Growth

Hormone

and

of Pituitary

Prolactin

Secretion

Adenoma

Jun YOSHIDA, M.D., Naoki KAGEYAMA, M.D., Hisao and Masaki KANZAKI, M.D.** Department

of Neurosurgery,

School of Medicine,

Nagoya

SEO, M.D.,*

University

*The Research Institute of Environmental Medicine , Nagoya University **Koserem Kamo Hospital , Japan Summary

Tissue culture of 20 pituitary adenomas (four acromegalies, 15 non-functioning tumors, one Forbes-Albright syndrome) was performed with the Maxmow double cover glass assembly. Growth hormone and prolactin in the medium were measured during the culture. Morphological feature in vitro was also compared with that of the original tumor.
In chromophobe adenomas of clinically non-functioning, tumor cells had round secretory granules of not only 100, but also 300 mμ in diameter. The study of their tissue culture revealed that these tumor cells were secreting growth hormone in all the cases.
In acromegalic patients, the culture medium a month later still contained more than 1000 ng/ml of growth hormone and also many secretory granules remained in the cultured cells. Thus, it was suspected that the tumor cells in acromegalic patients had autonomous secretory functions of growth hormone.
In pituitary adenomas of Forbes-Albright syndrome and even non-functioning or acromegaly, prolactin secretion from tumor cells was demonstrated. Ovoid, ellipsoid, or irregular secretory granules of 500 to 1000 mμ in acromegalic patients and 150 to 500 mμ in the other cases were observed with an electron microscope. Key word: prolactin,

pituitary

adenoma,

acromegaly,

Introduction

Pituitary adenomas are usually classified into three categories; basophilic, eosinophilic, and chromophobe adenomas by their histologi cal properties with haematoxylin and eosin stain. From the endocrinological standpoint of view, they are divided into two groups, hormon ally functioning and nonfunctioning tumors. The concepts that eosinophilic adenomas are associated with acromegaly or gigantism, basophilic with Cushing's disease, and chro mophobe with hypopituitarism has been re cognized for quite some time. In recent years, however, chromophobe adenomas which had been confirmed histologi cally, were shown to be associated with acro megaly, Cushing's syndrome"), hyperthyroid ism"), Forbes-Albright syndrome'), and FSH producing tumor"). Thus, it appears reason able to assume that some chromophobe ade homas are hormonally functioning. Many electron microscopic studies of normal pituitary adenomas

Forbes-Albright

syndrome,

growth

hormone,

tissue culture.

glands' 7.10,19,23,27,29) and pituitary 13,11,'8.31.32.36) have been reported.

Luse24) and other investigators 13,32)described the sizes of secretory granules to be mainly 300 my in both eosinophilic and basophilic ade nomas and 100 my in most chromophobe adenomas. They also reported some chro mophobe adenomas with secretory granules of 300 my or both 100 and 300 my. On the other hand, histochemical identifica tion and hormone assay in the medium of tissue culture have been performed to investigate the hormone secretion of pituitary adenomas. Secretion or synthesis of hormones was demon strated only in a few cases of chromophobe adenomas'-, 21) To elucidate the discrepancy between these morphological and hormone assay studies on pituitary adenomas, human growth hormone (hGH) and human prolactin (hPr) were mea sured in the medium of tissue culture with the Maximow double cover glass assembly. The autonomy of hGH secretion in acro megaly is also discussed in this report. Materials

Tumor

tissues

from

and

Method

four

acromegalies,

15

non-functioning tumors and one Forbes Albright syndrome were removed by surgery and cultured by the Maximow method. Brain tumors other than pituitary adenoma and mouse cerebellum were also cultured as con trols. Tumor tissues were washed twice with Hank's BSS and cut into 0.5 mm' pieces. Aliquots of two fragments were explanted on to a cover slip previously coated with a gel of recon stituted rat-tail collagens) and fed a single drop of nutrient medium (0.04-0.045 ml), consisting of 50% Eagle's minimum essential medium, 25 % horse serum and 25 % Hank's BSS, also containing 600 mg % glucose, were sealed with paraffin. All preparations were inclubated at 37'C. The culture medium was renewed two or three times a week. The exchanged medium was kept frozen until hormones were assayed. As a rule, 32 slides were made for one case. Histolo gical examinations with May-Gimsa or with Haematoxylin-Eosin stains were carried out at various stages of the culture after daily dynamic observations under the phase contrast micro scope. For electron microscopic observation, samples were fixed with 2.5 % glutaraldehyde and 1 % osmic acid, embedded in vestopal or epon and stained with lead citrate and uranyl acetate. Growth hormone and prolactin levels in the culture medium were determined by radioimmunoassay using the double antibody method. The standards employed for the assay of hGH was Wilhelmi (NIH HS-1895) and prolactin was NIH Friesen # 1. Standard curve using medium was in complete accordance with assay using a buffer. In this system, GH sensitivity was I ng/ml and prolactin was 2.5 ng/ml. Results

I.

Morphological observations

1) in vivo a) light microscopic observation Histological examination of 20 pituitary adenomas with haematoxylin and eosin stain disclosed that two of four acromegalic patients were eosinophilic and the others were mixed adenomas. One case of Forbes-Albright syn drome and 15 cases of non-functioning tumors were chromophobe adenoma (Table 1).

Table

I

NON-FUNCTIONING

E.A.:

Eosinophilic

M.A.:

Mixed

C.A.:

Chromophobe

TUMOR

Adenoma

Adenoma Adenoma

b) electron microscopic observation Electron microscopic study of eight chro mophobe adenomas revealed round secretory granules, 100 my in diameter in all, and 300 my in diameter in some cases. Two eosinophilic adenomas contained secretory granules of 300 to 500 my in diameter diffusely in the cytoplasm. One mixed adenoma, case A3, showed two kinds of cells; one of them contained round secretory granules of about 300 my like eosinophilic adenoma and the other ovoid, ellipsoid, or irregular secretory granules of 500 to 1000 my (Fig. 1), which were similar to prolactin granules of human or animal lacto trophs. Case Cl of non-functioning and case FAI of Forbes-Albright syndrome had similar but smaller granules of 150 to 500 my (Fig. 2). Electron micrograph revealed that these irreg ular granules were formed by the fusion of round granules (Fig. 3). 2) in vitro a) light microscopic observation

Fig. 1. Electron micrograph, adenoma cells of acromegaly. STH : Growth hormone secreting cell with round secretory granules of about 300 mp in diameter. LTH: prolactin secreting cell with ovoid or irregular (right upper) secretory granules of 500 to 1000 mp. x 6500.

Fig. 2.

Electron

shaped

micrograph,

secretory

granules

adenoma

cell

of 150

to 250

of clinically mp

in the

non-functioning cytoplasm.

tumor. 25000.

Ovoid

or rod

Fig.

5.

Chromophobe

culture. H Fig.

3.

Electron

micrograph,

adenoma

cells

syndrome.

Irregular

secretory

Forbes-Albright granules x 10000.

are formed

by the fusion

of round

of

granules.

Under the phase contrast microscope, tumor cells partly showed a cord-like or acinous structure after one or two day of culture (Fig. 4). After three to four days, they formed tubular structures. At the stage of two or three weeks,

&

tumor

H

&

E,

x 60.

with

4.

Phase

Two

day with

contrast

culture a

cord

microscope

cells like

or x 250.

of

chromophobe

acinous

structure.

Similar

tubular

Left

Right:

four

structure

to

:

two

week

week

culture.

the

original

formation.

Chromophobe

micrograph, adenoma

x 100.

culture cells showed similar histological struc ture to that of the original tumor (Fig. 5). After this stage, proliferation of the tubular structure was not vivid but futher culture was stable for more than one month. b) electron microscopic observation Electron microscopic study of chromophobe adenomas revealed some difference between initial and late stages of the culture. As shown in Fig. 6, adenoma cells of three-day culture were rich in secretory granules of 100 and 300 mµ in diameter, but after three weeks they dis appeared from the cytoplasm (Fig. 7). Ultrastructural changes during the culture in eosinophilic adenomas were different from those in chromophobe adenomas. That is, round secretory granules were rich even after three weeks. And they were larger than those of the original tumor. Futhermore, these granular cells had many microvilli and formed the follicular structure (Fig. 8).

Fig. 6. Fig.

adenoma.

E,

granules Electron Secretory

adenoma

adenoma.

Left :

Electron

cells

round

secretory

with

of 100 and 300 m,u in diameter. Right : micrograph of three day culture cells. granules still in the cytoplasm. x 7000.

Table HPr

Fig.

7.

Chromophobe

micrograph,

adenoma.

adenoma

cells

secretory

granules

Electron

micrograph

of

Secretory ~~2600.

granules

disappear

II.

of

100

Left:

have mp

three

in

round

diameter.

Right:

culture the

in tissue

2 culture

fluid

Electron

numerous week

from

levels

cells.

cytoplasm.

Hormone levels in the culture medium

1) growth hormone a) chromophobe adenoma Radioimmunoassay of hGH in the culture medium revealed that the concentration was highly than 25 ng/ml in five cases while it was one to five ng/ml in the other five at the early stage of the culture of 10 chromophobe adenomas. However, hGH concentrations of control groups such as other kind of brain tumor (seven cases) and mouse cerebellum (one case) were all less than 1 ng/ml (Table 1). In the chromophobe adenoma, hGH levels in the medium decreased gradually during the culture. After three weeks, it was not detected in most cases. In this series of tissue culture, it might have been possible that high hGH levels were caused by the contamination of normal pituitary tissue which remained in the peripheral portion of the tumor. In most cases, histological examination of the tumor revealed that the compressed normal pituitary tissue was present in the peripheral portion of the adenoma. To investigate this possibility, both central and peripheral portions of the tumor were cultured separately in five chromophobe ade nomas, case C l , C6, C7, C8, and C9. The peripheral portions yielded higher hGH level than those of the center. Even in the center of the tumor, however, hGH levels were more than 20 ng/ml in three adenomas and 4.1 and 3.5 ng/ml in the other two (Table 1, Fig. 9). b) acromegaly In all of four acromegalic patients, hGH levels remained high through longer periods of

C.A.: Chromophobe Adenoma, E.A.: Adenoma M.A.: Mixed Adenoma

Eosinophilic

culture compared to chromophobe adenoma (Table 1). Two of them showed extremely high levels of hGH of above 880 ng/ml for more than a month. 2) prolactin High hPr levels of more than 50 ng/ml in the culture medium were detected in not only the Forbes-Albright syndrome, but also acromegaly or non-functioning tumor with hyperpro lactinemia (Table 2). Discussion

Electron

microscopic

studies

of

human

pituitary adenoma were reported by Luse in 196224), Schelin in 196232), Fukumitsu and Kageyama in 196313), and other investi gators l4,ls,31,36> From these reports it has been well-known that round secretory granules of 300 my in diameter were demonstrated invariably in both eosinophilic and basophilic adenomas and those of not only 100 but also 300 my in most chromophobe adenomas. Recently it has become possible to identify various hormones in pituitary cells with peroxidase-labeled antibody method26). Using this method, Zimmerman et al. in 1974 de scribed that hGH was located in the tumor cells in seven of eight acromegalic patients and also hPr in 10 of 12 pituitary adenomas with hyperprolactinemia. They also reported that tumor hGH was found in one patient with normal plasma hGH, while tumor hPr was found in one patient with normal plasma hPr. On the other hand, quantitative assay of hormone in the culture medium of pituitary adenomas has been reported by Kohler2 ),

Fig. 8. Electron micrograph, adenoma cells of acromegaly. Lower: Round secretory granules of 300 to 500 mµ in diameter diffusely in the cytoplasm. Upper: Electron micrograph of three week culture cells. Many secretory granules of 400 to 600 mp in diameter, forming follicular structure with microvilli (upper right). x 6500.

Fig. 9. Difference of hGH levels in the culture medium between peripheral portion (normal pitui tary tissue) and central portion (chromophobe adenoma).

Batzdorf4), and Teraoka34). Kohler et al. reported that hGH was detected in all five acromegalic patients and only one out of five chromophobe adenomas. Batzdorf cultured eight chromophobe adenomas and detected hGH in two cases. In our study of 15 clinically non-functioning chromophobe adenomas, significant levels of hGH were demonstrated in all cultured me diums. With the Maximow double cover glass method, pituitary adenoma can be cultured with similar structure to that of the original tumor. This is important in studying the pro perty of the tumor because changes in the nature of tumor cells had been reported in some culture methods'). Another advantage of this method is that the only small amounts of culture medium are required for the amount of explanted tissue. This makes it possible to detect even small amounts hormones in the medium. In our series, hGH concentration in the medium was less than 5 ng/ml in seven of 15

chromophobe adenomas. It may arouse suspi cion as to the accuracy of these values. How ever, GH sensitivity in our assay system was 1 ng/ml, and hGH levels in all adenomas were more than 1 ng/ml as compared with less than 1 ng/ml in all of the control groups. Moreover gradual decrease in hGH levels during the culture occurred in parallel with morphological disappearance of secretory granules. Thus, it seems sure that all the chromophobe adenomas in our series had the capacity to secrete growth hormone. In pituitary adenomas of acromegalic pa tients, histological examination showed them to be either eosinophilic or chromophobe. Elec tron microscopic observation revealed that adenoma cells had secretary granules of 300 to 500 my in diameter diffusely in the cytoplasm. With tissue culture, high hGH was measured in the medium. Recently it has been recognized that hGH hypersecretion in acromegalic patients re sponds to various stimuli to the hypothalamus in some cases. A suppression of hGH levels following glucose ingestion',"'), and its rise subsequent to insulin hypoglycemia, arginine infusion 6,11), and provocation with thyrotropin releasing hormone") have been reported. From these clinical data, Sherman et al. suggested that pituitary tumors could result from chronic hyperstimulation through hypothalamic mecha nisms33), and Daughaday et al. have hypothe sized that a defect in hypothalamic regulation of GH secretion may underlie the development of acromegaly in some patients6). One the other hand, Allen et al. reported two acromegalic cases in which remission with maintenance of normal endocrine function lasting one year was achieved by selective pituitary adenomectomy. They suggested that acromegaly might be due to autonomous pituitary tumor. From experimental data of tissue culture, Kohler et al.") described that adenoma cells from acromegaly seemed to retain such a differ entiated function as producing a hormone in vitro, but normal pituitary cells were likely to exhibit dedifferentiation in the early stages of culture. Batzdorf et al.') reported that ade nomas from acromegaly synthesized and se creted GH independently from hypothalamic control.

In our study of tissue culture, gradual decrease of GH levels in chromophobe adenoma occurred in parallel with morphological dis appearance of secretory granules. In acro megaly, all GH levels were over ten-fold as compared with those in chromophobe adenoma and remained high even after one-month culture. Many secretory granules were also observed in the cytoplasm at this stage. The fact that the capacity to secrete hGH remains longer than that of normal pituitary tissue or non-functioning tumor in vitro suggests that the tumor cells in acromegaly have the capacity to secrete hGH independently from hypo thalamic control. Excessive prolactin secretion is frequently encountered in patients with pituitary tumors'' .12.28.35) While some patients with excessive prolactin secretion show galactorrhea, often no manifestation are recognized clinically. In Forbes-Albright syndrome and also non functioning tumor or acromegaly, secretion of prolactin as well as growth hormone was demonstrated in the culture medium. Electron microscopic study revealed big ovoid, ellipsoid, irregular secretory granules of 500 to 1000 mp in acromegalic patients, while granules of similar shape but of smaller size of 150 to 500 my were noted in non-functioning tumor and Forbes-Albright syndrome. According to elec tron microscopic studies of human pituitary gland, lactotrophs were said to contain ir regular or ovoid homogeneously dense granules as large as 400 to 800 mµ'19'28). Our prolactin secreting adenoma cells are strikingly similar to the above descriptions of human lactotrophs. Thus prolactin secretion from tumor cells is probably

attributed

to these granules.

3)

4)

5)

6)

7)

8)

9)

grant

work was supported from the Ministry

by cancer

of Health

and Welfare.

References 1)

Allen, J. P., Cook, D. M., Greer, M. A., Paxton, H., and Castro, M. A.: Evidence that acromegaly is not a hypothalamic disease. Trans. Assoc. Am. Physicians. 82: 272-283, 1973.

2)

Armen, H., and Tashjian, A. H. Jr.: Animal cell cultures and a sourse of hormones. Biotechnol. and Bioeng. 11: 109-126, 1969.

pituitary tumors. In Kohler, P. O., and Ross, G. T.: “Diagnosisand Treatment of Pituitary Tumors,” Excepta medica, American Elsevier, 1973, pp. 26-34. Daughaday, W. H.: The Adenohypophysis , In Williams, R. H.: “Textbookof Endocrinology”, Saunders, 1974, pp. 31-79. Earll, J. M., Sparks, L. L., and Forsham, P. H.: Glucose suppression of serum growth hormone in the diagnosis of acromegaly. J.A.M.A. 201: 628630, 1967. Forbes, A. P., Henneman, P. H., Criswold, G. C., and Albright, F.: Syndrome characterized by and low FSH. J. clin.

10) Forcin, J. F.: Etudes sur .'hypophyse human au microscope electronique. Path. Biol. (Paris) 14: 893-902,1966. 11)

12)

Forsyth, Francis,

I. A., Besser, G. M., Edwards, C. R. W., L., and Myres, R. P.: Plasma prolactin

activity in inappropriate lactation. Brit. med. J. 3: 225-227,1971. Friesen, H., Webster, B. R., Hwang, P., Guyda, H., Munro, R. E., and Read, L.: Prolactin synthesis and secretion in a patient with the Frobes-Albright syndrome. J. clin. Endocr. 34: 192-199, 1972.

13)

Fukumitsu, T., and Kageyama, N.: Hormone secretion from chromophobe adenoma, Electron microscopic study. Sindan-to-Chiryo (Japan) 51: 8-15, 1963. 14) Gusek, W.: Comparative light and electron microscope studies on pituitary adenoma in acro megaly. Endokrinologie. 42: 257-283, 1962. 15)

research

pituitary tumors. J. Neurosurg. 34: 741-748, 1971. Bornstein, M. B.: Reconstituted rat-tail collagen used as substrate for tissue cultures on coverslips in Maximow slide and in roller tubes. Lab. Invest. 7: 134-137, 1958. Daughaday, W. H., Cryer, P. E., and Jacobs, L. S.: The role of hypothalamus in the pathogenesis of

galactorrhoea, amenorrhoea Endocr. 14: 265-271, 1954.

Acknowledgement

This

Barnes, B. G.: Electron microscope studies on the secretory cytology of the mouse anterior pituitary. Endocr. 71: 618-628, 1962. Batzdorf, U., Gold, V., Matthews, N., and Brown, J.: Human growth hormone in cultures of human

Hrubesch, M., Böckel, K., and Vosberg, H.: Hyperthyreosis caused by a TSH-producing chromophobe hypophyseal adenoma. Verh. Dtsch. Ges. Inn. Med. 78: 1529-32, 1972 (Ger).

16) Irie, M., and Tsuchima, T.: Increase of serum growth hormone concentration following thyro tropin-releasing hormone injection in patients with acromegaly or gigantism. J. clin. Endocr. 35: 97100, 1972. 17)

Jacobs, L. S., Mariz, I. K., and Daughaday, W. H.: A mixed heterologous radioimmunoassay for human prolactin. J. clin. Endocr. 34: 484-490, 1972.

18)

Kageyama, N., Kobayashi, T., Yoshida, J. , Asai, T., Yonezawa, T., and Tsuji, Y.: Hormone secreion from pituitary adenomas. Neurol. Surg. (Japan) 1: 295-309, 1973.

19)

20)

Kansaki, M., and Uei, Y.: An electron microscopic study on the human adenohypophysis. Japan J. Clin. Electron Microscopy 5: Nos 2-3, 1972. Kurosumi, K.: Functional classification of cell type of the anterior pituitary gland accomplished by electron microscopy. Arch. histol. jap. 29: 329362, 1968.

21)

Kohler, Kohler,

P. O., S.

Bridson,

E.:

pituitary adenoma 788, 1969.

22)

23)

24)

25)

26)

27)

28)

29)

W. E.,

Hormone in culture.

Rayford,

production Metabolism

30)

31)

P. L., and by

human 18: 782-

Lawrence, A. M., Goldfine, I. K., and Kirsteins, L.: Growth hormone dynamics in acromegaly. J. clin. Endocr. 31: 239-247, 1970. Lawzewitsch, I., Dickmann, G. H., Amezua, L., and Pardal, C.: Cytological and ultrastructural characterization of the human pituitary. Acta anat. 81: 286-316, 1972. Luse, S. A.: “Electron Microscopy of Brain Tumors in Biology and Treatment of Intracranial Tumors”, Ist Ed. Fields, W. S., & Sharkey, P. C., C. C. Tomas, Springfield, 1962. MaCormic, W. F., and Halmi, N. S.: Absence of chromophobe adenoma from a large series of pituitary tumors. Arch. Path. 92: 231-238, 1971. Nakane, P. K.: Classification of anterior pituitary cell type with immunoenzyme histochemistry. J. Histochem. Cytochem. 18: 9-20, 1972. Paiz, C., and Henniger, G. R.: Electron microscopy and histochemical correlation of human anterior pituitary cells. Am. J. Path. 59: 43-73, 1970. Peake, G. T., McKeel, D. W., Jarett, L., and Daughaday, W. H.: Ultrastructural histologic and hormonal characterization of a prolactin-rich human pituitary tumor. J. clin. Endocr. 29: 13831393, 1969. Rinehalt, J. F., and Fargubar, M. G.: Electron

32)

33)

microscopic studies of the anterior pituitary gland. J. Histochem. Cytochem. 1:93-113, 1953. Salassa, R. M., Kearns, T. P., Kernohan, J. W., Sprague, R. G., and McCarthy, C. S.: Pituitary tumors in patients with Cushing syndrome. J. clin. Endocr. 19: 1523-1539, 1959. Schechter, J.: Electron microscopic studies of human pituitary tumors, I. chromophobe ade nomas. Am. J. Anat. 138: 371-386, 1973. Schelin, U.: Chromophobe and acidophil adenoma of the human pituitary gland, A light and electron microscopic study. Acta path. microbiol. scand. 158: 5-80, 1962. Sherman, L., and Kolodny, H. D.: The hypo thalamus, catecholamine and drug therapy for gigantism and 682-685,1971.

acromegaly.

The Lancet

april

3:

34)

Teraoka, A.: Morphological and functional properties of pituitary adenoma in culture. Neurol. Med. Chir. (Tokyo) 12: 52-63, 1972.

35)

Turkington,

36)

37)

38)

R.

W.:

Secretion

of

prolactin

by

patient with pituitary and hypothalamic tumors. J. clin. Endocr. 34: 159-164, 1972. Wechsler, V. W., and Hossmann, K. A.: Elektro nenmikroskopische Untersuchungen Chromo phober Hypophysen-Adenome des Menschen. Zbl. Neurochir. 26: 105-122, 1965. Woolf, P. D., and Schenk, E. A.: An FSH produc ing pituitary tumors in a patient with hypo gonadism. J. clip. Endocr. 38: 561-568, 1974. Zimmerman, E. A., Defendini, R., and Frantz, A. G.: Prolactin and growth hormone in patients with pituitary adenomas, A correlative study of hormone in tumor and pladma by immunoperoxi dase technique and radioimmunoassay. J. clin. Endocr. 38: 577-585, 1974.

Growth hormone and prolactin secretion of pituitary adenoma.

Growth Hormone and of Pituitary Prolactin Secretion Adenoma Jun YOSHIDA, M.D., Naoki KAGEYAMA, M.D., Hisao and Masaki KANZAKI, M.D.** Departmen...
888KB Sizes 0 Downloads 0 Views