Imprinting of the G_sα gene GNAS1 in the pathogenesis of acromegaly

Approximately 40% of growth hormone–secreting pituitary adenomas have somatic mutations in the GNAS1 gene (the so-called gsp oncogene). These mutations at codon 201 or codon 227 constitutively activate the α subunit of the adenylate cyclase–stimulating G protein G_s. GNAS1 is subject to a complex pattern of genomic imprinting, its various promoters directing the production of maternally, paternally, and biallelically derived gene products. Transcripts encoding G_sα are biallelically derived in most human tissues. Despite this, we show here that in 21 out of 22 gsp-positive somatotroph adenomas, the mutation had occurred on the maternal allele. To investigate the reason for this allelic bias, we also analyzed GNAS1 imprinting in the normal adult pituitary and found that G_sα is monoallelically expressed from the maternal allele in this tissue. We further show that this monoallelic expression of G_sα is frequently relaxed in somatotroph tumors, both in those that have gsp mutations and in those that do not. These findings imply a possible role for loss of G_sα imprinting during pituitary somatotroph tumorigenesis and also suggest that G_sα imprinting is regulated separately from that of the other GNAS1 products, NESP55 and XLαs, imprinting of which is retained in these tumors.

The GNAS1 gene on chromosome 20q13 encodes the α subunit of the heterotrimeric guanosine 5′triphosphate–binding (GTP-binding) protein, G_s. G_s couples hormonal stimulation of various cell-surface receptors to the activation of adenylate cyclase. A number of endocrine syndromes are caused by heterozygous mutations that affect G_sα. McCune-Albright syndrome (polyostotic fibrous dysplasia, skin hyperpigmentation, and autonomous endocrine hyperfunction) results from somatic mosaicism for a codon 201 mutation, which constitutively activates G_sα by impairing its intrinsic GTPase activity (1). Codon 201 mutations are also found in some endocrine tumors (2) (see below).

In contrast, germline mutations that inactivate G_sα result in the autosomal dominant syndrome Albright’s hereditary osteodystrophy (AHO; characterized by short stature, brachydactyly, and subcutaneous calcification) (3, 4). However, when such mutations are maternally inherited, in addition to AHO, resistance to the peripheral action of several hormones, notably parathyroid hormone (PTH), results (pseudohypoparathyroidism type Ia; PHP-Ia). It was this parental origin effect in AHO/PHP that first suggested that GNAS1 might be an imprinted gene (5).

Despite this indirect evidence for G_sα imprinting, in both AHO and PHP-Ia the activity of G_sα in erythrocytes is 50% of normal (6, 7), and RNA studies of several human fetal tissues showed biallelic GNAS1 expression (8). It is therefore believed that G_sα is monoallelically expressed only in a few sites, particularly the proximal renal tubule, the main PTH target tissue. This would be consistent with the fact that in PHP-Ia patients, renal cAMP response to PTH is absent or severely impaired (9), suggesting target cell G_sα levels much less than the 50% seen in erythrocytes.

Molecular studies have shown that both human GNAS1 and murine Gnas are indeed imprinted, but in a highly complex way. In addition to G_sα, GNAS1 encodes two other proteins, extra-large α_s (XLαs) and 55-kDa neuroendocrine secretory protein (NESP55). All three mRNA species include exons 2–13, spliced in each case to a distinct first exon. NESP55 transcripts are maternal, while XLαs transcripts, initiating 14 kb further downstream, are paternal (10, 11). A spliced, paternally expressed, noncoding antisense transcript arises approximately 2.5 kb upstream of XLαs and traverses the NESP55 exon (12). G_sα transcripts, initiating at exon 1, approximately 37 kb downstream of XLαs, are biallelic. Finally, a fourth sense promoter, approximately 2 kb upstream of exon 1, directs expression of paternal transcripts that are probably not translated, since they contain an alternative first exon (1A or A/B) lacking an initiation codon (13–15).

In humans, monoallelic expression of G_sα transcripts has not been demonstrated. In the mouse, though, G_sα does appear to be predominantly maternal in origin in renal cortex and in brown and white fat. In these tissues, G_sα protein and mRNA levels are much lower in mice with a heterozygous knockout mutation of the maternal Gnas allele, compared with those with a paternal knockout (16).

While G_sα is ubiquitous, NESP55 and XLαs are highly expressed only in certain neurosecretory tissues, including pituitary (17, 18). XLαs, like G_sα, can bind G βγ subunits and activate adenylate cyclase (17, 19), but a hormone receptor able to activate XLαs has not been identified. The function of NESP55 is unclear; it was originally suggested to be the proteolytic precursor for certain bioactive peptides (20).

Approximately 40% of pituitary somatotroph adenomas contain heterozygous mutations in GNAS1 that constitutively activate G_sα by substitutions at arginine 201 (R201) or glutamine 227 (Q227) (2, 21–23). Less frequently, the same mutations occur in thyroid tumors (24), Leydig cell tumors of the ovary and testis (25), nonsomatotroph (corticotroph or nonfunctioning) pituitary adenomas, pheochromocytomas, and parathyroid adenomas (26, 27). In Japan only 4–10% of somatotroph tumors have GNAS1 mutations (28, 29). The role of activated GNAS1 in neoplasia has earned it an alternative title — the gsp oncogene. In somatotroph tumors lacking a gsp oncogene, the factors involved in tumorigenesis are largely unknown (30).

G_sα-encoding transcripts are monoallelically expressed in the normal pituitary. We identified four adult pituitaries heterozygous for the exon 5 polymorphism (at codon 131 of G_sα). RT-PCRs were performed, each using the same downstream primer in exon 10 paired with a primer specific for one of the four alternative first exons. These reactions selectively amplify GNAS1 transcripts specific for one of the three protein-coding mRNAs (NESP55, XLαs, or G_sα) or the noncoding 1A transcript.

Figure 1 shows the sequence across the polymorphic site of cDNAs from two of these samples. NESP55 and XLαs are expressed from opposite alleles, as shown previously in a range of human fetal tissues, in all of which NESP55 is maternally and XLαs paternally derived (11). Exon 1A transcripts were shown recently to be paternally expressed in blood cells of two adults (15). We found similar paternal-specific expression of 1A transcripts in a range of first-trimester fetal tissues (data not shown). Figure 1 shows that the same is also true of adult pituitary.

Figure 1

Imprinting of GNAS1 transcripts in normal human pituitary. Sequences are shown for each of the four GNAS1 transcript types (G_sα, exon 1A, XLαs, and NESP55) across the site of the exon 5 A/G polymorphism (reverse strand). Results from two normal pituitaries (HN1 and HN3) are shown. In each case the polymorphic residue is indicated by an arrow. The G_sα transcripts are exclusively (HN3) or predominantly (HN1) maternal in origin, i.e., derived from the NESP55-expressing chromosome.

Examining G_sα transcripts, however, we found that in contrast to the biallelic expression seen in all human tissues examined previously, in the pituitary these transcripts are monoallelic. In each case, the expressed allele is the same as that represented in NESP55 transcripts. XLαs and exon 1A transcripts, in contrast, contain only the opposite allele. These results suggest that in the pituitary, G_sα transcripts, like NESP55, are derived from the maternal allele. The same result was obtained in all four adult pituitary cDNA samples; the maximum proportion of paternal G_sα transcript in any of these samples was 14%.

Somatic activating mutations in somatotroph adenomas occur on the maternal allele. Codons R201 and Q227, the sites of the gsp mutations, lie in exons 8 and 9 and hence are present in the XLαs and NESP55, as well as the G_sα, transcripts. Examination of maternal NESP55 and paternal XLαs transcripts for the presence of the gsp mutation should therefore permit the inference of which parental allele carries the mutation.

We examined 22 gsp⁺ adenomas in this way, including examples with mutations at codons 201 and 227, and also different mutations at each codon. From each tumor, we generated three RT-PCR products (specific for G_sα, XLαs, or NESP55) and sequenced across the gsp mutation site (Figure 2a). In 21 cases (15 R201C, one R201S, one R201H, three Q227L, and one Q227R) the gsp mutation was present in the NESP55 but not the XLαs transcript. This suggests that in somatotroph adenomas, oncogenic gsp mutations virtually always occur on the maternal allele.

Figure 2

(a) Sequences across codon 201 of the three protein-coding GNAS1 transcripts (G_sα, XLαs, and NESP55) in two R201C gsp⁺ adenomas (A103, A106). In both examples, the normal allele CGT is expressed from the paternal (XLαs) chromosome, while only the mutated allele TGT (arrow) is represented in the maternal NESP55 transcripts. (b) Sequence across codon 201 (normally CGT) of the G_sα transcript in six gsp⁺ adenomas, illustrating the variation in relative expression level of the two G_sα alleles. All these adenomas carry the gsp mutation on the maternal allele, apart from NB89, which has the gsp mutation on the paternal allele. In each case, the mutant allele is indicated by an arrow.

In a single adenoma, the mutation (R201H) was found in XLαs (and 1A) transcripts, but not in NESP55 transcripts, suggesting it had occurred on the paternal allele. This mutation creates a NlaIII restriction site. NlaIII digestion of the RT-PCR products was performed to verify the sequencing result, confirming the presence of the mutation on the XLαs but not the NESP55 transcript (not shown). The other R201H adenoma, however, had the mutation on the maternal allele (also verified by NlaIII digests), indicating that there is nothing peculiar to the R201H mutation that requires it to occur on the paternal chromosome.

Dysregulated imprinting of G_sα transcripts in somatotroph tumors. In gsp⁺ pituitary tumors, the mutant is usually more abundant than the normal transcript (34). Previously, this has been attributed to a proliferative advantage from high expression of a dominant oncogene. However, a simpler explanation is that gsp mutations occur on the more highly expressed maternal G_sα allele. Figure 2b shows that among gsp⁺ tumors the ratio of mutant to normal G_sα transcript is quite variable. However, none of our 22 adenomas expressed solely mutant G_sα, even although most showed a bias in favor of the mutant G_sα transcript. This implies that although in most tumors transcription of G_sα from the maternal chromosome remains predominant, there is relaxation of the monoallelic expression pattern, at least in some. We addressed this point further in the context of gsp^– somatotroph adenomas.

Sixty percent of somatotroph adenomas are gsp negative (gsp^–), and the genetic events involved in their development are largely unknown. In several types of tumor relaxation of IGF2/H19 imprinting occurs (35–37). We therefore wondered whether altered GNAS1 imprinting might be a factor in the pathogenesis of gsp^– somatotroph tumors. We examined 19 gsp^– adenomas that were heterozygous at the exon 5 polymorphism. Figure 3a shows typical examples of the G_sα transcript sequences. The allelic contributions to G_sα expression vary considerably, ranging from exclusively maternal (e.g., A75) to approximately equal (e.g., 368).

Figure 3

(a) Reverse-strand sequence across the exon 5 A/G polymorphism (arrow) in G_sα transcripts from gsp^– adenomas, illustrating the variation in relative expression level of the two G_sα alleles. (b) Scatter diagram showing estimated proportion of paternally derived G_sα transcripts in four normal adult pituitaries (N), five nonfunctioning pituitary adenomas (NF), and 19 gsp^– somatotroph tumors (S). The ratios are derived from intensities of bands corresponding to the alleles of the exon 5 polymorphism, as described in the text.

A semiquantitative index of this variability was derived. To control for uneven sample loading, we normalized the intensity of the polymorphic band to that of four neighboring bands in the same lane and then calculated the paternal contribution as a proportion of the sum of the normalized values for the two alleles. (The maternal or paternal allele assignment was based on analysis of NESP55 and XLαs transcripts, as before.) In addition to 19 gsp^– somatotroph tumors, for comparison we analyzed the four heterozygous normal adult pituitaries and five nonfunctioning pituitary adenomas.

Although there is wide variation among the gsp^– somatotroph tumors, a substantial number shows a much greater proportion of paternally derived G_sα than is seen in either the normal samples or the nonfunctioning tumors (Figure 3b). In the normal pituitaries, G_sα expression was either exclusively from or very strongly biased toward the maternal allele. In contrast, in almost two-thirds of gsp^– adenomas, G_sα expression was approximately biallelic (paternal proportion 0.3–0.7).

As judged from codon 201 or 227 analysis, G_sα expression was also biallelic in several of the 22 gsp⁺ adenomas (e.g., Figure 2b, tumors A103, A84, NB89). Thus there appears to be relaxation of G_sα imprinting both in gsp⁺ and gsp^– somatotroph tumors. This loss of imprinting may therefore be a secondary feature of adenoma progression. We have shown previously that the GNAS1 transcript levels in gsp^– somatotroph adenomas are more variable than in gsp⁺ adenomas (38). It is possible that part of that variability could be due to relaxation of G_sα imprinting control, resulting in an increase in G_sα transcript.

We have found that in normal human pituitary, unlike most tissues, G_sα is monoallelically derived. Furthermore, this imprinted expression pattern is relaxed in a substantial proportion of growth hormone–secreting (GH-secreting) adenomas. In contrast, the opposing monoallelic expression patterns of NESP55 and XLαs are not altered in the GH-secreting tumors. This dissociation between imprinting in the exon 1 region and the far upstream NESP55/XLαs region is consistent with the fact that germline imprinting mutations in PHP-Ib patients can also have discordant effects on the two regions (15).

The five hormone-secreting cell types of the anterior pituitary derive from a common progenitor (39, 40). Cell type–specific hormone production can be observed by week 12 of human development (41). Since we observed predominantly monoallelic G_sα expression in whole adult pituitary, it is likely that all regions of the pituitary express G_sα monoallelically and therefore that this imprinted expression pattern is established in an early common precursor cell. However, biallelic expression of G_sα in a minority cell population of normal pituitary might not be detectable by our analysis, so it remains theoretically possible that G_sα expression is biallelic in the normal somatotroph. We feel this is unlikely, though, given the extreme allelic bias in origin of gsp mutations.

Given the role of G_sα in mediating the pituitary response to GHRH (42), its monoallelic expression is somewhat paradoxical. One might predict that PHP-Ia patients, who carry maternal null GNAS1 mutations, would have no functional G_sα in the pituitary. Although GH deficiency has been reported in PHP-Ia (43), it is not typical. Possibly, therefore, when an inactive GNAS1 allele is inherited maternally, compensatory expression of the paternal G_sα allele occurs sufficient to allow normal pituitary function. This seems possible, given our observation of a small, variable proportion of paternally derived G_sα transcript in two of the normal pituitaries analyzed. There may be some polymorphic or temporal variation in the degree of imprinting of G_sα, but larger numbers of normal pituitaries need to be examined to assess this point.

Like others, we find that in gsp⁺ tumors the mutant is often overexpressed relative to the normal allele. This might indicate a need for high-level gsp expression to allow tumor progression, but we find that this inequality in gsp and normal transcript levels is not invariable and is therefore unlikely to be required for tumor development. It is more likely that the higher gsp expression simply reflects the fact that the mutations occur on the maternal allele, which is normally expressed at a higher level in the pituitary. Presumably, a gsp mutation on a paternal allele would be silent in most cases.

Nonetheless, all gsp⁺ tumors do express some G_sα (often considerably) from the (nonmutated) paternal allele. This could reflect either an absolute need for some normal G_sα function within the cell or a secondary relaxation of imprinting as part of the tumorigenic process. In support of the latter interpretation is our finding that a similar relaxation of G_sα imprinting is also seen at least as frequently in gsp^– tumors. Therefore, for gsp⁺ tumors we favor a model in which the monoallelic origin of G_sα in normal pituitary dictates that the initial gsp mutation must occur on a maternal allele in order to have a phenotypic effect, but that during subsequent tumor growth this monoallelic expression pattern is lost. As a caveat, it is not known whether the gsp⁺ and normal mRNAs are equally stable. If gsp⁺ transcripts were less stable, then the steady-state preponderance of maternal (mutated) over paternal (normal) mRNAs would be less pronounced than in normal cells. Nonetheless, such a mechanism would not easily explain the variability in maternal/paternal ratio shown in Figure 2b and also cannot explain the apparent relaxation of imprinting in gsp^– tumors.

This work was supported by the Wellcome Trust (grant number 053701).

1. Weinstein, LS, et al. Activating mutations of the stimulatory G protein in the McCune-Albright syndrome. N Engl J Med 1991. 325:1688-1695.
   View this article via: PubMed Google Scholar
2. Landis, CA, et al. GTPase inhibiting mutations activate the alpha chain of Gs and stimulate adenylyl cyclase in human pituitary tumours. Nature 1989. 340:692-696.
   View this article via: PubMed CrossRef Google Scholar
3. Patten, JL, et al. Mutation in the gene encoding the stimulatory G protein of adenylate cyclase in Albright’s hereditary osteodystrophy. N Engl J Med 1990. 322:1412-1419.
   View this article via: PubMed Google Scholar
4. Weinstein, LS, et al. Mutations of the Gs alpha-subunit gene in Albright hereditary osteodystrophy detected by denaturing gradient gel electrophoresis. Proc Natl Acad Sci USA 1990. 87:8287-8290.
   View this article via: PubMed CrossRef Google Scholar
5. Davies, SJ, Hughes, HE. Imprinting in Albright’s hereditary osteodystrophy. J Med Genet 1993. 30:101-103.
   View this article via: PubMed Google Scholar
6. Farfel, Z, Brickman, AS, Kaslow, HR, Brothers, VM, Bourne, HR. Defect of receptor-cyclase coupling protein in pseudohypoparathyroidism. N Engl J Med 1980. 303:237-242.
   View this article via: PubMed Google Scholar
7. Levine, MA, Jap, TS, Mauseth, RS, Downs, RW, Spiegel, AM. Activity of the stimulatory guanine nucleotide-binding protein is reduced in erythrocytes from patients with pseudohypoparathyroidism and pseudopseudohypoparathyroidism: biochemical, endocrine, and genetic analysis of Albright’s hereditary osteodystrophy in six kindreds. J Clin Endocrinol Metab 1986. 62:497-502.
   View this article via: PubMed Google Scholar
8. Campbell, R, Gosden, CM, Bonthron, DT. Parental origin of transcription from the human GNAS1 gene. J Med Genet 1994. 31:607-614.
   View this article via: PubMed Google Scholar
9. Chase, LR, Melson, GL, Aurbach, GD. Pseudohypoparathyroidism: defective excretion of 3’,5′-AMP in response to parathyroid hormone. J Clin Invest 1969. 48:1832-1844.
   View this article via: JCI PubMed CrossRef Google Scholar
10. Hayward, BE, et al. The human GNAS1 gene is imprinted and encodes distinct paternally and biallelically expressed G proteins. Proc Natl Acad Sci USA 1998. 95:10038-10043.
    View this article via: PubMed CrossRef Google Scholar
11. Hayward, BE, Moran, V, Strain, L, Bonthron, DT. Bidirectional imprinting of a single gene: GNAS1 encodes maternally, paternally, and biallelically derived proteins. Proc Natl Acad Sci USA 1998. 95:15475-15480.
    View this article via: PubMed CrossRef Google Scholar
12. Hayward, BE, Bonthron, DT. An imprinted antisense transcript at the human GNAS1 locus. Hum Mol Genet 2000. 9:835-841.
    View this article via: PubMed CrossRef Google Scholar
13. Swaroop, A, Agarwal, N, Gruen, JR, Bick, D, Weissman, SM. Differential expression of novel Gs alpha signal transduction protein cDNA species. Nucleic Acids Res 1991. 19:4725-4729.
    View this article via: PubMed CrossRef Google Scholar
14. Liu, J, Yu, S, Litman, D, Chen, W, Weinstein, LS. Identification of a methylation imprint mark within the mouse Gnas locus. Mol Cell Biol 2000. 20:5808-5817.
    View this article via: PubMed CrossRef Google Scholar
15. Liu, J, et al. A GNAS1 imprinting defect in pseudohypoparathyroidism type IB. J Clin Invest 2000. 106:1167-1174.
    View this article via: JCI PubMed CrossRef Google Scholar
16. Yu, S, et al. Variable and tissue-specific hormone resistance in heterotrimeric Gs protein alpha-subunit (Gsalpha) knockout mice is due to tissue-specific imprinting of the gsalpha gene. Proc Natl Acad Sci USA 1998. 95:8715-8720.
    View this article via: PubMed CrossRef Google Scholar
17. Pasolli, H, Klemke, M, Kehlenbach, R, Wang, Y, Huttner, W. Characterization of the extra-large G-protein α-subunit XLαs. I. Tissue distribution and subcellular localization. J Biol Chem 2000. 275:33622-33632.
    View this article via: PubMed CrossRef Google Scholar
18. Lovisetti-Scamihorn, P, Fischer-Colbrie, R, Leitner, B, Scherzer, G, Winkler, H. Relative amounts and molecular forms of NESP55 in various bovine tissues. Brain Res 1999. 829:99-106.
    View this article via: PubMed CrossRef Google Scholar
19. Klemke, M, et al. Characterization of the extra-large G-protein α-subunit XLαs. II. Signal transduction properties. J Biol Chem 2000. 275:33633-33640.
    View this article via: PubMed CrossRef Google Scholar
20. Ischia, R, et al. Molecular cloning and characterization of NESP55, a novel chromogranin-like precursor of a peptide with 5-HT1B receptor antagonist activity. J Biol Chem 1997. 272:11657-11662.
    View this article via: PubMed CrossRef Google Scholar
21. Lyons, J, et al. Two G protein oncogenes in human endocrine tumors. Science 1990. 249:655-659.
    View this article via: PubMed CrossRef Google Scholar
22. Landis, CA, et al. Clinical characteristics of acromegalic patients whose pituitary tumors contain mutant Gs protein. J Clin Endocrinol Metab 1990. 71:1416-1420.
    View this article via: PubMed Google Scholar
23. Yang, I, et al. Characteristics of gsp-positive growth hormone-secreting pituitary tumors in Korean acromegalic patients. Eur J Endocrinol 1996. 134:720-726.
    View this article via: PubMed Google Scholar
24. Suarez, HG, et al. gsp mutations in human thyroid tumours. Oncogene 1991. 6:677-679.
    View this article via: PubMed Google Scholar
25. Fragoso, MC, et al. Activating mutation of the stimulatory G protein (gsp) as a putative cause of ovarian and testicular human stromal Leydig cell tumors. J Clin Endocrinol Metab 1998. 83:2074-2078.
    View this article via: PubMed CrossRef Google Scholar
26. Williamson, EA, Ince, PG, Harrison, D, Kendall-Taylor, P, Harris, PE. G-protein mutations in human pituitary adrenocorticotrophic hormone-secreting adenomas. Eur J Clin Invest 1995. 25:128-131.
    View this article via: PubMed CrossRef Google Scholar
27. Williamson, EA, Johnson, SJ, Foster, S, Kendall-Taylor, P, Harris, PE. G protein gene mutations in patients with multiple endocrinopathies. J Clin Endocrinol Metab 1995. 80:1702-1705.
    View this article via: PubMed CrossRef Google Scholar
28. Hosoi, E, et al. Analysis of the Gs alpha gene in growth hormone-secreting pituitary adenomas by the polymerase chain reaction-direct sequencing method using paraffin-embedded tissues. Acta Endocrinol (Copenh) 1993. 129:301-306.
    View this article via: PubMed Google Scholar
29. Yoshimoto, K, Iwahana, H, Fukuda, A, Sano, T, Itakura, M. Rare mutations of the Gs alpha subunit gene in human endocrine tumors. Mutation detection by polymerase chain reaction-primer-introduced restriction analysis. Cancer 1993. 72:1386-1393.
    View this article via: PubMed CrossRef Google Scholar
30. Spada, A, Lania, A, Ballare, E. G protein abnormalities in pituitary adenomas. Mol Cell Endocrinol 1998. 142:1-14.
    View this article via: PubMed CrossRef Google Scholar
31. Barlier, A, et al. Expression of functional growth hormone secretagogue receptors in human pituitary adenomas: polymerase chain reaction, triple in-situ hybridization and cell culture studies. J Neuroendocrinol 1999. 11:491-502.
    View this article via: PubMed CrossRef Google Scholar
32. Barlier, A, et al. Pronostic and therapeutic consequences of Gs alpha mutations in somatotroph adenomas. J Clin Endocrinol Metab 1998. 83:1604-1610.
    View this article via: PubMed CrossRef Google Scholar
33. Hayward, BE, Bonthron, DT. Structure and alternative splicing of the ketohexokinase gene. Eur J Biochem 1998. 257:85-91.
    View this article via: PubMed CrossRef Google Scholar
34. Ballare, E, et al. Activating mutations of the Gs alpha gene are associated with low levels of Gs alpha protein in growth hormone-secreting tumors. J Clin Endocrinol Metab 1998. 83:4386-4390.
    View this article via: PubMed CrossRef Google Scholar
35. Ogawa, O, et al. Relaxation of insulin-like growth factor II gene imprinting implicated in Wilms’ tumour. Nature 1993. 362:749-751.
    View this article via: PubMed CrossRef Google Scholar
36. Rainier, S, et al. Relaxation of imprinted genes in human cancer. Nature 1993. 362:747-749.
    View this article via: PubMed CrossRef Google Scholar
37. el-Naggar, AK, et al. Frequent loss of imprinting at the IGF2 and H19 genes in head and neck squamous carcinoma. Oncogene 1999. 18:7063-7069.
    View this article via: PubMed CrossRef Google Scholar
38. Barlier, A, et al. 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. J Clin Endocrinol Metab 1999. 84:2759-2765.
    View this article via: PubMed CrossRef Google Scholar
39. Watkins-Chow, DE, Camper, SA. How many homeobox genes does it take to make a pituitary gland? Trends Genet 1998. 14:284-290.
    View this article via: PubMed CrossRef Google Scholar
40. Sheng, HZ, Westphal, H. Early steps in pituitary organogenesis. Trends Genet 1999. 15:236-240.
    View this article via: PubMed CrossRef Google Scholar
41. Asa, SL, Kovacs, K, Singer, W. Human fetal adenohypophysis: morphologic and functional analysis in vitro. Neuroendocrinology 1991. 53:562-572.
    View this article via: PubMed CrossRef Google Scholar
42. Mayo, KE, et al. Regulation of the pituitary somatotroph cell by GHRH and its receptor. Recent Prog Horm Res 2000. 55:237-266.
    View this article via: PubMed Google Scholar
43. Scott, DC, Hung, W. Pseudohypoparathyroidism type Ia and growth hormone deficiency in two siblings. J Pediatr Endocrinol Metab 1995. 8:205-207.
    View this article via: PubMed Google Scholar