Effects of mitotane treatment on human steroid metabolism: implications for patient management

Mitotane (o,p'-DDD), an oral adrenolytic agent for treatment of advanced adrenocortical carcinoma (ACC), is reported to inhibit cortisol biosynthesis in vitro and enhance production from exogenous cortisol of urinary 6β-hydroxycortisol and unidentified polar unconjugated metabolites. We examined urinary steroid profiles by gas chromatography-mass spectrometry of patients with histologically confirmed ACC following surgery, receiving a) hydrocortisone alone (three males and three females) and b) mitotane and hydrocortisone (six males and 11 females). Samples were collected after plasma mitotane had reached the therapeutic range of 14-20 mg/l. Increased excretion of polar unconjugated steroids during mitotane treatment was confirmed, with 6β-hydroxycortisol and 6β-hydroxy-20-dihydrocortisols predominating. The proportion of additionally hydroxylated metabolites was <2% in untreated controls and 52, 35-52% (mean, range) in the mitotane plus hydrocortisone group. Ratios of 5α-/5β- and 20β-/20α-metabolites of administered cortisol were decreased 50-, 15-fold, and 14-, 8-fold respectively (males, females - mean values) but with no change in metabolite ratios that reflect oxidoreduction at C11 or C20. Patterns of decrease in 5α- relative to 5β-reduced metabolites were similar to those of patients with 5α-reductase 2 deficiency or on treatment with the 5α-reductase 2 inhibitor finasteride but different from those of patients on dutasteride, indicating specific inhibition of 5α-reductase 2. We conclude that mitotane causes consistent changes in cortisol catabolism, most of which have not been previously recognised. These need not interfere with early detection of ACC recurrence. Induction of 6β-hydroxylation offers an explanation for a reported decrease in cortisol bioavailability. Mitotane also has potential as a unique steroid metabolic probe for 20β-reduction.

The relationship between steroidogenic factor 1 and DAX-1 expression and in vitro gonadotropin secretion in human pituitary adenomas

The orphan nuclear receptors, steroidogenic factor 1 (SF-1) and DAX-1, are involved in gonadotroph differentiation, and SF-1 has been shown to activate the LH-beta and glycoprotein hormone alpha-subunit (alpha GSU) gene promoters. Pituitary adenomas from 34 patients [13 somatotroph tumors, 4 prolactinomas, and 17 clinically nonfunctioning pituitary adenomas (NFPAs)] were enzymatically dispersed and cultured in vitro for 48 h. Tissue culture medium was collected and assayed for LH, FSH, and alpha GSU; messenger RNA was extracted from adherent cells, and expression of SF-1 and DAX-1 messenger RNA was determined by RT-PCR and verified by direct DNA sequencing. The presence of DAX-1 protein in tumor tissue was confirmed by immunocytochemistry. DAX-1 was demonstrated in all NFPAs, 7 of 13 somatotroph tumors and 0 of 4 prolactinomas. SF-1 expression occurred in 8 of 16 NFPAs, 4 of 12 somatotroph tumors, and 1 of 4 prolactinomas. LH secretion in vitro was greater in NFPAs that were SF-1 positive (P < 0.05). Neither FSH secretion nor alpha GSU secretion in vitro were significantly related to the expression of SF-1 or DAX-1. SF-1-positive somatotroph tumors immunostained positively for LH-beta and/or FSH-beta and secreted gonadotropins in vitro. SF-1 expression is associated with the in vitro secretion of LH by NFPAs. A proportion of somatotroph tumors also express SF-1 and DAX-1 and secrete gonadotropin hormones in vitro.

Expression and secretion of neural cell adhesion molecules by human pituitary adenomas

Neural cell adhesion molecules (NCAMs) are found predominantly in neural, muscle and endocrine cells. Recent interest has focused on their potential role in tumorigenesis. We have analysed the expression and secretion of NCAM in a series of 48 human pituitary adenomas. Immunocytochemical analysis of 19 adenomas demonstrated NCAM expression in all tumours with, in each case, diffuse cytoplasmic staining being found with variable membrane accentuation. There were no apparent differences in the expression of immunoreactivity seen on sections between individual tumours. Cell culture media from 43 dispersed human pituitary tumours were analysed by immunoassay for the secretion of soluble NCAM and all the pituitary hormones. In contrast to the immunocytochemical studies, soluble NCAM was released from only 27% of human pituitary tumours, but this was not related to tumour type nor was the amount of soluble NCAM released correlated with the amount of pituitary hormone secreted by each adenoma. NCAM expression is common to all human pituitary adenoma types and the observed differences in release of soluble NCAM between individual tumours may reflect different molecular mechanisms, altering adhesive interactions between normal and adenomatous tissue.

Steroidogenic factor 1-DNA binding: a kinetic analysis using surface plasmon resonance

Basal expression of the glycoprotein hormone alpha-subunit gene in pituitary gonadotrophs is partially dependent on a gonadotroph specific element (GSE) which binds the nuclear receptor, steroidogenic factor-1 (SF-1). We have used surface plasmon resonance (SPR) to determine the association (kappa ass), dissociation (kappa diss) and affinity (KA) constants of SF-1 binding to immobilized oligonucleotides containing either the GSE consensus motif or a GSE mutant with a 2 bp substitution in the GSE site (GSEMUT). In vitro translated SF-1 protein bound the consensus GSE with a threefold increase in affinity constant (P<0.01) compared with the GSEMUT. This was due primarily to a significant increase (P<0.05) in the kappa ass for SF-1 to the GSE and a slower kappa diss (P<0.05). The binding interaction was specific and could be significantly inhibited (P<0. 001) by either anti-SF-1 antibody or excess non-biotinylated GSE. The addition of 14 bp wild-type flanking sequences significantly reduced the affinity of SF-1 to both the GSE (P<0.05) and the GSEMUT (P<0.01). This was due to a significant (P<0.01) decrease in kappa ass for the wild-type and mutant long oligonucleotides compared with the short GSE. Nuclear extracts from alphaT3-1 gonadotroph cells also bound the GSE and GSEMUT, giving kappa diss values which were two- to threefold slower than those obtained with in vitro translated SF-1. Thus, SPR is a powerful technique for examining kinetic interaction between SF-1 and its binding site, and is able to demonstrate the effects of mutations and flanking sequences on that interaction.

Expression of the growth hormone secretagogue receptor in pituitary adenomas and other neuroendocrine tumors

Synthetic GH secretagogues (GHSs; GH-releasing peptides and their nonpeptide mimetics) stimulate GH release, activate the hypothalamo-pituitary-adrenal axis, and release PRL in vivo. Patients with acromegaly show an exuberant GH response to GHSs, whereas patients with pituitary-dependent ACTH-secreting tumors show an exaggerated rise in ACTH and cortisol. We, therefore, studied the presence of GHS receptor (GHS-R) messenger ribonucleic acid (RNA) in 38 human pituitary tumors of different cell types, 3 ectopic ACTH-secreting tumors, a pancreatic gastrinoma, 3 insulinomas, and a non-secreting thymic carcinoid as well as in 7 normal pituitary glands. Certain pituitary tumors were also studied by in vitro cell culture with measurement of secreted GH, ACTH, PRL, FSH, LH, alpha-subunit, and TSH. RNA was extracted from tissue samples and, after RT, a duplex PCR reaction with primers for the GHS-R gene and for the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase was performed, allowing semiquantitation of GHS-R expression. All the somatotroph adenomas (n = 8) showed a 2-10 times higher expression of the GHS-R gene compared to normal pituitaries. Higher than normal expression was shown in 5 of 18 tumors from patients with ACTH-secreting pituitary adenomas and in 1 of 3 ectopic ACTH-secreting carcinoid tumors. Two of the pituitary ACTH-secreting adenoma samples showed completely absent expression of the GHS-R, 8 showed expression similar to that of normal pituitary tissue, and 3 of the corticotroph adenoma tissue samples and 2 ectopic ACTH-secreting tumors showed a very low level of expression. One of 4 prolactinoma samples showed a high level of expression, 1 showed expression similar to that of normal pituitary, and 2 samples showed a very low level of expression. Nonfunctioning pituitary adenoma samples showed either absent or very low level expression of the GHS-R. The pancreatic gastrinoma sample showed expression similar to that of normal pituitary tissue, whereas 3 insulinomas showed low level expression of the GHS-R gene; a nonsecreting thymic carcinoid tumor showed no detectable expression. In summary, although GHS-R messenger RNA is abundant in human somatotroph adenomas, it is also present in other pituitary adenomas, particularly ACTH-secreting tumors. These findings may explain the in vivo responses to GHSs in patients harboring such tumors. It also appears from our study that GHS-R may be expressed in other neuroendocrine tumors.

Free alpha-subunit and intact TSH secretion in vitro are closely associated in human somatotroph adenomas

OBJECTIVE

GH-secreting pituitary adenomas frequently co-secrete prolactin and glycoprotein hormone alpha-subunit (alphaSU), but expression of additional hormones is considered unusual. The aim of this study was to establish the frequency with which acromegalic tumours secrete intact glycoprotein hormones LH, FSH and TSH, in comparison with other types of pituitary adenoma.

DESIGN AND METHODS

Pituitary tumours were studied by cell culture, measuring the basal secretion of anterior pituitary hormones in vitro. Light microscopy was used to exclude tumours where normal pituitary tissue was present, and immunocytochemistry was employed to confirm the clinical diagnosis and for comparison with tissue culture data.

RESULTS

TSH secretion was observed in vitro in 15/23 somatotroph adenomas, but from only 1/8 lactotroph, 4/29 null cell, 2/12 gonadotroph and 1/10 corticotroph adenomas; moreover, somatotroph adenomas secreted the largest amounts of TSH (P < 0.(001). Somatotroph adenomas also secreted LH (7/23) and FSH (2/23) but less frequently than gonadotroph adenomas. Immunocytochemistry demonstrated glycoprotein expression in somatotroph adenomas (LHbeta: 13%, FSHbeta: 26%, TSHbeta: 30%, alphaSU: 46%) more frequently than in lactotroph, corticotroph and null cell adenomas. A strong correlation was found between alphaSU secretion and TSH secretion in somatotroph adenomas (rho= 0.683, P < 0.001.

CONCLUSIONS

TSHbeta is frequently expressed by somatotroph adenomas, often associated with alphaSU expression. Both GH and TSHbeta are dependent on the transcription factor, Pit-1, which is frequently expressed in somatotroph adenomas, although the expression of alphaSU requires an alternative explanation. Increased expression of alphaSU compared with TSHbeta may account for the secretion of free alphaSU by somatotroph adenomas.

Mechanism of action of pituitary adenylate cyclase-activating polypeptide on human glycoprotein hormone alpha-subunit transcription in alphaT3-1 gonadotropes

Pituitary adenylate cyclase activating polypeptide (PACAP) has been shown to increase glycoprotein hormone alpha-subunit synthesis and release from pituitary cells. We have used alphaT3-1 clonal gonadotropes to investigate the intracellular mechanisms involved in PACAP regulation of alpha-subunit gene transcription; and using deletion, mutation, and heterologous constructs of the alpha-promoter linked to a luciferase reporter gene, we have defined DNA sequences responsive to PACAP. Stimulation of alphaT3-1 cells for 24 h with PACAP, GnRH, or vasoactive intestinal peptide (VIP) resulted in a time- and concentration-dependent increase in alpha-promoter transcription at 100 nM for GnRH (17.5-fold, P < 0.001), PACAP (12.7-fold, P < 0.01), and VIP (4.1-fold, P < 0.05). Incubation of alphaT3-1 cells in calcium-depleted medium suggested that the transcriptional response to PACAP was less dependent on changes in intracellular calcium concentration, in contrast to the results seen with GnRH or VIP, where alpha-subunit transcription was significantly reduced. Transfection of an alpha-promoter construct containing a mutant cAMP response element (CRE) suggested that the CRE region is involved in PACAP and VIP responsiveness, with stimulatory effects on the mutant construct by PACAP (11.1-fold) and VIP (7.6-fold) being significantly (P < 0.001) reduced, compared with their stimulatory effects (PACAP: 25.6-fold, VIP: 23.1-fold) on the native alpha-promoter. In the same experiment, the transcriptional response of the mutant CRE construct and the native CRE construct to GnRH was not significantly different. Both PACAP and VIP enhanced GnRH-stimulated alpha-subunit gene transcription, but this additive effect was lost when their combined effects on the mutant CRE were examined. Deletion analysis indicated that sequences between -244 and -195 bp were involved in mediating the response to PACAP, with a dramatic reduction in fold-stimulation by PACAP (2.0-fold) of the -195-bp construct, compared with the -244-bp construct (15.8-fold). Constructs containing only upstream alpha-promoter sequences from -517 bp to -98 bp, fused to the heterologous thymidine kinase promoter, exhibited a similar loss of responsiveness to PACAP below -298 bp. Thus, our studies show that, unlike GnRH, PACAP stimulation of alpha-subunit gene transcription in alphaT3-1 cells is less dependent on changes in intracellular calcium concentration; and full transcriptional activation of the alpha-subunit by PACAP requires an intact CRE. PACAP responsiveness involves sequences between -244 and -195 bp of the alpha-promoter. These sequences have been implicated also in GnRH-responsiveness and may thus provide a mechanism for coordinated regulation of the alpha-subunit gene by PACAP and GnRH in alphaT3-1 cells.

Calcium and glycoprotein hormone alpha-subunit gene expression and secretion in alpha T3-1 gonadotropes

GnRH stimulates both transcription and secretion of the alpha-subunit in pituitary cells, but the precise role of the calcium- signaling mechanisms mediating these actions are unclear. We have examined the role of calcium using alpha T3-1 gonadotropes transfected with alpha-promoter constructs linked to a luciferase reporter gene and concomitant measurement of alpha-subunit secretion. The calcium channel agonist, BayK8644 (1 microM) significantly stimulated alpha-subunit transcription (4.9-fold, P < 0.05) but to a lesser extent than either GnRH (100 nM, 20.7-fold, P < 0.001) or phorbol-12-myristate-13-acetate (TPA, 100 nM, 8.7-fold, P < 0.05). The transcriptional response to a combination of BayK8644 and TPA was approximately additive. Despite stimulating alpha-subunit gene expression, BayK8644 had no effect on alpha-subunit secretion at 24 h, and co-addition of BayK8644 and TPA did not produce any further stimulation of alpha-subunit secretion (3.0-fold, P < 0.001) compared with TPA alone (3.2-fold, P < 0.001). Pretreatment of alpha T3-1 cells with the calcium channel blocker, nifedipine (1 microM for 5 min), essentially blocked GnRH-stimulated alpha-promoter activity without affecting GnRH-stimulated alpha-subunit release. In contrast, thapsigargin pretreatment (1 microM for 5 min), which depletes intracellular calcium stores, significantly reduced basal and GnRH-stimulated secretion without affecting the ability of GnRH to increase alpha-promoter activity. Incubation of alpha T3-1 cells in calcium-depleted media showed that the transcriptional response was dependent on extracellular calcium concentration, with maximum stimulation by GnRH seen at a calcium concentration of 1.7 mM. Deletion analysis indicated that sequences between -420 and -244 bp were involved in mediating the response to BayK8644. Constructs containing only upstream alpha-promoter sequences from -517 to -98 bp, fused to the heterologous thymidine kinase promoter, exhibited loss of responsiveness to BayK8644 below -298 bp. These upstream elements were also found to be important for basal expression of the alpha-promoter and for mediating the response to TPA but were distinct from GnRH responsiveness of the human promoter in alpha T3-1 cells. These studies suggest differential regulation of GnRH-stimulated alpha-subunit gene transcription and secretion by extracellular calcium influx and intracellular calcium mobilization. The transcriptional response to extracellular calcium influx is mediated through two or more elements between -420 and -244 bp, which are also involved in basal and TPA-stimulated expression of the alpha-subunit promoter.

Glycoprotein hormone alpha-subunit production and plurihormonality in human corticotroph tumours--an in vitro and immunohistochemical study

Glycoprotein hormone alpha-subunit (alpha SU) is a recognized product of clinically non-functioning, glycoprotein hormone-secreting and somatotroph adenomas but has not been studied systematically in corticotroph tumours. We have performed immunohistochemistry for alpha SU in a consecutive series of four corticotroph tumours causing Nelson's syndrome, three corticotroph macroadenomas, 12 corticotroph microadenomas and one adrenocorticotrophin-secreting bronchial carcinoid tumour. In addition we have assessed alpha SU secretion in vitro in corticotroph adenomas from two subjects with Cushing's disease and two subjects with Nelson's syndrome. Immunohistochemistry, performed after microwave treatment of sections to enhance antigen retrieval, demonstrated alpha SU positivity in 3/4 Nelson's tumours, 2/3 corticotroph macroadenomas, 7/12 microadenomas and one bronchial carcinoid. Eight of the 13 tumours positive for alpha SU were also immunostained after microwave pretreatment of sections for thyrotrophin (six positive), follicle-stimulating hormone (four positive), luteinizing hormone (three positive), beta-chorionic gonadotrophin (five positive), growth hormone (three positive) and prolactin (two positive) immunoreactivity. In vitro cell cultures of all four tumours studied secreted adrenocorticotrophin and three secreted alpha SU, with the variable presence of luteinizing hormone, follicle-stimulating hormone, thyrotrophin, growth hormone and prolactin, in basal culture. The alpha SU secretion was augmented by phorbol ester (160 +/- 15%, SEM, n = 3 wells; p < 0.01) and 8-bromo-cAMP (138 +/- 8%; p < 0.05) in one tumour. These data indicate that plurihormonality and, in particular, alpha SU elaboration and secretion by corticotroph tumours is more common than hitherto recognized. Possible mechanisms include abnormal or deregulated gene expression, autocrine or paracrine effects or a stem cell origin of tumour.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of pituitary adenylate cyclase-activating polypeptide on hormone secretion by human pituitary adenomas in vitro

Hormone release in culture in response to pituitary adenylate cyclase-activating polypeptide (PACAP) was examined in 28 human pituitary adenomas: 10 null cell adenomas, 4 gonadotropin-, 6 GH-, 6 ACTH-, and 2 PRL-producing adenomas. The effects of PACAP38 were compared with those of the classical hypothalamic releasing hormones and other activators of intracellular signaling pathways. PACAP38 significantly stimulated GH release from 1 somatotrope tumor (125 +/- 3% of control; P < 0.05) and ACTH release from 3 corticotrope tumors (134 +/- 6%, 136 +/- 7%, and 137 +/- 9% of control; P < 0.05). The effects of PACAP38 were less potent than either GHRH on GH release in the somatotrope tumor or CRH on ACTH release in the corticotrope tumors but similar to the responses seen with the cAMP analog 8-bromo-cAMP (8-Br-cAMP). No detectable effects of PACAP38 on hormone release from null cell, gonadotropin-, or PRL-producing adenomas were observed. Of the 5 somatotrope tumors that failed to respond to PACAP38, all also failed to respond to either 8-Br-cAMP, TRH, or GHRH. Of the corticotrope tumors that failed to respond, 2 of the 3 also failed to respond to CRH. In addition to eliciting hormone release appropriate to the tumor type, PACAP38 also stimulated glycoprotein hormone alpha-subunit (alpha SU) release from one somatotrope tumor (229 +/- 35% of control, P < 0.01) and one corticotrope tumor (149 +/- 4% of control; P < 0.01). This response was not mimicked by 8-Br-cAMP in the somatotrope tumor, but in the corticotrope tumor a significant alpha SU release was also seen after stimulation with the protein kinase C activator 12-O-tetradecanoyl-phorbol-13-acetate and 8-Br-cAMP. These results suggest that the novel hypothalamic peptide PACAP38 has a modest role in the regulation of GH, ACTH, and alpha SU secretion from some human tumourous pituitary corticotropes and somatotropes. Further studies are needed to elucidate the intracellular signaling pathways that mediate the effects of PACAP on hormone secretion by these tumor types.