0013.7227/92/1315-2083$03.00/0 Endocrinology Copyright 0 1992 by The Endocrine

Pituitary Growth

Vol. 131, No. 5 Printed in U.S A.

Society

Adenomas in Mice Transgenic Hormone-Releasing Hormone*

SYLVIA L. ASA, KALMAN KOVACS, LUCIA STEFANEANU, NILS BILLESTRUP, CONSUELO GONZALEZ-MANCHON, Department of Pathology, St. Michael’s Hospital University Ontario, Canada M5B 1 W8; Hagedorn Research Laboratory Salk Institute (C.G.M. W. V.), La Jolla, California 92037

for

EVA HORVATH, AND WYLIE VALEt

of Toronto (S.L.A., K.K., L.S., E.H.), Toronto, (N. B.), DK-2820 Gentofte, Denmark; and The

ABSTRACT

reactivity; in situ hybridization demonstrated focal PRL mRNA in 3 of 5 immunohistochemically positive tumors. a-Subunit was positive by immunohistochemistry in 8 adenomas, and TSHfl was localized in tumor cells of 5 adenomas. The adenomas had variable ultrastructural appearances, ranging from cells that resembled somatotrophs or mammosomatotrophs to cells with features of the glycoprotein hormone cell line. These findings provide conclusive evidence that protracted GRH stimulation of secretory activity can result in proliferation, hyperplasia, and adenoma of adenohypophysial cells. (Endocrinology 131: 2083-

It has been shown that mice transgenic for human GH-releasing hormone (GRH) develon hvnernlasia of nituitarv somatotroohs, lactotrophs, and mammosomatotrophs, cells capable of producing both GH and PRL, by 8 months of age. We now report that GRH transgenic mice lo-24 months of age develop pituitary adenomas, which we characterized by histology, immunohistochemistry, in situ hybridization, and electron microscopy. Of 13 animals examined, all developed GH-immunoreactive neoplasms that had diffuse positivity for GH mRNA by in situ hybridization. Eleven also contained PRL immuno-

2089,1992)

H

hormones regulate hormone syntheYPOTHALAMIC sis and release by adenohypophysial cells and some pituitary tumors. It has been suggested that increased stimulation or decreased inhibition by the hypothalamic peptides that physiologically regulate the hormonal activity of adenohypophysial cells may also play a role in the development of pituitary adenomas (l-5). Acromegaly and gigantism are due to excessive GH secretion by the pituitary; these disorders are usually caused by a pituitary adenoma producing GH (6). Hormone release by these adenomas can be stimulated by GH-releasing hormone (GRH) and suppressed by somatostatin (7). Prolonged GRH excess occurs rarely in patients with GRH-secreting extracranial tumors; chronic overproduction of GRH increases pituitary GH release, giving rise to elevated blood GH levels and the development of acromegaly or gigantism (8). The pituitaries of patients bearing extracranial GRH-secreting tumors have been reported to be enlarged and in most cases show diffuse or nodular hyperplasia of somatotrophs (8). In one case, a patient with a pancreatic GRHproducing tumor was documented to have a GH-containing pituitary adenoma (8). GRH-containing hypothalamic gangliocytomas have been associated with pituitary somatotroph adenoma (9), and it has been suggested that GRH excess played a role in the development of those pituitary tumors

(9). To obtain deeper insight into the role of GRH in the development of pituitary tumors, we studied mice transgenic for GRH. These mice are known to have increased body weight and size as well as hyperplasia of the adenohypophysial target cells of GRH (10, 11). Using this model, we studied the effect of isolated hormone gene overexpression and protracted hormone excess. In a previous report we suggested that long term GRH stimulation resulted in the development of pituitary adenoma based on a small number of animals (12). We now report the consistent development of pituitary adenomas in mice older than 10 months of age; these results provide strong evidence that sustained GRH excess plays a role in pituitary tumorigenesis.

Materials Generation

of transgenic

and Methods

mice

The human (h) GRH/mouse metallothionein-I (MT-Q/simian virus40 small t fusion gene, described previously (1 l), was used as the DNA fragment to develop transgenic mice. Briefly, the construct consisted of a 713.basepair fragment of the mouse MT-1 promoter, containing elements responsible for metal induction and transcription initiation, fused to 220 basepairs of the hGRH gene, encoding the NH2-terminal 31. amino acid signal peptide and the 40-amino acid form of hGRH. The polyadenylation signal was provided by fusion to an 847.basepair fragment of the simian virus-40 virus small t poly(A); this portion of the gene is involved only in polyadenylation and is not translated. Fertilized zygotes of the B6D2Fl hybrid strain of mice, produced by mating C57BL/6 females and DBA/2 males, were microinjected with purified DNA fragments. Transgenic pups in litters of a single transgenic line were identified at the time of weaning; tail DNA was analyzed for the presence of hGRH by a DNA dot blotting method. From 6 weeks of age, some animals were maintained on water containing 25 mM ZnSOl and laboratory chow ad libitum; there was no effect of zinc on growth rate, total body weight, or serum GH or GRH, and addition of zinc to

Received May 20, 1992. Address all correspondence and requests for reprints to: Dr. Sylvia L. Asa, Department of Pathology, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario, Canada M5G 1X5. *This work was supported by grants from the Medical Research Council of Canada, the NIH, and Phillips Petroleum Co. The research was conducted in part by the Clayton Foundation for Research, California Division. t Senior Clayton Foundation investigator.

2083

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.

PITUITARY

2084

drinking water, therefore, was discontinued. the tails of 8-week-old animals to measure and hGRH by RIA (11).

ADENOMAS

IN GRH TRANSGENIC

Blood was collected from serum levels of mouse GH

Morphological methods Transgenic mice (n = 13) and age- and sex-matched controls (n = 15) were killed by decapitation. At autopsy, the pituitaries were removed and weighed; the other organs were carefully inspected. For light microscopy, portions of each pituitary were fixed in buffered formalin and embedded in paraffin; sections 4-5 pm thick were stained with hematoxylin and eosin and with the Gordon-Sweet silver method to demonstrate the reticulin fiber network. Immunohistochemical stains to localize adenohypophysial hormones were performed using the avidin-biotin-peroxidase complex, as described previously (6, 11). Primary antisera were donated by Dr. A. F. Parlow at the National Pituitary Agency (NIDDK, Bethesda, MD). The specificity of the reactions was verified by replacement of primary antiserum with normal rabbit serum and primary antiserum preabsorbed with homologous and heterologous antigens. In situ hybridization was performed on six tumors. Oligonucleotide probes were synthesized to hybridize with mouse GH-(145-151) and PRL-(64-70) residues and were purified using uolvacrvlamide gel electrophbresis by Genosys Biotechnologies, Inch [Woodlands TX): Probes were labeled by the 3’-end method with [35S]deoxy-ATPol, using a NEP100 kit (DuPont Canada, Inc., Mississauge, Ontario, Canada) and purified with a NENSORB-TM 20 cartridge. The details of in situ hybridization performed on 5-Grn thick paraffin sections, including prehybridization, hybridization, signal development, and controls, were described previously (13). Controls to assure the specificity of the reaction included competition with nonlabeled probe, RNase predigestion, known negative tissue, and immunohistochemical staining to verify localization of the peptide product. For transmission electron microscopy, pieces of nine tumors were fixed in 2.5% glutaraldehyde, postfixed in 1% osmium tetroxide, dehydrated in graded ethanols and propylene oxide, and embedded in an Epon-Araldite mixture. Semithin sections were stained with toluidine blue; ultrathin sections of selected areas were stained with uranyl acetate and lead citrate and examined with a Philips 41OLS electron microscope (Philips, Mahway, NJ). Ultrastructural immunocytochemistry was performed, using the double immunoaold techniaue to localize GH and PRL with protein-A-10 nm gold complex (GH)’ and immunoglobulin G-40 nm gold particle complex (PRL), as previously reported (6, 11, 14).

Results Gross appearance

and biochemistry

All animals identified as transgenic for hGRH had significantly increased body weight and elevated blood levels of GH and hGRH, similar to the changes described in these transgenic mice at 8 months of age (11); GH levels ranged from 16-560 rig/ml, and circulating GRH levels from 21-86 rig/ml. All 13 transgenic animals aged 10 months or older when killed had markedly enlarged and congestedpituitary glands; all but 2 extended upward out of the sella turcica and infiltrated the adjacent brain. There was no correlation noted between the size of the pituitary and other parameters, including age of the animal. Histological

findings

Histological examination showed hyperplasia of somatotrophs, lactotrophs, and bihormonal GH- and PRL-containing cells called mammosomatotrophs, similar to the morphological findings in B-month-old mice reported previously

MICE

Endo. Vol131.

1992 No 5

(11). In addition, all 13 animals had histological evidence of a pituitary tumor (Fig. 1, a-c); in some animals, lobulated tumors had the appearance of multifocal neoplasms,and in 2 instances, the suggestion of multiple neoplasmswas supported by differences in the profiles of hormone production, asdocumented by immunohistochemistry and in situ hybridization (see Table 1). The tumors infiltrated adjacent brain structures in several animals. Sheetsof monomorphous cells contained occasional binucleate cells (Fig. lc), and mitotic figures were readily identified. Some tumors showed marked nuclear pleomorphism. The Gordon-Sweet silver stain documented the presenceof a distorted reticulin network at the periphery of the adenomas, with absenceof reticulin fibers within the tumors (Fig. lb). These features fulfilled the criteria for the diagnosisof adenoma (6). Immunohistochemistry

and in situ hybridization

The majority of tumor cellsin all adenomasshowed strong immunoreactivity for GH (Fig. 2a). By in situ hybridization, the signal for GH mRNA was diffusely localized in tumor cells; however, the intensity of the signal varied from cell to cell (Fig. 2b). Eleven tumors exhibited focal PRL positivity by immunohistochemistry (Fig. 2~); in situ hybridization was positive only in scattered tumor cells in three of five immunopositive tumors examined by this technique (Fig. 2d). In eight adenomas, the a-subunit of glycoprotein hormones was found, usually in a diffuse cytoplasmic distribution in the majority of tumor cells, but in a few adenomas, the positivity was focal. In five tumors, TSHP was localized in a large subpopulation of tumor cells (Fig. 2e). In each case, negative controls verified the specificity of the reactions detected by immunohistochemistry and in situ hybridization. ACTH, FSHP, and LHP were detected in cells within the hyperplastic areas, but not within the tumors; the cells containing these hormones were of normal size and shape. Ultrastructural

features

Electron microscopy revealed that the tumors had variable ultrastructural morphology. Four tumors (Table 1, cases9 and 11-13) were composedof large densely granulated cells. They contained slightly eccentric ovoid nuclei, with one or two nucleoli. The large cytoplasm harbored abundant rough endoplasmic reticulum disposed in parallel arrays and well developed Golgi complexeswith forming secretory granules. The majority of cells exhibited numerous spherical secretory granules with high electron density, measuring 300-500 nm in diameter (Fig. 3a). In scattered cells, the secretory granules were smaller and less numerous. In many cells, fusion of secretory granules gave rise to large deposits of secretory material. A prominent feature was the extrusion of secretory granules and large deposits of secretory material into the intercellular space (Fig. 3b) or toward the capillary. Mitochondria were present in moderate numbers; a few phagolysosomesand lipid droplets were occasionally seen. These cells resembled densely granulated somatotrophs, but the presenceof large irregular secretory granules and extrusions of secretory material are thought to be features of mammo-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.

PITUITARY

ADENOMAS

IN GRH

TRANSGENIC

2085

MICE

1. Histology of pituitary adenomas in GRH-transgenic mice. a, In the pituitary of a GRH transgenic mouse, there is evidence of hyperplasia (top) as well as an area of disrupted architecture composed of sheets of large cells that do not form acini (bottom). Hematoxylin and eosiq stain; magnification, x190. b, The Gordon-Sweet silver stain confirms the presence of distended acini in the hyperplastic adenohypophysis (top) and total breakdown of the reticulin fiber network in the adenoma (bottom). Magnification, x190. c, A pituitary tumor in a GRH transgenic mouse is composed of highly pleomorphic cells, some of which are bi- and multinucleate (arrows). Hematoxylin and eosin stain; magnification, x240. FIG.

TABLE

1. Morphological

characteristics of tumors in GRH transgenic mice Age (months) 10 11

4 5 6

12 13 13 13

7 8

13 16

9 10 11 12 13

16 17 17 17 24

In SAL

Electron microscopy

hybridization GH a) GH b) GH, PRL GH, PRL* GH; PRL* GH, PRL, (Y,TSHp a) GH. PRL* bj GH; PRL, (Y,TSHP GH, PRL, cy GH, PRL, CX* GH, PRL, cy,TSH/3 GH, PRL, cr,TSHj3 GH, PRL, CY* GH GH, PRL, LY,TSHfl

MG cells with granule extrusions MG cells with granule extrusions

GH” GH” GH,” PRL* GH” GH”

SG cells, no granule extrusions MG cells with eranule extrusions

GH,” PRL*

DG SG DG DG DG

::pa

PRL*

cells with granule extrusions cells, no granule extrusions cells with granule extrusions cells with granule extrusions cells with granule extrusions

a and b are two adenomas. DG, Densely granulated, MG, moderately granulated; SG, sparsley granulated. a Diffuse. * Focal. somatotrophs. The extruded material contained two cases, PRL also, as shown by immunogold

GH and, in labeling (Fig.

3c). Two tumors (Table 1, cases 5 and 10) were composed of very sparsely granulated cells, with no resemblance to nontumorous somatotrophs. The abundant cytoplasm was occupied by contorted parallel arrays of rough endoplasmic reticulum. In some cells the rough endoplasmic reticulum was represented by short dilated cistemae or abundant small vesicles. The Golgi complexes were very prominent and sometimes occupied a large part of the cytoplasm. These

features are characteristic of cells that produce glycoprotein hormones. In most cells, secretory granules were small and few (Fig. 4a). Occasionally larger secretory granules were seen. Abnormal secretory granules with rectangular or angular shape were present in the Golgi area of some cells (Fig. 4b). Long needle-shaped deposits of secretory material were also present in the cytoplasm of some cells. Three adenomas were composed mainly of cells with a moderate number of secretory granules, with a diameter ranging between 200-300 nm. Extruded secretory granules were frequently seen. In these tumors, some cells had rare

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.

PITUITARY

ADENOMAS

IN GRH TRANSGENIC

MICE

Endo. ‘Jo1131

l

1992 No 5

FIG. 2. Immunohistochemistry and in situ hybridization of pituitary adenomas in GRH transgenic mice. a, The majority of tumor cells have strong cytoplasmic GH immunoreactivity. Avidin-biotinperoxidase complex technique; magnification, x140. b, By in situ hybridization, the signal for GH mRNA is found diffusely in tumor cells, but the strength of the signal (i.e. the number of silver grains) varies from cell to cell. Magnification, X230. c, In some transgenic animals, PRL immunoreactivity is found in scattered tumor cells (arrows), with variable intensity and diffuse cytoplasmic positivity. In contrast, the adjacent nontumorous adenohypophysis (bottom) has cells with strong juxtanuclear globular positivity. Avidin-biotin-peroxidase complex technique; magnification, X140. d, In situ hybridization confirms strong signal for PRL mRNA in the nontumorous adenohypophysis (bottom) and identifies scattered cells within the adenoma that express PRL mRNA (arrows). Magnification, x230. e, Intense staining for TSHfl is found in clusters of tumor cells in some adenomas. Avidinbiotin-peroxidase complex technique; magnification, X140.

small secretory granules and had the same ultrastructural features as those of the two sparsely granulated adenomas. Control animak

Nontransgenic littermates served as controls in this study; in 15 control mice, aged 12-27 months, there was no evidence of adenohypophysial hyperplasia or neoplasia on examination by histology, immunohistochemistry, or electron microscopy. Discussion Hormonal stimulation may play a role in the development of several neoplasms(4). Hormone dependenceis considered to be important in the development and progression of carcinomasarising in breast, endometrium, and prostate (1517); among endocrine organs, TSH has been implicated as a stimulus of the growth of thyroid tumors (18), sometumors of the adrenal cortex are thought to be dependent on ACTH stimulation (19), and in rodents, estrogen administration results in pituitary lactotroph hyperplasia and adenoma (20). Functional insufficiency resulting in chronic compensatory overstimulation has been implicated in the formation and

growth of thyroid, adrenal, and gonadal neoplasms(18, 19, 21, 22). The lossof feedback inhibition may account for the development of parathyroid adenomasin tertiary hyperparathyroidism (23) and of somepituitary adenomascomposed of corticotrophs, thyrotrophs, or gonadotrophs (6); the complex regulation of these cells raisesthe possibility that their proliferation is modulated by sustained hormonal stimulation. Previous studies have proven that stimulation of rat adenohypophysial cells by GRH in vitro causes somatotroph proliferation, as measured by [3H]thymidine uptake (24); however, there was no direct proof that protracted GRH excessis implicated in adenoma formation. Our preliminary report of smaller numbers of animals brought attention to this possibility (12). In this report we provide a model that confirms the tumorigenic potential of prolonged excessive hormonal stimulation on the development of pituitary adenomas. There was no other factor in the gene construct inserted into the transgenic animals that can be implicated in neoplastic transformation. Pituitary adenomas are not known to occur spontaneously in the parent strains of the hybrid mouse strain studied (Russell,J. D., SimonsenLaboratories, Inc., Gilroy, CA, personalcommunication), and control animals had no evidence of adenohypophysial hyperpla-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.

PITUITARY

ADENOMAS

IN GRH TRANSGENIC

MICE

FIG. 3. Ultrastructure of mammosomatotroph adenomas in GRH transgenic mice. a, The majority of tumors are composed of large, densely granulated cells with well developed rough endoplasmic reticulum (*) and large Golgi complexes (G). The numerous secretory granules are spherical, with high electron density. These are features of somatotrophs; however, there are subtle differences indicating mammosomatotroph differentiation, including pleomorphic secretory granules (arrows) and extrusion of secretory material at the lateral cell borders (arrowheads). Magnification, ~3,060. b, Fusion of secretory granules gives rise to large deposits of secretory material, and extrusion of these deposits into the intercellular space is prominent in this tumor, which is a characteristic mammosomatotroph adenoma. Magnification, X6,730. c, The immunogold technique confirms the presence of GH (10 nm gold particles) and PRL (40 nm gold particles) in the secretory granules and extruded deposits of secretory material in this mammosomatotroph adenoma. Magnification, ~13,910.

FIG. 4. Ultrastructure of sparsely granulated adenomas in GRH transgenic mice. a, Some animals harbored tumors composed of sparsely granulated cells with abundant cytoplasm containing parallel arrays of rough endoplasmic reticulum (*), prominent Golgi complexes that almost fill the cytoplasm (G), and few small secretory granules; these cells resemble cells of the glycoprotein hormone cell line. Magnification, ~2,725. b, Some of the sparsely granulated cells harbor abnormal secretory granules of an angular or rectangular shape, most often in the Golgi region. Magnification, x13,490.

sia or neoplasia. In contrast, in the pituitaries of these 13 transgenic animals, there were lesionsthat fulfilled the morphological criteria of adenoma unequivocally (6); there was a discrete, well demarcated nodule compressing adjacent tissue, the reticulin pattern was disrupted, and the cellular composition was homogeneous and distinctively different from the remainder of the gland. Thus, it is reasonable to assumethat the prolonged exposure to GRH in these old transgenic mice plays a major role in the development of pituitary adenoma. The morphological appearances of these adenomas are quite variable. In younger, GRH transgenic mice, there was evidence of hyperplasia of somatotrophs, lactotrophs, and mammosomatotrophs(11). The tumors that developed in the

background of similar hyperplasia all contained GH and expressedGH mRNA; in addition, some had PRL immunoreactivity and demonstrated ultrastructural features thought to reflect mammosomatotrophic differentiation. A few expresseda-subunit, and a smaller number contained TSHP. The ultrastructure of some of these resembled mammosomatotrophs; this tumor type may produce a-subunit (25) and TSHP (26). The combination of GH, PRL, and TSH has been reported in a number of human tumors, and the ultrastructural morphology has varied from that of somatotrophs to that of thyrotroph-like cells (26, 27). A similar range of ultrastructural morphology was documented in these transgenie mice. The relationships underlying these different cell types may be directly due to GRH stimulation or may impli-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.

2088

PITUITARY

ADENOMAS

IN GRH

cate other factors, such as the POU homeodomain tissuespecific transcription factor Pit-l, which is expressed in somatotrophs, lactotrophs, and thyrotrophs (28). Our study provides evidence for the first time that introduction of a single hormone gene into the genome can lead to tumor formation. It should be stressed, however, that other pathogenetic factors may also be necessary for neoplastic transformation in these animals. According to the multistep theory of carcinogenesis (29), an irreversible initiating event is required to permanently alter the genome of a cell and predispose it to the neoplastic phenotype, while promotion, long term stimulation of cell proliferation, is necessary for an initiated cell to express the transformed phenotype. More work is needed to clarify the role of GRH

in the multistep processthat occurs during pituitary tumorigenesis. GRH may promote the development of tumors by increasing the population of proliferating cells that are susceptible to oncogenic factors or mutation. GRH may also act as an initiator indirectly by stimulating CAMP biosynthesis and/or its sequela, such as expression of the c-fos protooncogene (30). GRH-stimulated cell proliferation may be mediated by other factors, such as various growth factors that are known to exist in the anterior pituitary (31, 32); those substances may also be involved in transformation to the

neoplastic phenotype (33, 34). Alternatively, that cell transformation

it is possible

may be the direct effect of long term

GRH stimulation. The neoplasms in these mice transgenic

for GRH that would simulate the protracted stimulation only in affected cells. It has been shown that some human pituitary somatotroph adenomas contain mutations of the G, proteins, resulting in altered adenylate cyclase activity anal-

ogous to continuous GRH stimulation (35). Further studies of models such as these transgenic mice may clarify the roles and hormonal

stimulation

Acknowledgments J. Price, The Salk Institute, Biotechnology/Industrial Associates, Inc. (San Diego, CA), prepared the transgenic animals. The authors acknowledge the technical assistance of Z. Cheng, D. Lietz, R. Logan, N. Nelson, and F. Rotondo, and the secretarial help of Colette Drvodelic. References 1. Asa SL, Kovacs K 1984 Development and proliferation of adenohypophysial cells. In: Falkmer &S, Hikanson R, Sundler F (eds) Evolution and Tumour Patholonv of the Neuroendocrine Svstem. , Elsevier, Amsterdam, pp 399-4s 2. Melmed

S, Braunstein

GD,

Horvath

E, Ezrin

C, Kovacs

K 1983

MICE

Endo * 1992 Voll31. No 5

Pathophysiology of acromegaly. Endocr Rev 4:271-290 3. Molitch ME 1987 I’athogenesis of pituitary tumors. Endocrinol Metab Clin North Am 16:503-527 4. Asa SL, Kovacs K 1991 Pathogenesis of endocrine tumors. In: Kovacs K, Asa SL (eds) Functional Endocrine Pathology. Blackwell, Boston, pp 1005-1013 5. Asa SL 1991 The role of hypothalamic hormones in the pathogenesis of pituitary adenomas. Path01 Res Pratt 187:581-583 6. Kovacs K, Horvath E 1986 Tumors of the Pituitary Gland. fascicle 21, ser 2. Armed Forces Institute of Pathology, Washington DC 7. Reichlin S 1987 Control of GH secretion: an overview. In: Liidecke DK, Tolis G (eds) Growth Hormone, Growth Factors, and Acromegaly. Raven Press, New York, pp l-11 8. Sano T, Asa SL, Kovacs K 1988 Growth hormone-releasing hormone-producing tumors: clinical, biochemical, and morphological manifestations. Endocr Rev 9:357-373 9. Asa SL, Scheithauer BW, Bilbao JM, Horvath E, Ryan N, Kovacs K, Randall RV, Laws Jr ER, Singer W, Linfoot JA, Thorner MO, Vale W 1984 A case for hypothalamic acromegaly: a clinicopatho-

logical study of six patients with hypothalamic gangliocytomas producing growth hormone-releasing factor. J Clin Endocrinol Metab 58:796-803 10. Mayo

KE, Hammer RE, Swanson LW, Evans RM 1988 Dramatic pituitary

Brinster

RL,

Rosenfeld

hyperplasia in transgenic mice expressing a human growth hormone-releasing factor gene. Mol Endocrinol2:606-612 MG,

11. Stefaneanu trup

L, Kovacs K, Horvath E, Asa SL, Losinski NE, BillesN, Price J, Vale W 1989 Adenohypophysial changes in mice

transgenic for human growth hormone-releasing factor: a histological, immunocytochemical, and electron microscopic investigation. Endocrinology 125:2710-2718

12. Asa SL, Kovacs

K, Stefaneanu C, Vale W

L, Horvath

E, Billestrup

N, Gon-

1990 Pituitary mammosomatotroph adenomas develop in old mice transgenic for growth hormonereleasing hormone. Proc Sot Exp Biol Med 193:232-235 zalez-Manchon

for GRH arose in association with diffuse hyperplasia due to prolonged hormonal stimulation. In contrast, most human pituitary GHproducing adenomas are solitary lesions in a nonhyperplastic gland (6). Although it is unlikely that the substantial majority of human neoplasms are the result of excess hormonal stimulation, the oncogenic potential of GRH shown by this model suggests that GRH and possibly other stimulating peptides may play a role in adenoma formation in humans. The mechanism may involve abnormalities of hormone receptors

of the various oncogenic agents in endocrine tumorigenesis.

TRANSGENIC

13. Stefaneanu L, Rindi G, Horvath E, Murphy D, Polak JM, Kovacs K 1992 Morphology of adenohypophysial tumors in mice transgenic for vasopressin-SV40 hybrid oncogene. Endocrinology 130:17891795 14. Asa SL, Kovacs K, Horvath E, Losinski NE, Laszlo FA, Domokos I, Halliday WC 1988 Human fetal adenohypophysis. Electron microscopic and ultrastructural immunocytochemical analysis. Neuroendocrinology 48:423-431 15. Kirschner MA 1977 The role of hormones in the etiology of human breast cancer. Cancer 39:2716-2726 16. Salmi T 1979 Risk factors in endometrial carcinoma with special reference to the use of estrogens. Acta Obstet Gynecol Stand [Suppl] 86:1-119 17. Smith Jr JA 1987 New methods of endocrine management in prostatic cancer. J Urol 137:1-10 18. Williams ED 1979 The aetiology of thyroid tumours. Clin Endocrino1 Metab 8:193-207 19. Neville AM, O’Hare MJ 1982 The Human Adrenal Cortex. Pathology and Biology-An Integrated Approach. Springer-Verlag, Berlin 20. Furth J, Ueda G, Clifton KH 1973 The pathophysiology of pituitaries and their tumors: methodologic advances. In: Busch H (ed) Methods in Cancer Research. Academic Press, New York, pp 201277 21. Newell ME, Lippe BM, Ehrlich RM 1977 Testicular tumors associated with congenital adrenal hyperplasia: a continuing diagnostic and therapeutic dilemma. J Ural 117:256-258 22. Neubecker RD, Theiss EA 1962 Sertoli cell adenomas in patients with testicular feminization. Am J Clin Path01 38:52-59 23. AkerstrGm G, Malmaeus J, Grimelius L, Ljunghall S, Bergstriim R 1984 Histological changes in parathyroid glands in subclinical and clinical renal disease. An autopsy investigation. Stand J Ural Nephrol 18:75-84 24. Billestrup N, Swanson LW, Vale W 1986 Growth hormone-releasing factor stimulates proliferation of somatotrophs in vitro. Proc Nat1 Acad Sci USA 83:6854-6857 25. Asa SL, Kovacs K, Horvath E, Singer W, Smyth HS 1992 Hormone

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.

PITUITARY

ADENOMAS

secretion in vitro by plurihormonal pituitary adenomas of the acidophil cell line. J Clin Endocrinol Metab 75:68-75 26. Horvath E, Kovacs K, Scheithauer BW, Randall RV, Laws Jr ER, Thorner MO, Tindall GT, Barrow DL 1983 Pituitary adenomas producing growth hormone, prolactin, and one or more glycoprotein hormones: a histologic, immunohistochemical, and ultrastructural study of four surgically removed tumors. Ultrastruct Path01 5:171183 27. Simard M, Mire11 CJ, Pekary AE, Drexler J, Kovacs K, Hershman JM 1988 Hormonal control of thyrotropin and growth hormone secretion in a human thyrotrope pituitary adenoma studied in vitro. Acta Endocrinol (Copenh) 119:283-290 28. Crenshaw EB, Kalla K, Simmons DM, Swanson LW, Rosenfeld MG 1989 Cell-specific expression of the prolactin gene in transgenic mice is controlled by synergistic interactions between promoter and enhancer elements. Genes Dev 3:959-972

IN GRH

TRANSGENIC

MICE

29. Farber E 1981 Chemical carcinogenesis. N Engl J Med 305:13791389 30. Billestrup N, Mitchell RL, Vale W, Verma IM 1987 Growth hormone-releasing factor induces c-fos expression in cultured primary pituitary cells. Mol Endocrinol 1:300-303 31. Ezzat S, Melmed S 1990 The role of growth factors in the pituitary. J Endocrinol Invest 13:691-698 32. Webster J, Ham J, Bevan JS, Scanlon MF 1989 Growth factors and pituitary tumors. Trends Endocrinol Metab 1:95-98 33. Cross M, Dexter TM 1991 Growth factors in development, transformation, and tumorigenesis. Cell 64:271-280 34. Heldin C-H, Westermark B 1989 Growth factors as transforming proteins. Eur J Biochem 184:487-496 35. Vallar L, Spada A, Giannattasio G 1987 Altered G, and adenylate cyclase activity in human GH-secreting pituitary adenomas. Nature 330:566-568

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 July 2015. at 21:10 For personal use only. No other uses without permission. . All rights reserved.