Pit-1 expression in the human fetus, we examined fetal pituitaries (n = 23) at various stages of gestation from 6 weeks to term using in situ hybridization (ISH) for pit-1 mRNA, immunohistochemical localization of Pit-1 protein, and combined ISH for pit-1 mRNA with immunohistochemistry for pituitary hormones. At 6 and 7 weeks of gestation, the cells surrounding both limbs of Rathke's cleft showed a moderate specific signal for pit-1 mRNA. At 7 weeks, only a few cells were immunoreactive for ACTH and there was no definite colocalization of that hormone with pit-1 mRNA. At 8 and 9 weeks of gestation, there was definite preferential expression of pit-1 mRNA in cells containing growth hormone (GH) but not ACTH, as well as in cells with no detectable hormone immunopositivity. At midgestation and after, there was clear correlation between pit-1 mRNA expression and hormone content; cells with GH, prolactin and/or thyrotropin immunoreactivity had abundant pit-1 mRNA, whereas those containing ACTH, FSH or LH were negative for pit-1 mRNA by ISH and its protein by immunohistochemistry. The signal for pit-1 mRNA was stronger in intensity and present in more cells in fetal than in adult adenohypophyses. The ontogeny of pit-1 mRNA expression indicates that it precedes the onset of pituitary hormone detection. Its abundance in the human pituitary early in gestation may reflect its role in cytodifferentiation and cell proliferation. The correlation between pit-1 mRNA detection and Pit-1 protein localization is consistent with a cell-type-specific pretranslational regulatory mechanism for Pit-1 expression in the developing human adenohypophysis.

Cellular localization of the growth hormone receptor/binding protein in the human anterior pituitary gland.

The increasing use of GH therapy has led to the description of its target cells in human tissues, but no data are yet available on the localization of the GH receptor in the human pituitary. In the present study, we used immunocytochemistry to detect the presence of GH receptor/binding protein (GHR/BP), and we examined its distribution among the different types of human anterior pituitary cells. Human pituitaries were taken from autopsies and were processed for embedding in paraffin wax. Immunocytochemistry was performed by using monoclonal antibody 263 raised against purified rat and rabbit GHR/BP, which cross-reacts with the human GH receptor. In order to determine the types of cells that display immunoreactivity for GHR/BP, adjacent pituitary sections were used to detect immunoreactivity for GH, PRL, ACTH, TSH, LH, and FSH. Several controls were carried out to verify the specificity of the immunostaining. Receptor immunoreactivity was found in the cytoplasm and nuclei of the somatotrophs, lactotrophs, and gonadotrophs but not in the thyrotrophs or corticotrophs. In order to demonstrate that the detected GHR/BP immunoreactivity was not caused solely by a cellular capture, we also investigated the cellular distribution of GHR gene expression. This was performed by in situ hybridization with use of complementary oligonucleotide radioactive probes encoding distinct domains of the GHR. Several tests were carried out to validate the detection of gene expression. In situ hybridization demonstrated the presence of GHR messenger RNAs in the anterior lobe of human pituitary, and examination of the signal strengthened the cell-specificity of GHR gene expression. These results demonstrate the presence of GHR/BP in discrete human pituitary cells and indicate a paracrine, autocrine, or intracrine role for GH in the pituitary.

Cellular localization of the growth hormone receptor/binding protein in the human anterior pituitary gland

Cellular localization of the growth hormone receptor/binding protein in the human anterior pituitary gland

Characterization and localization of an immunoreactive growth hormone- releasing hormone precursor form in normal and tumoral human anterior pituitaries

Arguments favor an in situ synthesis of GH-releasing hormone (GHRH) in the normal and tumoral human anterior pituitary. These tissues may express human (h) GHRH messenger RNA, contain hGHRH-(1-44)-NH2, and secrete in vitro an immunoreactive form (ir-form) of the peptide. Here, we characterize and localize the precursor of hGHRH in human anterior pituitary tissues using RIAs specific for the C-terminus or the midportion of hGHRH-(1-44)-NH2, size-exclusion chromatography, HPLC, Western blotting, and immunocytochemistry. The anterior pituitary ir-forms were compared to those found in hypothalamus, posterior pituitary, and GHRH-secreting endocrine pancreatic tumors. Three ir-forms of hGHRH with mol wt of 30-45, and 5 kilodaltons (kDa) were detected. The 30- to 45-kDa ir-form was very likely to consist of hGHRH bound to proteins. The 5-kDa ir-form represented mature forms of hGHRH. It was the major form in tissues actively synthesizing and/or secreting hGHRH. Nontumoral anterior pituitaries contained significant amounts of mature hGHRH. The 10-kDa form was identified as a hGHRH precursor ir-form. In addition to its expected presence in the hypothalamus and GHRH-secreting tumors, normal and tumoral human anterior pituitaries contained an identical ir-form of the hGHRH precursor. Cells immunoreactive for the hGHRH precursor were observed in pituitary adenomas. Evidence for precursor and mature ir-forms of hGHRH in anterior pituitary tissues provides conclusive arguments for the endogenous synthesis of the neuropeptide.

Characterization and localization of an immunoreactive growth hormone-releasing hormone precursor form in normal and tumoral human anterior pituitaries.

Arguments favor an in situ synthesis of GH-releasing hormone (GHRH) in the normal and tumoral human anterior pituitary. These tissues may express human (h) GHRH messenger RNA, contain hGHRH-(1-44)-NH2, and secrete in vitro an immunoreactive form (ir-form) of the peptide. Here, we characterize and localize the precursor of hGHRH in human anterior pituitary tissues using RIAs specific for the C-terminus or the midportion of hGHRH-(1-44)-NH2, size-exclusion chromatography, HPLC, Western blotting, and immunocytochemistry. The anterior pituitary ir-forms were compared to those found in hypothalamus, posterior pituitary, and GHRH-secreting endocrine pancreatic tumors. Three ir-forms of hGHRH with mol wt of 30-45, and 5 kilodaltons (kDa) were detected. The 30- to 45-kDa ir-form was very likely to consist of hGHRH bound to proteins. The 5-kDa ir-form represented mature forms of hGHRH. It was the major form in tissues actively synthesizing and/or secreting hGHRH. Nontumoral anterior pituitaries contained significant amounts of mature hGHRH. The 10-kDa form was identified as a hGHRH precursor ir-form. In addition to its expected presence in the hypothalamus and GHRH-secreting tumors, normal and tumoral human anterior pituitaries contained an identical ir-form of the hGHRH precursor. Cells immunoreactive for the hGHRH precursor were observed in pituitary adenomas. Evidence for precursor and mature ir-forms of hGHRH in anterior pituitary tissues provides conclusive arguments for the endogenous synthesis of the neuropeptide.

Stimulatory effect of gonadotropin-releasing hormone (GnRH) on in vitro prolactin secretion and presence of GnRH specific receptors in a subset of human prolactinomas.

The purpose of this study was to determine whether gonadotropin-releasing hormone (GnRH) may exert a direct action on human prolactinomas. On a series of 17 adenomas, we studied the effect of GnRH on the in vitro prolactin (PRL) secretion of dispersed and perifused cells of 10 cases and the [125I]GnRH agonist binding on frozen sections of three out of the adenomas studied in perifusion and on the membrane preparations of seven other cases. Two 20-min pulses of GnRH (10(-7) mol/l) stimulated the in vitro PRL secretion of three adenomas out of 10 (increase of 200, 444 and 205%, respectively, above basal levels). The GnRH receptors of three adenomas bound GnRH agonist (Des-Gly10-(D-Ala6)-GnRH ethylamide). The binding was specific, with a high affinity (Kd = 0.60, 0.48 and 0.40 nmol/l) similar to that of two human anterior pituitaries obtained post-mortem (Kd = 0.70 and 0.40 nmol/l). Indirect immunoperoxidase revealed that the majority of the cells (60-90%) in all the adenomas studied contained immunoreactive PRL. Four of them also contained cells immunoreactive to the alpha-subunit of the glycoprotein hormones. In none of the prolactinomas were cells immunoreactive to antiserum of anti-beta-luteinizing hormone, anti-beta-follicle-stimulating hormone or anti-beta-thyrotropin. All the prolactinomas that were responsive to GnRH in perifusion experiments and/or bound specifically to [125I]GnRH agonist were also immunoreactive for alpha-subunit. These results show that GnRH, via GnRH specific receptors, exerts a stimulation on in vitro PRL secretion in a subset of prolactinomas characterized by the presence of alpha-subunit.

Is estrogen-induced pituitary hyperplasia and hyperprolactinaemia mediated by angiotensin II?

Human and rat anterior pituitary glands (APG) contain local renin-angiotensin systems (RAS). In the rat APG renin, renin mRNA, angiotensin converting enzyme (ACE) and angiotensin II (A II) have been identified within gonadotrophs (Naruse et al. 1986; Deschepper et al. 1986; Thomas and Sernia 1990). Angiotensinogen (AGEN) and AGEN mRNA have also been found in the rat APG but their localization remains still unclear (Ganong and Deschepper 1989; Healy 1992). Main target sites for A II action in APG seem to be lactotrophs. Specific A II binding sites have been identified mainly on lactotrophs and A II has been shown to stimulate prolactin (PRL) secretion (Aguilera et al. 1982; Denef and Schramme 1985). Recently, we have found that A II stimulated the proliferation of lactotrophs isolated from the estrogen-induced pituitary tumors in the rat (Kunert-Radek and Pawlikowski 1992). Moreover, an ACE inhibitor, enalapril, was found to prevent the pituitary proliferative response to a single injection of estradiol in rats (Mucha et al. 1993). Estrogens are well known to stimulate the pituitary cell proliferation and tumorigenesis and the estrogen-induced APG hyperplasia involves mainly lactotrophs (Zondek 1936; Mietkiewski 1959; Lloyd et al. 1975). Estrogens have been also found to enhance the prolactin secretion (Nicoll and Meites 1962). On the other hand, estrogens were found to increase the AGEN plasma levels (Menard et al. 1970) and to enhance AGEN mRNA expression in the liver, brain (Kunapuli et al. 1987) and pituitary (Healy et al. 1992). Thus, a question arises whether A II is involved in the estrogen action on APG. The aim of the experiments presented below was to test the hypothesis in question.

Messenger ribonucleic acid expression of platelet-derived growth factor subunits and receptors in pituitary adenomas.

Little is known about the expression of growth factors and their receptors in pituitary tumors or the relationship of growth factors to pituitary neoplasia. Platelet-derived growth factor (PDGF) is a potent mitogen that has been postulated to stimulate tumor growth through autocrine and/or paracrine loops in a number of human tumors. In the present study we demonstrate messenger ribonucleic acid expression of the PDGF subunits and receptors in a variety of human pituitary adenomas and in a normal human anterior pituitary gland. Northern blot analysis performed on 34 pituitary adenomas showed that all tumors expressed the PDGF-A and PDGF-B subunits, the majority (94%) expressed the PDGF-beta receptor, and a subset (44%) expressed the PDGF-alpha receptor. The normal anterior pituitary studied expressed all of the PDGF subunits and receptor subunits. Clinically, there was no correlation between expression of the PDGF subunits or receptors and tumor size or with invasion into adjacent structures (cavernous sinus, sphenoid sinus, or clivus). A higher proportion of the endocrinologically active adenomas expressed the PDGF-alpha receptor (5 of 8) compared to the endocrine inactive tumors (10 of 26). The function of PDGF in the pituitary is currently not known. The finding that PDGF subunit and receptor messenger ribonucleic acids are coexpressed in pituitary adenomas and normal anterior pituitary suggests that autocrine and/or paracrine loops involving PDGF may exist in pituitary adenomas and normal anterior pituitary.

Immunoelectron microscopic study of the GH cell in the anterior pituitary gland of normal human fetus.

Ultrastructural studies of growth hormone-producing cells (GH cells) in the anterior pituitary gland have been reported using several experimental animals. However, no attempt has yet been made to identify the ultrastructural heterogeneity of the GH cells within the human anterior pituitary gland. To this end, we employed immunogold electron microscopy to investigate the ultrastructural characteristics of GH cells in relation to gestational age in normal human fetuses. Based on ultrastructural characteristics, three distinct types of GH cells were identified by immunogold electron microscopy in the anterior pituitary glands of 34 normal human fetal pituitary glands. The age of the tissue samples ranged from 8 to 34 weeks. The Type-I GH cell is a small, round cell with a narrow cytoplasm containing a few small secretory granules (268 nm in mean diameter). The GH cells designated Type-II are polygonal and contain medium-sized secretory granules (347 nm), profiles of rough endoplasmic reticulum (RER) arranged in parallel lamellae, and a Golgi complex which is frequently encountered but only in this cell type. The Type-III GH cell is polygonal, large, and contains numerous large spherical-shaped secretory granules (404 nm). The Type-I was the predominant cell type until about 20 weeks of gestation; its incidence decreased thereafter. In contrast, the Type-II and Type-III cells increased in number starting at 20 weeks of gestational age. From these results, we suggest that Type-I is the most immature type of GH cell, Type-III the most mature, and the Type-II is intermediate in development. The marked difference in the incidence of each GH cell type between the first and second half of gestation appears to be a reflection of the development of the hypothalamic regulation of the anterior pituitary gland, which is reported to be established at around 20 weeks of gestation.

Human pituitary somatotropes express transforming growth factor-alpha and its receptor.

Transforming growth factor-alpha (TGF-alpha) is a growth-regulatory peptide produced by a variety of transformed and non-transformed cells. Among non-transformed cells, TGF-alpha has been identified in the prolactin (PRL)- and GH-secreting cells of the bovine anterior pituitary gland. In this report, we have examined the expression of TGF-alpha in human anterior pituitary glands by Western analysis and immunohistochemistry. For the Western analysis, human pituitary glands were extracted in acid/ethanol, an acetic acid-soluble fraction was ether-precipitated and dialysed, and TGF-alpha was partially purified by C18 chromatography. TGF-alpha was then identified by immunostaining of Western transfers. Anterior pituitary extracts exhibited a major band(s) migrating at 19 kDa that was immunoreactive with a monoclonal antibody directed against the mature TGF-alpha. However, no evidence of the fully processed 6 kDa TGF-alpha was observed. We next identified TGF-alpha by immunohistochemistry. Using both monoclonal and polyclonal antibodies, specific immunoreactivity was identified in a population of secretory cells in the anterior pituitary gland. Using antibodies specific for the COOH and NH3 terminals of the TGF-alpha precursor, a comparable number of TGF-alpha-positive cells were found to contain TGF-alpha precursor sequences. These results indicate that the 19 kDa form of TGF-alpha expressed in the human pituitary gland may exist as the transmembrane form. We next sought to determine which cells express TGF-alpha in a human male pituitary gland. On frontal sections, TGF-alpha-immunopositive cells were evenly distributed in a manner and number indistinguishable from GH-immunopositive cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Cell type-specific expression of the pituitary transcription activator pit-1 in the human pituitary and pituitary adenomas.

Pit-1 is a transcription factor that has been shown to be critical for pituitary-specific activation of the GH and PRL genes. In rodents and humans, differentiation and/or maintenance of somatotroph, lactotroph, and thyrotroph phenotypes are dependent on expression of a functional pit-1 gene. In rodents, Pit-1 protein is detectable in only these three cell types; however, pit-1 mRNA transcripts appear to be present at comparable levels in all adenohypophysial cell types, suggesting that translational controls may dictate the pattern of Pit-1 expression. We examined the distribution of pit-1 transcripts in the human pituitary and pituitary adenomas. All tumors were characterized by immunocytochemistry, electron microscopy, and tissue culture for accurate classification. Northern blot analysis demonstrated abundant levels of pit-1 mRNA in somatotroph, mammosomatotroph, and lactotroph adenomas. Two clinically silent adenomas that expressed TSH as well as gonadotropins contained detectable levels of pit-1 mRNA. No pit-1 expression was otherwise detected in corticotroph, gonadotroph, null cell, or oncocytic adenomas. In situ hybridization localized pit-1 mRNA transcripts in adenomas that contained GH, PRL, or TSH, but not in adenomas composed of other cell types. Pit-1 mRNA was also localized to selected subpopulations of the human nontumorous adenohypophysis that contained immunoreactivity for GH, PRL, and/or TSH. Pit-1 protein immunoreactivity was detected in the nuclei of adenomas that expressed pit-1 mRNA, but not in those that were negative for pit-1 mRNA; it was also localized only in cells containing GH, PRL, or TSH beta in the nontumorous adenohypophysis. These data demonstrate selective expression of the human pit-1 gene in adenohypophysial cell types responsible for GH, PRL, and/or TSH synthesis and are consistent with a predominantly pretranslational regulatory mechanism for Pit-1 expression in the human.

In vitro modulation of growth hormone (GH) secretion from early to midgestation human fetal pituitaries by GH-releasing factor and somatostatin: role of Gs-adenylate cyclase-Gi complex and Ca2+ channels.

Using explant cultures of human fetal anterior pituitary glands (9-19 weeks fetal age) and an acute (3-h) test protocol, we investigated the role of two signal transduction pathways (Gs-adenylate cyclase-Gi, Ca2+ channels) in GH-releasing factor (GRF)/somatostatin (SRIF) regulation of GH secretion during the first half of gestation. Data have been analyzed for ontogenic changes using three age groups: 9-10, 12-13, and 15-19 weeks fetal age. The fetal somatotrope shows dose-related responses to forskolin (0.1-10 microM), dibutyryl cAMP (dBcAMP; 0.1-1 mM), and theophylline (0.01-1 mM), all factors that increase intracellular concentrations of cAMP; there is a significant age-related increase in the stimulatory effects of 1 mM dBcAMP and 1 mM theophylline. When any one of these three factors is added with 10 nM GRF, there are no significant additive or synergistic effects on GH secretion. Although 10 nM SRIF has no effect at 9-10 weeks, it is inhibitory in the 12-13 and 15-19 week groups, significantly suppressing the stimulatory effect of 1 microM forskolin and completely blocking the effects of 1 mM dBcAMP or theophylline. Pretreatment of cultures with pertussis toxin completely blocks SRIF inhibition of both basal and GRF-stimulated GH release. KCl (5-50 mM) and Bay K 8644 (0.1-10 microM), both activators of Ca2+ channels, have dose-related stimulatory effects on GH release; 50 mM KCl shows an age-related increase in stimulatory activity. If 10 nM GRF, 1 microM forskolin, 1 mM theophylline or 1 mM dBcAMP is added with either 50 mM KCl or 1 microM Bay K 8644, there is an additive response. SRIF (10 nM) completely blocks the stimulatory effects of 1 microM Bay K 8644 and markedly inhibits the effects of 50 mM KCl from as early as the ninth week of fetal age. The Ca2+ channel blocker nifedipine (1-10 microM) significantly inhibits basal as well as stimulated (GRF, forskolin, dBcAMP, theophylline, KCl, and Bay K) GH release from as early as 9 weeks fetal age; in contrast, the calmodulin blocker trifluoperazine (0.01-10 microM) has no effects on basal GH secretion and only slightly inhibits the stimulatory effects of KCl. Pretreatment with 10 nM GRF for 24 h significantly decreases a subsequent 3-h response to 10 nM GRF, but does not alter the subsequent response to 1 mM theophylline or 50 mM KCl.(ABSTRACT TRUNCATED AT 400 WORDS)

Effects of growth hormone (GH)-releasing factor and somatostatin on GH secretion from early to midgestation human fetal pituitaries.

Using explant cultures of human anterior pituitary glands (9-19 weeks fetal age) and an acute (3-h) test protocol, we investigated fetal somatotrope responsiveness to human GH-releasing factor [hGRF-(1-44)] and somatostatin [SRIF-(1-14) and SRIF-(1-28)] as a function of age. Ontogenic data were analyzed using three age groups: 9-10, 12-13, and 15-19 weeks fetal age. Both daily (24-h) and acute test (3-h) basal GH secretion increases as a function of fetal age, with the greatest increase occurring after 12-13 weeks; however, the 3 h/24 h secretion ratio remains unchanged, at approximately 12%. GRF (0.1-10 nM) stimulates GH release in a dose-related fashion, regardless of fetal age; there is a significant increase in the response to both 1 nM (P < 0.05) and 10 nM (P < 0.01) GRF between 9-10 and 15-19 weeks fetal age and to 10 nM GRF (P < 0.05) between 12-13 and 15-19 weeks. Pretreatment of cultures (9-19 weeks) with 1 or 10 nM GRF for 24 h does not alter basal 3-h GH secretion, but significantly decreases subsequent responses to 1 or 10 nM GRF, respectively (P < 0.01). Pretreatment with 1 nM GRF does not alter a subsequent 3-h response to 10 nM GRF. SRIF-(1-14) (1-100 nM) causes a dose-related inhibition of basal GH secretion from as early as the ninth week of fetal life; there is a small age-related increase in the somatotrope response to 100 nM SRIF-(1-14) between 12-13 and 15-19 weeks fetal age (P < 0.05). In a group of 11- to 14-week-old fetal pituitaries, SRIF-(1-28) had a significantly (P < 0.05) greater inhibitory effect than SRIF-(1-14) at both 1 and 10 nM; the two peptides decreased basal GH secretion to a similar extent at 100 nM (52.8 +/- 4.0% of control value; P < 0.01). SRIF-(1-14) (10 and 100 nM) does not significantly alter 10 nM GRF-stimulated GH release from 9- to 10-week-old fetal pituitaries. However, by 12-13 weeks, 10 nM SRIF-(1-14) reduces GRF-stimulated GH secretion by 60% (P < 0.01), while 100 nM SRIF-(1-14) decreases it by 80% (P < 0.01); similar inhibitory effects are observed with 15- to 19-week-old fetal somatotropes.(ABSTRACT TRUNCATED AT 400 WORDS)

Folliculo-Stellate Cells in the Human Anterior Pituitary Express Cytokeratin

Recent studies have suggested that the anterior pituitary is derived not from Rathke's pouch, but instead from neuroectoderm. To address this controversy and the differentiation of the adenohypophysis, we evaluated folliculo-stellate cells from 7 human pituitaries using a panel of immunohistochemical stains. The panel included antibodies to a low molecular weight cytokeratin (CK-8), a broad spectrum of cytokeratins (AE 1/AE 3), S-100 protein, vimentin, and glial fibrillary acid protein. The most intense immunostaining was observed with the antibody to broad spectrum cytokeratins. Less intense immunohistochemical staining of folliculo-stellate cells was observed with the antibodies against glial fibrillary acidic protein and S-100 protein. Immunoblotting of anterior pituitary tissue obtained from 3 autopsied patients showed that cytokeratins identified in the anterior pituitary were of low molecular weight (40 to 56 Kd). These results indicate that folliculo-stellate cells show epithelial differentiation, and support the hypothesis that the anterior pituitary is derived from Rathke's pouch.

Human monoclonal islet cell antibodies from a patient with insulin-dependent diabetes mellitus reveal glutamate decarboxylase as the target antigen.

The autoimmune phenomena associated with destruction of the beta cell in pancreatic islets and development of type 1 (insulin-dependent) diabetes mellitus (IDDM) include circulating islet cell antibodies. We have immortalized peripheral blood lymphocytes from prediabetic individuals and patients with newly diagnosed IDDM by Epstein-Barr virus transformation. IgG-positive cells were selected by anti-human IgG-coupled magnetic beads and expanded in cell culture. Supernatants were screened for cytoplasmic islet cell antibodies using the conventional indirect immunofluorescence test on cryostat sections of human pancreas. Six islet cell-specific B-cell lines, originating from a patient with newly diagnosed IDDM, could be stabilized on a monoclonal level. All six monoclonal islet cell antibodies (MICA 1-6) were of the IgG class. None of the MICA reacted with human thyroid, adrenal gland, anterior pituitary, liver, lung, stomach, and intestine tissues but all six reacted with pancreatic islets of different mammalian species and, in addition, with neurons of rat cerebellar cortex. MICA 1-6 were shown to recognize four distinct antigenic epitopes in islets. Islet cell antibody-positive diabetic sera but not normal human sera blocked the binding of the monoclonal antibodies to their target epitopes. Immunoprecipitation of 35S-labeled human islet cell extracts revealed that a protein of identical size to the enzyme glutamate decarboxylase (EC 4.1.1.15) was a target of all MICA. Furthermore, antigen immunotrapped by the MICA from brain homogenates showed glutamate decarboxylase enzyme activity. MICA 1-6 therefore reveal glutamate decarboxylase as the predominant target antigen of cytoplasmic islet cell autoantibodies in a patient with newly diagnosed IDDM.

Detection of thyrotropin-releasing hormone (TRH) mRNA by the reverse transcription-polymerase chain reaction in the human normal and tumoral anterior pituitary

To investigate the presence of TRH mRNA in the human anterior pituitary tissue, total RNA from human normal and tumoral anterior pituitary, hypothalamus (positive control) and muscle tissues (negative control) was reverse transcribed (RT) to the first strand of cDNA. RT products were then amplified by polymerase chain reaction (PCR) using a set of three exon-specific primers (two external 5' and 3' primers and one internal 3' primer) for a target sequence of the TRH gene including an intronic sequence of about 650 base pairs (bp). Southern analysis of the RT-PCR products specifically hybridizing with a 45-mer TRH probe showed two bands of the predicted sizes (399 and 351 bp) far more intense in hypothalamus than in normal and tumoral anterior pituitary tissue. The 399 and 351 bp RT-PCR products contained the BglII enzyme restriction site included in the TRH cDNA sequences spanned by the primers and the two respective digested fragments which were, as predicted, 337 and 289 bp long, hybridized with the TRH probe. Based on these results, we can conclude that the RT-PCR products generated from RNA tissue were the target TRH sequences in the human normal and tumoral anterior pituitary tissue as well as in the hypothalamus. Our data imply TRH gene expression in the human anterior pituitary.

Localization and quantification of pro-opiomelanocortin mRNA and glucocorticoid receptor mRNA in pituitaries of suicide victims.

Suicidal behavior has been associated with hypothalamic-pituitary-adrenal overactivity in humans, as measured by increased corticosteroid secretion. To investigate whether this overactivity is reflected at the pituitary level, we have studied the localization of pro-opiomelanocortin (POMC) mRNA, and glucocorticoid receptor (GR) mRNA, in human anterior pituitaries, and quantified these messages relative to controls. Pituitaries from 7 suicide victims and 11 cardiac deaths were sectioned into 10-microns slides, stained with thionin and processed for in situ hybridization using a riboprobe complementary to human POMC mRNA. To correct for possible postmortem cell loss, hybridization with P1B15, a cDNA complementary to rat cyclophillin mRNA, was used in adjacent sections. POMC mRNA containing cells were found to be localized in clusters and were highly associated with corticotropin-releasing hormone (CRH) receptors. In contrast, GR mRNA containing cells were distributed through the pituitary, although areas of increased density were associated with POMC mRNA cells. Quantification with a computerized image analysis system revealed a 25% increase in POMC message in suicide victims. Analysis of the corticotrophic cell clumps showed that the suicide victims had higher POMC mRNA density per cell (p = 0.04) and larger corticotrophic cell size (p = 0.04) than the cardiac death victims. No differences in GR mRNA were detected between the two groups, although GR and POMC mRNA levels were highly and significantly correlated (r = 0.8, p < 0.001). There were no differences in P1B15 message between the two groups. We conclude that in situ hybridization is a useful tool to study gene regulation in human neuroendocrine tissue and that suicide victims show evidence of chronic hypothalamic-pituitary-adrenal axis activation.

The asparagine-linked oligosaccharides at individual glycosylation sites in human thyrotrophin.

The asparagine-linked carbohydrate structures at each of the three glycosylation sites of human thyrotrophin were investigated by 400 MHz 1H-NMR spectroscopy. Highly purified, biologically active human thyrotrophin (hTSH) was dissociated into its subunits hTSH alpha (glycosylated at Asn 52 and Asn 78) and hTSH beta (glycosylated at Asn 23). The alpha-subunit was further treated with trypsin which gave two glycopeptides that were subsequently separated by reverse-phase HPLC and identified by amino acid sequence analysis. The oligosaccharides were liberated from hTSH alpha glycopeptides and from intact hTSH beta by hydrazinolysis, and were fractionated as alditols by anion-exchange and ion-suppression amine-adsorption HPLC preparatory to structural analysis. The N-glycans present on hTSH were mainly diantennary complex-type structures with a common Man alpha 1-3 branch that terminated with 4-O-sulphated GalNAc. The Man alpha 1-6 branch displayed structural heterogeneity in the terminal sequence, with chiefly alpha 2-3-sialylated Gal and/or 4-O-sulphated GalNAc. The relative amounts of the two major complete diantennary oligosaccharides and their core fucosylation differed according to glycosylation site; the sulphated/sialylated diantennary oligosaccharide was most abundant at the two sites on the alpha-subunit, whereas the disulphated, core-fucosylated oligosaccharide was more plentiful on the beta-subunit. Some interesting structural features, not previously reported for the N-glycans of hTSH, included 3-O-sulphated galactose (SO4-3Gal) and peripheral fucose (Fuc alpha 1-3GlcNAc) in the Man alpha 1-6 branch of some diantennary structures; the former suggests the presence of a hitherto uncharacterized galactose-3-O-sulphotransferase in thyrotroph cells of the human anterior pituitary gland.

Cell types in the fetal pars tuberalis of the human adenohypophysis at mid-gestation

The pars tuberalis of the adenohypophysis was investigated in three human fetuses at mid-gestation by electron microscopy or immunohistochemistry. In addition to gonadotrophs and thyrotrophs, identified by immunohistochemistry and ultrastructural morphology, electron microscopy revealed the existence of an additional differentiated cell type closely resembling "pars tuberalis-specific" cells known from other species. The role of this cell type in the human endocrine regulation remains to be elucidated.

Twenty-Four-Hour Rhythms in Plasma Concentrations of Adenohypophyseal Hormones Are Generated by Distinct Amplitude and/or Frequency Modulation of Underlying Pituitary Secretory Bursts*

To evaluate the nature of anterior pituitary secretory events in vivo, we have applied a novel waveform-independent deconvolution technique that dissects the underlying secretory behavior of endocrine glands quantitatively from available serial plasma hormone concentration measurements assuming one- or two-compartment elimination kinetics. We used this new tool to ask the following physiological questions. 1) Does the pituitary gland secrete exclusively in randomly dispersed bursts, and/or does a tonic (constitutive) mode of interburst hormone secretion exist? 2) What secretory mechanisms generate the nyctohemeral rhythms in plasma hormone concentrations? Analysis of 24-h plasma concentration profiles of GH, LH, FSH, PRL, TSH, ACTH, and beta-endorphin (n = 6-8 men/group) revealed that 1) pituitary secretion in vivo occurs in an exclusively burst-like mode for all hormones except TSH and PRL (for the latter two, a mixed burst and constitutive mode pertained); 2) significant nyctohemeral regulation of secretory burst frequency alone was not demonstrated for any hormone; 3) prominent 24-h variations in secretory burst amplitude alone were delineated for ACTH and LH; 4) TSH, GH, and beta-endorphin were both frequency and amplitude controlled; 5) no significant diurnal variations in FSH secretory parameters occurred; and 6) a fixed hormone half-life yielded fits of the 24-h data series with a normalized residual variance of less than 8%. We conclude that the normal human anterior pituitary gland releases its multiple (glyco)protein hormones via punctuated secretory episodes unassociated with tonic basal (constitutive) hormone secretion, except in the case of TSH and PRL. Hormone-specific amplitude and/or frequency control of secretory burst activity over 24 h provides the mechanistic basis for the classically recognized 24-h rhythms in plasma concentrations of adenohypophyseal hormones in men.