Efficacy of chimeric molecules directed towards multiple somatostatin and dopamine receptors on inhibition of GH and prolactin secretion from GH-secreting pituitary adenomas classified as partially responsive to somatostatin analog therapy.

OBJECTIVE: This study compared the potency of a somatostatin receptor (sstr)2-sstr5 analog, BIM-23244, of an sstr2-dopamine D2 receptor (sstr2-DAD2) molecule, BIM-23A387 and of new somatostatin-dopamine chimeric molecules with differing, enhanced affinities for sstr2, sstr5 and DAD2, BIM-23A758, BIM-23A760 and BIM-23A761, to suppress GH and prolactin (PRL) from 18 human GH adenomas that are partially responsive to octreotide or lanreotide. MATERIALS AND METHODS: The sstr2, sstr5 and DAD2 mRNA levels were determined by RT-PCR. The effect of drugs was tested in cell cultures at various concentrations. RESULTS: In all tumors, the sstr2, sstr5 and DAD2 mRNA levels were coexpressed (mean levels+/-s.e.m. 0.4+/-0.1, 5.3+/-1.9 and 2.0+/-0.4 copy/copy beta-glucuronidase). In 13 tumors, the maximal suppression of GH secretion produced by BIM-23A387 (30+/-3%) and BIM-23244 (28+/-3%) was greater than that produced by octreotide (23+/-3%). In six out of 13 tumors, BIM-23A758, BIM-23A760 and BIM- 23A761 produced greater maximal suppression of GH secretion than octreotide (33+/-5, 38+/-2 and 41+/-2 vs 24+/-2%). Their EC(50) values were 10, 2 and 4 pmol/l. BIM-23A761 was more effective than BIM-23A387 in GH suppression (41+/-2 vs 32+/-4%). The new chimeric molecules produced maximal PRL suppression greater than octreotide (62+/-8 to 74+/-5 vs 46+/-11%). CONCLUSIONS: Novel dopamine-somatostatin chimeric molecules with differing, enhanced activity at sstr2, sstr5 and DAD2, consistently produced significatly greater suppression of GH and PRL than either octreotide or single-receptor-interacting ligands in tumors from patients classified as only partially responsive to octreotide therapy. The higher efficacy of the chimeric compounds was, at least partially, linked to their high affinity for sstr2 (IC50 1-10 pmol/l). The other mechanisms by which such molecules produce an enhanced inhibition of GH remain to be elucidated.

BIM-23A760, a chimeric molecule directed towards somatostatin and dopamine receptors, vs universal somatostatin receptors ligands in GH-secreting pituitary adenomas partial responders to octreotide.

We report the comparative efficacy of a somatostatin receptor 1 and 5 subtypes (SSTR2 and SSTR5), and dopamine D2 (DAD2) compound, BIM-23A760, in suppressing GH secretion, in cell culture from human GH-secreting tumors, from patients partially responsive to long-term treatments with octreotide or lanreotide. In 18 tumors tested, the SSTR2, SSTR5, and DAD2 mRNAs were coexpressed. The SSTR2-selective analog, BIM-23197, the SSTR5-selective analog, BIM-23268, and the dopamine (DA) analog, BIM-53097, produced a mean maximal suppression of GH secretion (24 +/- 3, 20 +/- 3, and 20 +/- 3%, respectively) that was similar to that obtained with octreotide (23 +/- 3%). Nevertheless, based on individual responses, 60% of the tumors were mostly sensitive to the SSTR2 analog while 19 and 21% of the tumors were mainly responsive to the SSTR5 analog and to the DA analog, respectively. Among a series of new chimeric compounds that bind the SSTR2, SSTR5, and DAD2 receptors with variable affinities, BIM-23A760 produced greater maximal suppression of GH secretion than octreotide (38 +/- 2 vs 24 +/- 2%; p<0.03). The EC50 for BIM-23A760 was 2 pmol/l. In the presence of sulpride, the dose response inhibition of GH secretion by the trihybrid molecule, BIM-23A760, was partially reversed. The trihybrid produced also a maximal suppression of PRL greater than octreotide (74 +/- 5 vs 46 +/- 11%). When SSTRs pan inhibitors such as BIM-23A779 (binding affinity for SSTR1, SSTR2, SSTR3, SSTR5, respectively: 2.5, 0.3, 0.6, 0.6 nmol/l) or SOM230 were tested for their suppressive effects on GH secretion, they were less potent than the previous dopastatin hybrid molecule. After a brief exposure to a SSTR2-selective analog, BIM-23197, or to a DA analog, BIM-53097, the maximal GH suppression was achieved during 12 h. Under exposure to BIM-23A760, in the same conditions, maximal suppression of GH secretion lasted for 24 h. Such a longer biological effect, yet not explained, probably participates in the higher efficacy of BIM-23A760. The higher efficacy of BIM-23A760 is, at least partially, linked to its high affinity for the SSTR2 receptor subtype (IC50: 3 pmol/l). As compared to the dopastatin compound, the lower efficacy of the universal somatostatin ligands in the inhibition of GH secretion of GH-secreting tumors argues for the use of drugs targeted, according to specific receptors expression and functionality which may vary among the various classes of tumors.

Demonstration of enhanced potency of a chimeric somatostatin-dopamine molecule, BIM-23A387, in suppressing growth hormone and prolactin secretion from human pituitary somatotroph adenoma cells.

In acromegaly, the combination of somatostatin (SS) and dopamine (DA) agonists has been shown to enhance suppression of GH secretion. In the present study, a new chimeric molecule, BIM-23A387, which selectively binds to the SS subtype 2 receptor (sst(2); K(i) = 0.10 nM) and to the DA D2 receptor (D2DR; K(i) = 22.1 nM) was tested in cultures prepared from 11 human GH-secreting tumors for its ability to suppress GH and prolactin (PRL) secretion. The chimeric compound was compared with individual sst(2) and D2DR agonists of comparable activity at the individual receptors. All tumors expressed both sst(2) and D2DR mRNAs (0.8 +/- 0.2 and 4.7 +/- 0.7 copy/copy beta-glucuronidase mRNA, respectively). In cell cultures from seven octreotide-sensitive tumors, the maximal inhibition of GH release induced by the individual sst(2) and D2DR analogs and by BIM-23A387 was similar. However, the mean EC(50) for GH suppression by BIM-23A387 (0.2 pM) was 50 times lower than that of the individual sst(2) and D2DR analogs, either used individually or combined. Similar data were obtained in four tumors that were only partially responsive to octreotide. The inhibition of GH release by BIM-23A387 was only partially reversed by the D2R2 antagonist, sulpiride, or by the sst(2) antagonist, BIM-23454. Only when both antagonists were combined was the GH suppressive effect of BIM-23A387 totally reversed. Finally, BIM-23A387 produced a mean 73 +/- 6% inhibition of PRL in six mixed GH plus PRL tumors. These data demonstrate an enhanced potency of the chimeric molecule, BIM-23A387, in suppressing GH and PRL secretion from acromegalic tumors, which cannot be explained merely on the basis of binding affinity for SS and/or DA receptors.

A luteinizing hormone-, alpha-subunit- and prolactin-secreting pituitary adenoma responsive to somatostatin analogs: in vivo and in vitro studies.

OBJECTIVE: Evaluation of the efficiency of somatostatin analogues in the treatment of a mixed luteinizing hormone (LH)-, alpha-subunit-, prolactin (PRL)-secreting pituitary adenoma. DESIGN: A 30-year-old woman, with amenorrhaea-galactorrhaea, presented with a pituitary macroadenoma. The endocrine evaluation showed high plasma levels of PRL, LH, and alpha-subunit inhibited by 65%, 65% and 33% respectively under octreotide test (200 microg, s.c.). Long-term treatment with slow release (SR) lanreotide (30 mg/10 days, i.m.) restored menstrual cycles and normalized PRL values. Due to persisting supranormal levels of LH and alpha-subunit, and to the absence of tumoral shrinkage, the adenoma was resected by the transsphenoidal route. METHODS: In vitro characterization of the somatostatin receptor subtypes (SSTR) expression and functionality. Real-time polymerase chain reaction was performed to quantify the expression of SSTR mRNAs and functionality of the SSTRs was assessed in cell culture studies with various concentrations of native somatostatin (SRIF-14) and of analogues preferential for SSTR2 or SSTR5. RESULTS: This adenoma presented with high levels of SSTR2, SSTR3 and SSTR5 mRNAs, as compared with a series of gonadotroph adenomas. In cell culture studies, PRL, LH and alpha-subunit were inhibited by 60%, 47% and 33% respectively by SRIF-14 at a concentration of 10 nmol/l. The SSTR2 (BIM-23197, lanreotide) and SSTR5 (BIM-23268) preferential analogues both produced a partial 21-38% inhibition of PRL, LH, and alpha-subunit release. DISCUSSION: In this plurihormonal-secreting adenoma, the high efficacy of somatostatin analogues to inhibit PRL, LH and alpha-subunit secretion in vivo may be explained by the unusually high level of expression and by the functionality of both SSTR2 and SSTR5 receptor subtypes.

Bim-23244, a somatostatin receptor subtype 2- and 5-selective analog with enhanced efficacy in suppressing growth hormone (GH) from octreotide-resistant human GH-secreting adenomas.

Although both somatostatin receptor subtype 2 (SSTR2) and SSTR5 messenger ribonucleic acid (mRNA) are consistently expressed in GH-secreting adenomas, SSTR2 has been believed to be the key modulator of somatostatin-mediated inhibition of GH release. The somatostatin agonists currently in clinical use, octreotide and lanreotide, are directed mainly to SSTR2 (IC(50) 12- to 18-fold higher than for SSTR5). Recently, however, it was demonstrated that an SSTR5 preferential agonist, BIM-23268, not only suppressed PRL release from prolactinomas and mixed GH-PRL adenomas, but also inhibited GH release in about half of GH adenomas. In addition, the SSTR5-preferring analog showed a slight additive effect when used in combination with SSTR2 preferential drugs at submaximal concentrations in octreotide partially sensitive adenomas. In the present study we quantified SSTR2 and SSTR5 mRNA expression and the GH-suppressive effects of somatostatin-14; octreotide; a SSTR2-preferential compound, BIM-23197; a SSTR5-preferential compound, BIM-23268; and a new SSTR2- and SSTR5-bispecific compound, BIM-23244, in GH-secreting tumors classified as either full responders to octreotide (n = 5) or partially sensitive to octreotide (n = 5). The octreotide-sensitive GH secretory adenomas presented with a high level of both SSTR2 and SSTR5 mRNA expression [222 +/- 61 and 327 +/- 136 pg/pg glyceraldehyde-3-phosphate dehydrogenase (GAPDH), respectively]. In these tumors the suppression of GH release was similarly achieved at picomolar ranges by octreotide, BIM-23197, and BIM-23244 (EC(50) = 25 +/- 15, 3 +/- 2, and 3 +/- 3 pmol/L, respectively). The compounds preferential for only SSTR5 were unable to inhibit GH release in such tumors. Among the octreotide partially responsive tumors, SSTR2 mRNA expression was 9-fold lower than in the octreotide-sensitive tumors (25 +/- 12 vs. 222 +/- 61 pg/pg GAPDH; P < 0.015), whereas SSTR5 mRNA expression was approximately 7-fold higher than in the octreotide-sensitive tumors (2271 +/- 1197 pg/pg GAPDH). In these octreotide partially responsive tumors, the SSTR5-preferential compound, BIM-23268, and the SSTR2- and SSTR5-bispecific compound, BIM-23244, were quite effective in suppressing GH secretion (EC(50) = 25 +/- 13 and 50 +/- 31 pmol/L, respectively). Similarly, BIM-23244, was able to suppress by 51 +/- 5% PRL release from five mixed GH- and PRL-secreting adenomas. These data indicate that due to heterogeneous expression of SSTR2 and SSTR5 receptor subtypes, in GH-secreting tumors, a bispecific analog, such as BIM-23244, that can activate both receptors could achieve better control of GH hypersecretion in a larger number of acromegalic patients.

Combined pituitary hormone deficiency due to the F135C human Pit-1 (pituitary-specific factor 1) gene mutation: functional and structural correlates.

The pituitary-specific transcription factor Pit-1 (pituitary-specific factor 1) is known to play a key role in the differentiation of PRL-, GH-, and TSH-secreting cells, and in the regulation of expression of the corresponding genes. In recent years, 12 distinct mutations of the Pit-1 gene have been shown to be responsible for a phenotype of multiple congenital pituitary hormone deficiency involving PRL, GH, and TSH. We had previously identified, in four siblings with GH, PRL, and TSH deficiencies, a mutation (F135C) resulting in a single amino acid change within the POU-specific binding domain of the Pit-1 molecule. In the present report, we have explored the functional effect of the F135C mutation. In vitro activity tests performed by transfection in human HeLa cells showed decreased transactivation capacity on the PRL, GH, and Pit-1 genes. The DNA binding experiments performed by gel shift showed that the F135C mutation generated a protein capable of binding to DNA response elements. To analyze how the F135C mutation might affect functionality of the transcription factor despite a normal DNA binding, we used a structure modelization approach and also analyzed two other Pit-1 mutant proteins (F135A and F135Y). The loss of functionality in these two mutants was similar to that of F135C. This finding was in keeping with our molecular modeling studies. According to structural data derived from the crystallographic analysis of the DNA/Pit-1 POU domain complex, the conformation of the first helix of the F135C-mutated POU-specific domain could be perturbed to such an extent that any interaction with other transcription cofactors might be definitively prevented.

Human somatostatin receptor subtypes in acromegaly: distinct patterns of messenger ribonucleic acid expression and hormone suppression identify different tumoral phenotypes.

Recently, studies using somatostatin (SRIF) analogs preferential for either the SRIF receptor 2 (SSTR2) or the SSTR5 subtype demonstrated a variable suppression of GH and PRL release from GH-secreting human adenomas. These data suggested the concept of SSTR subtype specificity in such tumors. In the present study the quantitative expression of messenger ribonucleic acid (mRNA) for the 5 SSTR subtypes and the inhibitory effects of SRIF14; SRIF28; octreotide; the SSTR2-preferential analog, BIM-23197; and the SSTR5-preferential analog, BIM-23268, on GH and PRL secretion were analyzed in cells cultured from 15 acromegalic tumors. RT-PCR analysis revealed a consistent pattern of SSTR2 and SSTR5 mRNA expression. SSTR5 mRNA was expressed at a higher level (1052 +/- 405 pg/pg glyceraldehyde-3-phosphate dehydrogenase) than SSTR2 mRNA (100 +/- 30 pg/pg glyceraldehyde-3-phosphate dehydrogenase). However, only SSTR2 mRNA expression correlated with the degree of GH inhibition induced by SRIF14, SRIF28, and BIM-23197. The SSTR5-preferential compound inhibited GH release in only 7 of 15 cases. In cells cultured from the 10 mixed adenomas that secreted both GH and PRL, RT-PCR analysis revealed a consistent coexpression of SSTR5, SSTR2, and SSTR1 mRNA. In all cases SRIF14, SRIF28, and the SSTR5-preferential analog, BIM-23268, significantly suppressed PRL secretion, with a mean maximal inhibition of 48 +/- 4%. In contrast, the SSTR2-preferential analogs, BIM-23197 and octreotide, were effective in suppressing PRL in only 6 of 10 cases. In cells cultured from adenomas taken from patients partially responsive to the SRIF analog, octreotide, partial additivity in suppressing both GH and PRL secretion was observed when the SSTR2- and SSTR5-preferring analogs, BIM-23197 and BIM-23268, were tested in combination. Our data show a highly variable ratio of the SSTR2 and SSTR5 transcripts, according to tumors. The SSTR2-preferring compound consistently inhibits GH release, whereas the SSTR5-preferring compound is the main inhibitor of PRL secretion. When both drugs are combined, the partial additivity observed in mixed GH- plus PRL-secreting adenomas may be of interest in the therapeutic approach of such tumors.

Quantitative and functional expression of somatostatin receptor subtypes in human prolactinomas.

Recently, it was demonstrated that somatostatin analogs preferential for the SSTR5 subtype suppress PRL release from prolactinoma cell cultures by 30-40%. These data supported the idea of somatostatin receptor subtype-specific control of PRL secretion in such tumors. The present study examines the quantitative profile of SSTRs messenger ribonucleic acid (mRNA) in 10 PRL-secreting tumors and correlates the expression with the ability of native somatostatins (SS14 and SS28), SSTR2 preferential analogs (octreotide and BIM-23197), and the SSTR5 preferential analog BIM-23268 to suppress PRL secretion. RT-PCR quantitative analysis showed a large predominance of SSTR5 mRNA [5648 +/- 1918 pg/pg glyceraldehyde-3-phosphate dehydrogenase (GAPDH)] vs. SSTR2 mRNA (148 +/- 83 pg/pg GAPDH). The SSTR1 transcript was also highly expressed in prolactinomas (1296 +/- 669 pg/pg GAPDH). SSTR5 mRNA expression correlated with PRL inhibition induced by both SRIF14 and SRIF28. Among the different analogs tested, only BIM-23268 produced inhibition of PRL release similar to that achieved with the native peptides. Its EC50 for PRL suppression was 0.28 +/- 0.10 nmol/L. No additive effects on PRL suppression were achieved by cotreatment of the tumor cells with SSTR2 and SSTR5 preferential analogs. In the same tumor cell cultures, quinagolide, a potent dopamine agonist, produced a dose-dependent inhibition of PRL with an EC50 at least 10 times lower than that of BIM-23268. Coincubation of quinagolide and BIM-23268, particularly in tumor cells resistant to dopamine agonist treatment, did not produce additive effects on PRL suppression. In conclusion, prolactinomas have a specific pattern of SSTR subtype mRNA expression (SSTR5 and SSTR1). SSTR5 expression is correlated to PRL regulation. These inhibitory effects are superimposable, at a higher concentration, to those of the dopamine agonists, but are not additive, particularly in the adenomas resistant to dopaminergic suppression of PRL release.

Impact of gsp oncogene on the expression of genes coding for Gsalpha, Pit-1, Gi2alpha, and somatostatin receptor 2 in human somatotroph adenomas: involvement in octreotide sensitivity.

The impact of the gsp oncogene on the expression of genes engaged in the somatotroph cell phenotype remains poorly understood in human somatotroph adenomas. As the gsp oncogene is associated with an increased octreotide (somatostatin agonist) sensitivity, a group of 8 somatotroph adenomas bearing the gsp mutation (gsp+) and another group of 16 adenomas without the mutation (gsp-) were analyzed, all of them presenting variable octreotide sensitivities. The expressions of genes encoding for G(s)alpha, Pit-1, G(i2)alpha, and SSTR2, involved in the regulation of secretory activity in somatotroph cells, were assessed by Northern blot. A decreased expression of the G(s)alpha gene was found in gsp + tumors, suggesting the existence of a negative feedback of the oncogenic protein upon its own messenger ribonucleic acid (mRNA). In contrast, G(i2)alpha, Pit-1, and GH messengers were not significantly different in the groups. A positive correlation between the in vitro and in vivo GH octreotide-induced secretory inhibition and the expression of SSTR2 mRNA was found. However, the expression of the gene for SSTR2 appeared not to be different between gsp + and gsp-, even when the octreotide sensitivity was significantly higher in the adenomas carrying the mutation. Interestingly, the SSTR2 gene expression was significantly correlated to those of G(i2)alpha and Pit-1. In the same way, the G(s)alpha mRNA expression was positively correlated with those of Gi2alpha and Pit-1. Such correlations strongly suggest a concerted dysregulation of the expression of these genes in both categories of adenomas. The loss of the octreotide sensitivity represents one aspect of the dysregulation process that partially results from the decreased SSTR2 expression. However, the improvement of the sensitivity associated with the presence of the gsp oncogene seems to proceed in a way different from SSTR2 expression.

Expression of functional growth hormone secretagogue receptors in human pituitary adenomas: polymerase chain reaction, triple in-situ hybridization and cell culture studies.

We examined the expression of functional growth hormone secretagogue receptors (GHS-R) in a series of 30 human pituitary adenomas-six secreting GH, three GH-PRL, six prolactin (PRL), five adrenocorticotrophic hormone (ACTH), one thyroid stimulating hormone (TSH), four gonadotroph and five non-secreting adenomas. By reverse transcriptase polymerase chain reaction (RT-PCR), the coexpression of the two GHS-R isoforms (Ia and Ib) was found in all the GH-, GH-PRL- and PRL-secreting adenomas, and only in two out of three corticotroph, two out of four gonadotroph and one out of five non-secreting tumours. They were absent in the TSH-secreting adenoma. The PCR products of GHS-R Ia and Ib were identical in size to those from two normal pituitaries. PCR cloning and sequencing of isoforms performed in two somatotroph adenomas revealed only two single, silent base mutations. Triple in-situ hybridization showed colocalization of GHS-R mRNA with messengers of GH and PRL, conjointly or separately, in individual cells of somatotroph, mammosomatotroph, and lactotroph adenomas. The presence of GHS-R mRNA in cells expressing PRL mRNA is emphasized. In cultured cells from six somatotroph and two mammosomatotroph adenomas, the powerful GHS MK-0677 stimulated GH release in a dose-dependent manner, with maximal effect at 6 h. Contrarily, when GHRH was applied, only three somatotrophs and two mamosomatotrophs were stimulated. In the two mammosomatotrophs, the PRL response to MK-0677 and to GHRH was similar to the GH response. An homologous desensitization of the GHS-R and the GHRH receptor was observed 24 h after a first stimulation by a single dose of the corresponding agonist. Heterologous desensitization was not observed. Interestingly, MK-0677 also stimulated, in a dose-dependent way, the hormone release of cells from all tested lactotroph and corticotroph adenomas. The existence of a functional expression of GHS-R in somatotroph, mammosomatotroph, lactotroph and corticotroph adenomas rises the question of the role played by GHS-R in pituitary adenomas, particularly those not engaged in GH secretion.

Expression of the members of the Ptx family of transcription factors in human pituitary adenomas.

A number of putative transcription factors described in the pituitary have been implicated as key elements in the processes that direct pituitary development. Three recently described proteins, Ptx1, Ptx2, and Ptx3, define a new family of transcription factors, the Ptx subfamily, within the paired-like class of homeodomain factors. In mice, Ptx1 and Ptx2 gene expression has been detected in the area of the pituitary primordium and is maintained throughout development in Rathke pouch and adult pituitary. In the present study, the expression of the Ptx1, Ptx2, and Ptx3 genes was characterized in the normal human pituitary and in the different types of human pituitary adenomas. Although no Ptx3 gene expression could be detected in these tissues, Ptx1 presented with a quite ubiquitous pattern of distribution, being expressed at quite constant levels in normal tissues and in all 60 pituitary tumors analyzed. The pattern of expression of the Ptx2 gene among the different subsets of pituitary adenomas was even more varied. No Ptx2 expression could be detected in corticotroph tumors. In contrast, high levels of Ptx2 messenger ribonucleic acid were measured in the gonadotroph tumors, although no specific correlation to other markers of the gonadotroph lineage differentiation, such as alphaGsu, LHbeta, or FSHbeta, could be evidenced. Finally, Ptx2 was also expressed in pure lactotroph adenomas and not in somatotroph adenomas. Ptx2 is, therefore, the first paired homeodomain pituitary transcription factor differentially expressed in these two lineages, which derive from a common precursor. These results support a role for Ptx2 in the terminal differentiation of somatotroph and lactotroph cell phenotypes.

Pronostic and therapeutic consequences of Gs alpha mutations in somatotroph adenomas.

Human pituitary somatotroph adenomas can be associated with mutations of the s alpha-subunit of G proteins. However, the impact of the gsp mutations on the tumoral phenotype is not well understood at present. This study aims to determine whether the detection of this mutation could impact on the management of acromegalic patients. We examined 30 acromegalic patients; 8 were gsp positive, and 22 were gsp negative. The gsp-positive adenomas appeared to secrete significantly more when the ratio of basal GH level/tumor size was considered. A better octreotide sensitivity of mutated adenomas was clearly shown under in vivo (short and long term) and in vitro conditions. During the acute octreotide test, the GH nadir was significantly lower in the gsp-positive adenomas (85% of maximal inhibition vs. 52%). Eighteen patients were treated with octreotide (300 micrograms/day) for at least 3 months before surgery: the percent inhibition of GH hypersecretion was higher in gsp-positive adenomas (76% vs. 47%). In cell culture, the octreotide-induced inhibition of GH release was significantly higher in gsp-positive adenomas (71% vs. 30%). Finally, during 2 yr of postoperative follow-up, GH hypersecretion was controlled in all patients with gsp mutation even in those in whom tumoral tissue remained after surgery. On the contrary, in the gsp-negative group, octreotide treatment was unable to control hypersecretion in 4 patients bearing tumoral remnants. The Gs alpha mutation could, therefore, be a new marker to foresee the susceptibility of the tumor to be controlled by somatostatin analogs, which improves prognosis.

Abnormal transduction mechanisms in pituitary adenomas.

Pituitary adenomas are differentiated tumors expressing their appropriate mature hormone. Tumoral cells sometimes present with a defective physiological inhibitory or stimulatory control, resulting in paradoxical responses or nonresponsiveness to regulatory neurohormones. These abnormalities can be explained by defects at the intracellular transduction mechanism level. Knowledge of these defective pathways has made progress in the understanding of the pathogenesis of pituitary adenomas possible. The discovery of mutations of Gs alpha named gsp oncogenes in 40% of human somatotropinomas represents one of the most important advances in this field. Other molecular alterations were identified but are rare and sporadic and the pathogenesis of pituitary adenomas remains largely unknown. Abnormal transduction mechanisms may also result in a variable sensitivity of tumors to pharmacological therapy. The dopamine agonist, bromocriptine, is able to normalize blood PRL levels and to reduce tumor size in the majority of patients with prolactinoma, but is ineffective in 8-15% of them. Under physiological conditions, PRL secretion is under the tonic inhibitory control of dopamine which binds D2 receptors negatively coupled to adenylyl cyclase. Several defects in the dopaminergic transduction pathways participate in this bromocriptine resistance. The mean D2-binding site density is decreased to 50% as compared to responsive tumors. This loss of D2 receptors can account for a lower transcription level of its gene and is accompanied by modifications in the messenger alternative splicing; the D2 short isoform receptor expression decreases preferentially. A reduction in Gi2 alpha protein expression is also observed and is correlated to that of the D2 receptor. Finally, the pituitary-specific transcription factor Pit-1 expression is affected. A highly significant correlation was seen between the D2 receptor mRNA and Pit-1 mRNA levels. These defects observed on many levels of the dopaminergic transduction cascade may be the first steps in the loss of the functional features of differentiated tumors toward more proliferative tumors.

Pit-1 gene expression in human pituitary adenomas.

The anterior pituitary-specific transcription factor Pit-1 (also known as GHF-1) was initially identified and cloned as a transactivator of the GH and PRL genes, and later as a regulator of the TSH beta gene. Analysis of Pit-1 expression during mouse embryogenesis revealed that initiation of its expression correlates both temporally and spatially with activation of its target genes. Immunocytochemical studies revealed a high expression of Pit-1 protein in the nuclei of only three cell types: somatotropes, lactotropes and thyrotropes. The importance of Pit-1 as a regulator of the anterior pituitary development has been further demonstrated by the absence of somatotropes, lactotropes and thyrotropes in the pituitary glands of Pit-1-defective mice and humans. Since Pit-1 is required for both cell phenotype and proliferation, one may ask if this transcription factor might be associated with development of pituitary tumors. Several investigators have reported Pit-1 gene expression in human pituitary adenomas. These studies, while not in total agreement, show that pituitary tumorigenesis does not seem to be associated with a gross alteration of Pit-1 gene expression in humans. Pit-1 transcripts, identical in size and sequence to those observed in normal pituitary, were described in human GH-, PRL- and TSH-secreting pituitary adenomas and in most cases the presence of Pit-1 transcripts correlated with the localization of Pit-1 protein. The biological relevance of Pit-1 expression reported in some nonfunctioning adenomas remains to be clarified. As expression of the PRL and GH genes is ultimately confined to distinct lactotropic and somatotropic populations despite the presence of Pit-1 protein in both cell types, there must be additional mechanisms that control the cell-specific activation of the PRL and GH promoters. The Pit-1 beta isoform, raised through alternative splicing of exon 2 of the Pit-1 gene, is a more potent inducer of GH transcription than the major Pit-1 form. This form could, at least in part, account for the cell-specific activation of the PRL and GH genes. Pit-1 beta was invariably found present in all the tumors expressing the Pit-1 major form, no significant difference in the Pit-1 beta/Pit-1 expression ratio being observed between tumors identified as pure GH- or PRL-producing tumors. This lack of selectivity together with its low level of expression is therefore not in favor of a key role for the beta-isoform in the cell type-specific expression of the GH and PRL genes in humans. The failure of somatotropes, lactotropes and thyrotropes to proliferate in Pit-1-defective mice and humans indicates that Pit-1 might be competent to activate genes required for cell proliferation or survival of the three cell types. Recent data indeed suggest that Pit-1 may directly or indirectly regulate somatotropes and lactotropes through activation of the receptors for GRF and SRIF on the one hand, and dopamine on the other hand. Such regulatory mechanisms could contribute to the differentiation of the somatomammotropic lineage in fully differentiated somatotropic and lactotropic cells.

[Homeoproteins and pituitary adenoma]. / Homéoprotéines et adénomes hypophysaires.

Several transactivating factors specifically involved in the differentiation and proliferation of anterior pituitary cell types have been recently identified. Among them Pit-1 a member of the POU-domain transcription factors family is specific of anterior pituitary cells, and was initially identified and cloned as a transactivator of the GH and PRL genes and as a regulator of the TSH beta gene. Pit-1 play a key role during embryogenesis in the differentiation and proliferation of somatotrophs, lactotrophs and thyreotrophs. The importance of Pit-1 as a regulator in the anterior pituitary development has been further demonstrated by naturally occurring mutations or delections in dwarf mouse strains. In the Snell and Jackson dwarf mice, the levels of Pit-1 gene expression are low or undetectable, GH, PRL and TSH beta gene expression are absent and lactotrophs, somatotrophs, and threotrophs fail to proliferate. Furthermore Pit-1 carries out similar functions in humans. This is supported by the fact that children with mutations of the Pit-1 gene present with a congenital combined GH, PRL and TSH deficiency analogous to the phenotype of the Snell and Jackson dwarf mice. In children who were born to healthy consanguinous parents and present such combined deficiencies we recently reported a Pit-1 mutation causing a transition from a Phe to a Cys in a region of the protein known to be involved in DNA binding. Pit-1 transcripts identical in size and sequence to those observed in normal pituitary were described in human GH, PRL and TSH secreting pituitary adenomas. The Pit-1 beta isoform, raised through alternative splicing of exon 2 of the Pit-1 gene, is a more potent inducer of GH transcription than the major Pit-1 form. However no difference in the level of expression of the different Pit-1 isoforms was observed between tumors identified as pure GH or PRL producing tumors. The results support the existence of other transcription factors interacting with Pit-1 to coordinately regulate the activity of the GH and PRL promotors in a cell specific manner. In contrast, variable Pit-1 expression was observed in prolactinomas, according to their sensitivity to bromocriptine treatment. A highly significant correlation was indeed evidenced between the D2 receptors mRNA and the Pit-1 mRNA levels. These results raise the possibility that Pit-1 may either directly or indirectly affect the transcription of the D2 dopaminergic receptor gene. In fact, receptors for other hypothalamic neurohormones such a GHRH and somatostatin are known to be potential Pit-1 target genes. Such mechanisms could be implicated in the differentiation and proliferation of lactotrophs and somatotrophs.

Pit-1 gene expression in human lactotroph and somatotroph pituitary adenomas is correlated to D2 receptor gene expression.

The expression of the pituitary-specific transcription factor Pit-1 gene was analyzed in a series of 30 human lactotroph and somatotroph pituitary tumors. Northern blot analysis failed to reveal any quantitative differences in Pit-1 gene expression between somatotroph and lactotroph tumors, and reverse transcription-PCR analysis showed similar patterns of Pit-1 isoforms expression in both populations of tumors. The expression of the D2 receptor gene was subsequently analyzed in the same adenomas. In the prolactinomas, which presented with a variable sensitivity to dopamine agonist treatment, the intensity of the D2 receptor transcripts (2.8 kilobases) was variable and was related to the sensitivity to the dopamine agonist treatment. Notably, the individual D2 receptor messenger ribonucleic acid (mRNA) levels were highly correlated to the Pit-1 mRNA levels measured in the same tumors (r = 0.90; P < 0.0001). In the GH-secreting tumors, a significant expression of the D2 receptor gene was evidenced by Northern blot in all mixed somato-lactotroph adenomas and in some of the pure somatotroph adenomas; again, a positive correlation was found between D2 mRNA and Pit-1 mRNA levels (r = 0.68; P < 0.01). These results suggest the existence of mechanisms responsible for a coordinate control of Pit-1 and D2 receptor genes that remain to be determined.

A new mutation of the gene encoding the transcription factor Pit-1 is responsible for combined pituitary hormone deficiency.

The pituitary-specific transcription factor Pit-1/GHF1 regulates the expression of PRL, GH, and TSH beta genes through binding to specific regions of the promoters of these genes. Mutations of the Pit-1 gene have been shown to be responsible for a syndrome of combined pituitary hormone deficiency (CPHD), including complete GH and PRL deficiencies and central hypothyroidism. We studied four siblings presenting with CPHD born to healthy consanguinous parents. All four affected children had complete GH deficiency diagnosed in early childhood. They later developed hypothyroidism and were found to have undetectable PRL levels. The pituitary gland was hypoplastic at magnetic resonance examination in one of the patients. Amplification of genomic DNA and subsequent sequencing of the six exons of the Pit-1 gene allowed identification in the four patients with CPHD of an as yet undescribed mutation in exon 3. A substitution of T go G induced a change from a Phe to a Cys residue at position 135 within the hydrophobic core of the POU-specific DNA-binding domain of the Pit-1 protein. All affected children were homozygous for the mutation, whereas the mother was heterozygous, suggesting a recessive mode of inheritance. Molecular studies in other affected families will allow instructive genotype-phenotype correlations concerning the Pit-1 gene.

Pit-1 gene expression in the human pituitary and pituitary adenomas.

The cellular basis for pituitary neoplasia is poorly understood. The POU domain protein Pit-1 is a pituitary-specific transcription factor involved in the generation, differentiation, and proliferation of three pituitary cell types: lactotrophs, somatotrophs, and thyrotrophs. In this study, we analyzed the expression of Pit-1 gene in a series of 15 different human pituitary tumors and compared it with that observed in normal tissue. Pit-1 transcripts, identical in size (2.4 and 4.5 kilobases) and sequence to those observed in normal tissue were evidenced in PRL-, GH-, and TSH-secreting tumors. Pit-1 is overexpressed (2.5- to 5-fold) in the PRL- and GH-secreting tumors, but to an extent consistent with the predominant cellular type of these adenomas. An isoform of Pit-1, with an insertion of 26 amino acids in the trans-activation domain as a result of alternative splicing, is also present in both normal and tumoral tissues. It is concluded that human pituitary tumorigenesis does not seem to be associated with a gross alteration of Pit-1 gene expression.

Polymorphism of thyrotropin and alpha subunit in human pituitary adenomas.

To understand better why patients with TSH-secreting pituitary tumors exhibit variable degree of hyperthyroidism, we analyzed the various isoforms of TSH and alpha-subunit secreted by 4 TSH-secreting adenomas in primary culture. All patients had macrodenomas clinically associated with hyperthyroidism with normal to elevated TSH plasma levels. The in vivo molar alpha/TSH ratio ranged from 18.4 to 3.8. The hormone material secreted over 4 to 48 h in culture was separated by gel isoelectrofocusing, eluted and estimated by immunoassays. The release of free alpha-subunit was noticeably different among adenomas. Three tumors were found to release an homogeneous and acidic (pI = 5.4-4.5) species totally unrelated to the alpha-subunit dissociated from intrapituitary TSH (5 isoforms, pI = 8.8-5.8) while another was more heterogeneous (pI = 8.8, 8.4, 7.6, 6.8, 5.8, 5.4-4.5). Tumoral TSH exhibited at least six detectable isoforms (pI = 8.6, 8.3-8.0, 7.5, 7.0, 6.5, 6.0) very similar to those present in a purified intrapituitary hormone preparation. While intrapituitary TSH was composed of 70% of alkaline (pI = 8.6-7.5), 25% of neutral (pI = 7.0-6.0) and 5% (pI = 5.8-4.5) of acidic forms, these species were found to be more evenly distributed in adenomatous secretion (43%/42%/15%). The TSH-secreting tumors thus appeared to relase preferentially neutral and acidic forms of TSH than alkaline components but for one tumor, this ratio could be modified by chronic incubation with TRH. When assayed for their capacity to stimulate 3H-thymidine incorporation in FRTL-5 cells, neutral TSH appeared definitely less potent than the alkaline and acidic isohormones. Altogether, these data show that pituitary adenomas synthesize normal forms of TSH but release them in variable amount in the medium. When circulating in the blood, the ratio between active and inactive isoforms of TSH may thus be responsible for the variable stimulation of the thyroid gland observed in the patients.