0021-972x/92/7402-0357$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society

Vol. 74, No. 2 Printed

Expression of Growth Gene in GH-Producing ICHIJI WAKABAYASHI, HITOSHI SUGIHARA, Department

of

Medicine,

Hormone Pituitary

KOITI INOKUCHI, SHIRO MINAMI

OSAMU

Factor

HASEGAWA,

AND

Nippon

Medical

School, Sendagi

l-l-5,

Bunkyoku,

ABSTRACT. Pituitary cells synthesize various neuropeptides that influence pituitary hormone secretion. GH-releasing factor (GRF) may also be produced by normal or pituitary tumor cells. We examined GRF gene expression in pituitary tumors. Standard techniques for the analysis of GRF gene expression did not appear to be suitable. Highly sensitive reverse transcription coupled to polymerase chain reaction was used. Specimens of pituitary adenoma were obtained by transsphenoidal adenomectomy from six patients with acromegaly and three patients with no clinical evidence of pituitary hormone overproduction; non-

G

(GH)-Releasing Adenoma*

in U.S.A.

Tokyo 113, Japan

functioning adenoma. Pituitary glands were collected at autopsy from three patients who died from nonendocrine disorders. A specific GRF gene transcript was detected in five out of six GHproducing pituitary adenomas, whereas this was not found in three separate specimens of nonfunctioning pituitary adenoma or anterior and posterior pituitary tissue. The data suggest that GRF is synthesized as an intrinsic product in human GHproducing pituitary adenoma. (J Clin Endocrinol Metab 74: 357361,1992)

pituitary adenoma is by far the most common cause of acromegaly. The pathogenesis of GH-producing adenoma is not known. GRF is known to stimulate release and synthesis of GH, as well as proliferation of GH-producing cells (1, 2). Human tumor GH cells are shown to respond to GH-releasing factor (GRF) similar to normal pituitary cells (3). GRF may exert influences on GH secretion and proliferation of tumor GH cells. It remains to be clarified how the release of GRF from the hypothalamus is modulated in patients with pituitary GH-producing adenoma (4). Alternatively, GRF may be produced by pituitary cells or tumor GH cells. Data have been presented that a number of neuropeptides known to influence anterior pituitary secretion, such as vasoactive intestinal polypeptide, neuropeptide Y, substance P, galanin, interleukin-6, interleukin-l@ are synthesized by pituitary cells (4-lo), and TRH and LHRH-like immunoreactivities are localized in anterior pituitary cells (11, 12). Although GRF is not detectably expressed in the pituitary in rats (13), a study of Joubert et al. (14) suggests GRF synthesis in normal pituitary and GH-producing pituitary adenoma in man (14).

The objective of studies described here is to examine GRF gene expression in specimens of pituitary adenoma. To this end, we attempted to detect GRF gene transcript using reverse transcription coupled to polymerase chain reaction (RT-PCR).

H-PRODUCING

Materials

and Methods

Specimens of pituitary adenoma were obtained from six with acromegaly and three patients (Table 1) with no clinical evidence of pituitary hormone overproduction; nonfunctioning adenoma, by selective transsphenoiclal adenomectomy. Small tissue fragments were fixed for light microscopy. Specimens of the pituitary gland and hypothalamic-pituitary stalk were collected at autopsy, 3-18 h after death, from three patients who had died from lung cancer, hepatoma, or myelodysplastic syndrome. All specimens were kept under -80 C until RNA extraction. RNA from the specimens was prepared by the guanidinium CsCl method (15). RT-PCR was performed according to the method of Roth et al. (16) with slight modifications. Approximately 0.5 pg total RNA was used as the template for complementary DNA (cDNA) synthesis by 10 U avian myeloblastosis virus reverse transcriptase (Boehringer Mannheim, Indianapolis, IN) using a primer, GRT (Fig. 1). Synthesis was carried out at 37 C for 60 min in 50 ~1containing 50 mM Tris-HCl, pH 8.3, 50 mM KCl, 8 mM MgCl*, 10 mM dithiothreitol, 500 PM dNTP, and 0.05 OD, U primer, GRT. Ten microliters of the cDNA reaction mixture was used for PCR with 0.0125 ODtGOU primers, Gl and G2 (Fig. 1). PCR was performed in 50 ~1 volume containing 10 mM Tris-HCl, pH8.3,50 mM KCl, 15 mM MgC12, 2 mM dithiothreitol, 200 MM dNTP, and 2 U Taq patients

Received March 11, 1991. Address requests for reprints to: Dr. Ichiji Wakabayashi, Department of Medicine, Nippon Medical School, Sendagi l-l-5, Bunkyoku, Tokyo 113, Japan. *This work was supported in part by a research grant from the Intractable Diseases Division, Japanese Ministry of Health and Welfare.

357

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 03:04 For personal use only. No other uses without permission. . All rights reserved.

358

WAKABAYASHI

ET AL.

JCE & M .1992 Vol74.No2

1. Profile of patients

TABLE

Patients Acromegaly K.Y. LT. S.K.B Y.M. G.M. S.F. Nonfunctioning K.T.h T.T.’ T.T.

hGH km*

PRL hm’

42 39 61 23 48 32

140 52 60 107 14 18

5 87 8 6 48 14

L L L L L’ M

69 70 43

3 3 4

7 11 15

Lf L’ L+

Sex

Age”

F F F F F F M M F

Tumor sized

Duration (yr)

GRF gene transcript’

12 16 4 4 16 Unknown

+ + + + +

adenoma

0Age when adenomectomy was performed. b.cMean of baseline levels. Normal values, GH below 5 pg/L, PRL below 14 rg/L. d L’, Macroadenoma with suprasellar extension; L, macroadenoma; M, microadenoma. e Estimated from history and photographs of patients. f +: GRF gene transcript was detected by RT-PCR. f Associated with hyperparathyroidism. h Complicated by LH, FSH, TSH, and ACTH deficiencies. ’ Complicated by LH, FSH, and TSH deficiencies. Ikb

EXONS; I

2

3

4

5

50bp FIG. 1. Diagram of the hGRF gene and transcript. The structure of the GRF gene is represented schematically in the upper part of the figure. The cross-batched boxes are the coding exons of GRF. Open boxes are indicated 5’- and 3’-nontranslated regions, and poly(A) region. The transcript of GRF is represented in the lowerpart. Positions of oligomers used for RT-PCR analysis is indicated by thick bars (Gl, G2, and GRT). GRT:5’-TTTGGCTACAGGTAGCCCGG-3’, antisense strand G1:5’-TATGCAGATGCCATCTTCAC-3’, coding strand G2:5’GGAGTTCCTGCTGTGCTTCT-3’, antisense strand.

polymerase (Perkin Elmer-Cetus, Emeryville, CA). PCR of 40 cycles was performed, consisting of denaturation (94 C, 30 s), annealing (55 C, 30 s) and extension (75 C, 1 min). As a negative control, RNA from the K562 cells; a human chronic myelogenous leukemia cell line, was similarly reverse transcribed and subjected to PCR as described above. Finally, the PCR products (50 ~1) were phenol-extracted, ethanol-precipitated, and electrophoresed in a 2% agarose gel. PCR products were transferred to a nylon membrane filter (Gene-Screen Plus, New England Nuclear, Boston, MA). In a study (Fig. 2), PCR products were visualized with ethidium bromide staining initially, then transferred to a nylon membrane. The membrane was prehybridized at 42 C for 2 h in the solution containing 50% deionized formamide, 5~ SSC, 1X PE (50 mM Tris HCl, pH 7.5, 0.1% sodium pyrophosphate, 1%

sodium dodecyl sulfate (SDS), 0.2% polyvinylpyrrolidone, 0.2% ficoll, 5 mM EDTA, 0.2% BSA), 150 pg/ml denatured salmon sperm DNA. Hybridization was performed at the same temperature overnight in the same solution containing a 32P-labeled human cDNA RNA probe. The membrane was washed for 15 min twice in 2~ SSC, 0.1% SDS at 55 C and for 15 min twice in 0.1~ SSC and 0.1% SDS at the same temperature. The membrane was exposed to Kodak XAR-5 film at -80 C for 5 h. The human GRF cDNA clone was kindly provided by Dr. Kelly E. Mayo. The BamHI-EcoRI 350 base pair (bp) fragment of the GRF cDNA was constructed into the BamHI-EcoRI site of pBluescript SK-II. After the SK-II vector was linearized with EcoRI enzyme, a high specific activity RNA probe was synthesized from the linearized vector with 32Pusing a Sp6/T7 transcription kit (Boehringer Mannheim, Indianapolis, IN) and T3 RNA polymerase (Boehringer Mannheim, Indianapolis, IN). The radioactive RNA probe was precipitated by 2 vol ethanol, solubilized in 1 X TE (10 mM Tris-HCl, pH 7.4, 1 mM EDTA), then added to prehybridizing medium.

Results Typical pituitary adenoma was ascertained in all hematoxylin-eosin stained specimens obtained by transsphenoidal pituitary adenomectomy. As expected, PCR conducted with hypothalamic tissue cDNA produced a single intense band smaller than 234 bp marker and larger thati 194 bp marker as viewed on ethidium stained gel. A specific GRF PCR-product was detected as much as 50 ng total RNA from hypothalamic tissue (Fig. 2). Amplification of cDNA from pituitary adenoma tissue from patients with acromegaly also produced a band equal in size to that of human hypothalamic tissue in

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 03:04 For personal use only. No other uses without permission. . All rights reserved.

GRF

GENE

EXPRESSION

IN PITUITARY

bps

ADENOMA

359

1,2,3,4,5,6,7,8,9,10,11,~,13,~?5,16,17,16,

4.368-

Ml2345

b 1098.

872.

603.

FIG. 3. Autoradiogram of the radioactively probed PCR products derived from specimens of pituitary tissue. Five hundred nanograms of total RNAs were investigated in this analysis. Lane 1, RNA from the human hypothalamus; lanes 2 and 18, RNA from the K562 cell line (chronic myelogenous leukemia cell line); lanes 3-5, RNA from anterior pituitary; lanes 6-8, RNA from posterior pituitary; lanes 9-11, RNA from nonfunctioning pituitary adenomas; lanes 12-17, RNA from GHproducing pituitary adenomas. The size of the molecular marker are indicated at the left. Arrows indicate the specific GRF PCR-products (225 bp).

FIG. 2. Detection of GRF transcript by RT-PCR. Various weights of total RNA extracted from the human hypothalamus were reversetranscribed, amplified, and electrophoresed through a 2% agarose gel. The ethidium bromide stained gel (A) and the autoradiograph (B) of this gel after transfer to nylon membrane and hybridization to a GRF cDNA RNA probe. Lane M, Size marker of HindIII-digested DNA and MspI-digested DNA; lane 1, 500 ng total RNA; lane 2, 100 ng total RNA, lane 3, 50 ng total RNA; lane 4, 10 ng of total RNA; lane 5, 1 ng total RNA. The sizes of the marker are indicated at the left. Arrows indicate a specific GRF PCR-product (225 bp).

five out of six cases (Fig. 3, Table 1). A specific GRF PCR-product was not detected in any tissue specimens obtained from patients with nonfunctioning pituitary adenoma or autopsy specimens. Finally, PCR conducted with cDNA from the K562 cells did not produce any bands that hybridized with a hGRF cDNA RNA probe. Discussion Several studies have found the presence of GRF-like immunoreactivity in gastric G cells, liver, lung, pancreatic islet cells, and placenta in man and rat (17-19). GRF-like immunoreactivity was also found in human tumors including pancreatic endocrine tumors, hypothalamic gangliocytoma, carcinoids of the lung, pheochromocytoma, medullary thyroid carcinoma and small cell carcinoma of the lung (17-21). These data suggest that

GRF is expressed widely in normal and tumor tissues other than the hypothalamus. However, they do not provide informations on the authenticity of GRF-like immunoreactivity. It is also not known whether GRFlike material is synthesized locally. We found that GRF gene is expressed in GH-producing pituitary adenoma. This is in stark contrast to undetectable expression of the GRF gene in normal pituitary or in nonfunctioning pituitary adenoma. Although structures expressing GRF need to be identified in future studies, the finding suggests that GRF is synthesized by GH-producing adenoma. Like hypothalamic GRF, tumor GRF is expected to stimulate GH release. Studies by Pagesy et al. (22) demonstrated that somatostatin gene was expressed in GH-producing pituitary adenoma, and the synthesis and release of somatostatin were increased in GH-producing pituitary adenomas with poor GH expression as compared to those with high GH expression (22). It may be that the effect of tumor GRF on GH secretion is also influenced by locally produced somatostatin and thus, the biological activity of GH-producing pituitary adenoma depends on the balance of interaction between GRF and somatostatin released by tumor cells. In addition, it is suggested that GRF serves not only as a trophic factor for somatotrophs, but also has oncogenic potential. GRF stimulates mitotic activity of somatotrophs and induces the expression of c-fos, a growth signal-transducing oncogene (1, 2). Mice transgenic for hGRF initially revealed a selective hyperplasia of somatotrophs, and later developed pituitary mammosomatotroph adenoma (23, 24). Asa et al. (21) described that

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 03:04 For personal use only. No other uses without permission. . All rights reserved.

360

WAKABAYASHI

hypothalamic gangliocytoma containing GRF was associated with pituitary somatotroph adenoma and acromegaly. We do not know the causal relationship between GRF gene expression and formation of adenoma in acromegalic patients. Our finding may have implications for understanding the role of locally produced GRF in controlling GH secretion from adenoma and/or somatotroph adenoma cell proliferation. Joubert et al. (14) reported the existence of GRF in both normal and GH secreting adenomatous pituitaries using a RIA (14). This appears to be in conflict with our observation, since the GRF gene was not detectably expressed in the normal pituitary tissue. GRF in the pituitary tissue may represent those produced locally and accumulated from blood-borne GRF, presumably of hypothalamic origin. In the normal pituitary tissue, the synthesis of GRF in the pituitary may be minute so that GRF gene expression could not be demonstrated, while the accumulated GRF could be measured by a RIA. In line with this suggestion, the existence of immunoreactive LHRH in normal rat pituitary tissue had been observed (14), whereas Azad et al. (25) could not demonstrate LHRH gene expression by RT-PCR. RT-PCR allows the detection of very low levels of GRF gene expression. Trace levels of sample contamination will complicate the interpretation of the results. We took precautions to cope with the problem. The guanidinium CsCl method was used to minimize DNA contamination, when total cellular RNA from tissue specimens was prepared. Amplification of genomic DNA will produce a band larger than 2 kilobases instead of 225 bp, because the specific PCR product contains the third GRF intron. To control cross-contamination, we included the K562 cells as GRF non-expressing cells. In addition, the finding that GRF gene transcript was consistently undetectable in three separate specimens of anterior and posterior pituitary or nonfunctioning adenoma supports that cross-contamination is highly unlikely. GRF gene expression can be examined by Northern blot analysis or in situ hybridization, but there were limitations to such cytogenetic techniques. The degree of GRF gene expression likely differs among tissue specimens. Messenger RNA may degrade during handling of the specimens of pituitary adenoma. We observed that Northern blot analysis of RNA from hypothalamic tissue obtained at autopsy revealed a smear band (data not shown). Taken together, we chose RT-PCR to examine GRF gene expression in various tissue specimens. The use of the method enabled us to detect a specific GRF gene transcript in a specimen of GH-producing adenoma weighing as much as 8 mg.

ET AL.

JCE & M - 1992

Voll4.No2

References 1. Billestrup N, Swanson LW, Vale W. Growth hormone-releasing factor stimulates proliferation of somatotrophs in uitro. Proc Nat1 Acad Sci USA. 1986;83:6854-7. 2. Billestrup N, Mitchell RL, Vale W, Verma IM. Growth hormonereleasing factor induces c-fos expression in cultured primary pituitary cells. Mol Endocrinol. 1987;1:300-5. 3. Webb CB, Thominet JL, Frohman LA. Ectopic growth hormone releasing factor stimulates growth hormone release from human somatotroph adenomas in vitro. J Clin Endocrinol Metab. 1983;56:417-9. 4. Melmed S. Acromegaly. N Engl J Med. 1990;322:966-77. 5. Aronin N, Morency K, Leeman SE, Braverman LE, Coslovsky R. Regulation by thyroid hormone of the concentration of substance P in the rat anterior pituitary. Endocrinology. 1984;114:2138-42. 6. Kaplan LM, Gabriel SM, Koenig JI, et al. Galanin is an estrogeninducible, secretory product of the rat anterior pituitary. Proc Nat1 Acad Sci USA. 1988;85:7408-12. 7. Jones PM, Ghatei MA, Steel J, et al. Evidence for neuropeptide Y synthesis in the rat anterior pituitary and the influence of thyroid hormone status: comparison with vasoactive intestinal peptide, substance P, and neurotensin. Endocrinology. 1989;125:334-41. 8. Segerson TP, Lam KSL, Cacicedo L, et al. Thyroid hormone regulates vasoactive intestinal peptide (VIP) mRNA levels in the rat anterior pituitary gland. Endocrinology. 1989;125:2221-3. 9. Vankelecom H, Carmeliet P, Van Damme J, Billiau A, Denef C. Production of interleukin-6 by folliculo-stellate cells of the anterior pituitary gland in a histiotypic cell aggregate culture system. Neuroendocrinology. 1989;49:102-6. 10. Koenig JI, Snow K, Clark BD, et al. Intrinsic pituitary interleukin16 is induced by bacterial lipopolysaccharide. Endocrinology. 1990;126:3053-8. 11. Childs CV (Moriarty), Cole DE, Kubek M, Tobin RB, Wilber JF. Endogenous thyrotropin-releasing hormone in the anterior pituitary: sites of activity as identified by immunocytochemical staining. J Histochem Cvtochem. 1978:26:901-8. 12. Li JY, Knapp- RJ, Sternberger LA. Immuncytochemistry of a “private” luteinizing-hormone-releasing hormone system in the pituitary. Cell Tissue Res. 1984;235:263-6. 13. Sawchenko PE, Swanson LW, Rivier J, Vale WW. The distribution of growth-hormone-releasing factor (GRF) immunoreactivity in the central nervous system of the rat: an immunohistochemical study using antisera directed against rat hypothalamic GRF. J Comp Neurol. 1985;237:100-15. 14. Joubert (Bression) D, Benlot C, Lagoguey A, et al. Normal and growth hormone (GH)-secreting adenomatous human pituitaries release somatostatin and GH-releasing hormone. J Clin Endocrinol Metab. 198368572-7. 15. Maniatis T, Fritsch FF, Sambrook J. Molecular cloning. A laboratorv manual. Cold Snrina Harbor: Cold Snrine - - Harbor Laboratory;i982;196. - 16. Roth MS, Antin JH, Bingham EL, Ginsburg D. Detection of Philadelphia chromosome-positive cells by the polymerase chain reaction following bone marrow transplant for chronic myelogenous leukemia. Blood. 1989;74:882-5. 17. Bosman FT, Assche CV, Nieuwenhuyzen Kruseman AC, Jackson S, Lowry PJ. Growth hormone releasing factor (GRF) immunoreactivity in human and rat gastrointestinal tract and pancreas. J Histochem Cytochem. 1984;32:1139-44. 18. Christofides ND, Stephanou A, Suzuki H, Yiangou Y, Bloom SR. Distribution of immunoreactive growth hormone-releasing hormone in the human brain and intestine and its production by tumors. J Clin Endocrinol Metab. 1984;59:747-51. 19. Shibasaki T, Kiyosawa Y, Masuda A, et al. Distribution of growth hormone-releasing hormone-like immunoreactivity in human tissue extracts. J Clin Endocrinol Metab. 1984;59:263-8. 20. Asa SL, Kovacs K, Thorner MO, Leong DA, Rivier J, Vale W. Immunohistochemical localization of growth hormone-releasing hormone in human tumors. J Clin Endocrinol Metab. 1985;60:4237. 21. Asa SL, Scheithauer BW, Bilbao JM, et al. A case for hypothalamic

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 03:04 For personal use only. No other uses without permission. . All rights reserved.

GRF GENE EXPRESSION acromegaly: a clinicopathological study of six patients with hypothalamic gangliocytomas producing growth hormone-releasing factor. J Clin Endocrinol Metab. 1984;58:796-803. 22. Pagesy P, Li JY, Rentier-Delrue F, et al. Growth hormone and somatostatin gene expression in pituitary adenomas with active acromegaly and minimal plasma growth hormone elevation. Acta Endocrinol (Copenh). 1990;122:745-52. 23. Hammer RE, Brinster RL, Rosenfeld MG, Evans RM, Mayo KE. Expression of human growth hormone-releasing factor in trans-

IN PITUITARY

ADENOMA

genie mice results in increased somatic growth. Nature. 1985;315:413-6. 24. Asa SL, Kovacs K, Stefaneanu L, et al. Pituitary mammosomatotroph adenomas develop in old mice transgenic for growth hormone-releasing hormone. Proc Sot Expl Biol Med. 1990;193:2325. 25. Azad N, Emanuele NV, Halloran MM, Tentler J, Kellely MR. Presence of luteinizing-releasing hormone (LH-RH) mRNA in rat spleen lymphocytes. Endocrinology. 1991;128:1679-1681.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 03:04 For personal use only. No other uses without permission. . All rights reserved.

Expression of growth hormone (GH)-releasing factor gene in GH-producing pituitary adenoma.

Pituitary cells synthesize various neuropeptides that influence pituitary hormone secretion. GH-releasing factor (GRF) may also be produced by normal ...
779KB Sizes 0 Downloads 0 Views