Mammotropes and somatotropes in the adenohypophysis of androgenized female miceMorphological and immunohistochemical studies by light microscopy correlated with routine electron microscopy.код для вставкиСкачать
THE ANATOMICAL RECORD 233:103-110 (1992) Mammotropes and Somatotropes in the Adenohypophysis of Androgenized Female Mice: Morphological and lmmunohistochemical Studies by Light Microscopy Correlated With Routine Electron Microscopy AKIKO YAMAJI, FUMIHIKO SASAKI, YOSHIE IWAMA, AND SHOJI YAMAUCHI Department of Veterinary Anatomy, University of Osaka Prefecture College of Agriculture, Sakai, Osaka, Japan (A.Y.,F.S., S.Y.); and Department of Anatomy, Aichi Medical University, Aichi, Japan (Y.I.) ABSTRACT Female mice were divided into androgenized (AF) and control (CF) groups. The AF mice were injected subcutaneously with testosterone propionate (Tp) and the CF mice with sesame oil a t 5 days of age. Mammotropes (PRL cells) and somatotropes (GH cells) in the adenohypophyses of these mice when they became adults were studied with immunohistochemistry and morphometry by light microscopy correlated with routine electron microscopy. In CF mice, almost all of the PRL-immunoreactive cells (about 43% of all parenchymal cells) were type I (classical) PRL cells, and almost all of the GH-immunoreactive cells (about 30% of all parenchymal cells) were type I (classical) GH cells. Type I1 PRL cells accounted for about 0.4% of parenchymal cells, and type I1 GH cells were about 2.5% of all parenchymal cells. In AF mice, the percentages of PRL and GH cells were not significantly different from those of CF mice. Mammosomatotropes (Ms cells) in both groups were less than 1%of all parenchymal cells. Numbers of all parenchymal, PRL and GH cells, however, were increased significantly in AF mice when compared to those in CF, because the adenohypophysis was increased in volume in AF mice. Type I PRL cells were larger in AF than in CF. The ultrastructure suggested that type I PRL cells may show increased PRL synthesis and secretion in AF mice. Furthermore, AF mice, in which the hypothalamus is masculinized by the neonatal treatment with Tp, retained feminine characteristics in the population and size of PRL cells and GH cells in the adenohypophysis. 0 1992 Wiley-Liss, Inc. There is a clear sex difference in adult mouse and rat hypothalamus due to the absence or presence of testosterone during the neonatal period: the hypothalamus in female and male types controls the cyclic and tonic secretion, respectively, of gonadotropins (Gn) from the anterior pituitary gland (Pfeiffer, 1936; Harris, 1964; Barraclough, 1967; Gorski, 1968; Yazaki, 1968; MacLusky and Naftolin, 1981). Barraclough and Leathem (1954) reported that the administration of testosterone propionate (Tp) to neonatal female mice causes the hypothalamus to become the male type and results in infertility characterized by polycystic ovaries that have numerous follicles of various sizes but no corpora lutea. In such androgenized animals, prolactin (PRL) levels in serum (Mallampati and Johnson, 1973, 1974; Cheng and Johnson, 1973174;Harlan and Gorski, 1977; Nakamura et al., 1988) and pituitary (Mallampati and Johnson, 1974) were reported to be increased significantly when compared with those of controls. However, no morphological evidence on the adenohypophysis in androgenized animals has been presented, to our knowledge. This study was undertaken to get new information on the influence of androgen on pituitary morphology in immature female mice, using immunohistochemistry by light microscopy correlated 0 1992 WILEY-LISS, INC. with routine electron microscopy in a stereological morphometric study. MATERIALS AND METHODS Animals Primiparous mice of the SMA strain, bred in our animal quarters, were housed in cages under controlled conditions with a constant temperature (24"C), 14 h r light, 10 h r dark cycle, with food and water ad libitum. To maintain uniform body growth, all of the litters were reduced to 6 pups on the day of parturition and weaned on day 25. Ten female pups were injected subcutaneously with Tp (1.00 mg) a t 5 days of age a s described by Barraclough and Leathem (1954). Ten female control mice received only injections of sesame oil. All animals were killed a t 3 months of age. Light Microscopic Study To observe the development of ovarian follicles and corpora lutea, ovaries of all mice were fixed for 22-24 Received February 11, 1991; accepted Setpember 25, 1991. Address reprint requests to Dr. F. Sasaki, Department of Veterinary Anatomy, University of Osaka Prefecture College of Agriculture, Mozuume-machi, Sakai, Osaka 591, Japan. 104 A. YAMAJI ET AL TABLE 1. Percentages (means f S.D.) of PRL cells and GH cells in the adenohypophysis of control and androgenized mice’ Group of mice (No.) Control (5) I Androgenized (5) Types of PRL cells (%) I1 I 43.3 2 7.7 40.0 2 4.4 0.4 t 0.3 0.3 0.1 * Types of GH cells (%) + I1 43.8 40.3 * 3.4 * 2.0 I TI 30.5 2 3.9 27.8 2 4.1 2.5 2 1.1 5.2 2 3.6 T + TT 33.1 t 3.6 33.0 t 5.4 Ms cells (%) 0-0.4 0.2-1.0 ~ ‘All data between two groups were analyzed by Student’s t test and Mann-Whitney’s U test. TABLE 2. Numbers ( x Group of mice (No.) Control (5) Androgenized (5) **P< 0.001. lo4) of PRL cells and GH cells (means f S.D.) in the adenohypophysis of control and androgenized female mice Types of PRL cells ( x lo4) I1 I + I1 Parenchymal cells ( x lo4) I 41.3 4.5 78.4 2 9.9** 17.9 2 4.1 30.6 2 6.0* * 0.1 t 0.1 0.3 2 0.1* 18.1 i- 4.0 31.0 5 6.0* Types of GH cells ( x r * 12.5 1.9 21.9 t 4.4* TT 1.0 t 0.4 4.0 t 2.8 ~~~ lo4) T + IT ~ 13.6 t 1.8 25.9 i- 5.0** *P < 0.01 when compared with control female mice by Student’s t test. h r in Bouin’s fluid, dehydrated in a graded series of alcohols, and embedded in paraffin wax. Serial sections (10 pm) were cut and stained with hematoxylin and eosin. To estimate the volume of the anterior pituitary gland, the glands were obtained from 5 control (CF) and 5 androgenized animals. After fixation in Bouin’s fluid for 10 min in situ, the glands were carefully excised from the base of the skull. Then they were fixed for 24 h r in the same fixative, dehydrated in a graded series of alcohols, and embedded in paraffin wax. Serial sections (10 pm in thickness) were cut and stained with hematoxylin and eosin. Areas of the anterior pituitary gland (Al, A,, . . ., A,) were measured with a Profile projector (V-10, Nikon, Tokyo, Japan) and a Photopattern analyzer (PC-8001, Nippon Electric, Tokyo, Japan). The volume (V) of the anterior pituitary gland (pm3) was expressed as V = 10 x A pm3, as described in a previous study (Sasaki and Sano, 1982). The specificity of the immunohistochemical staining was tested as follows. Normal rabbit serum was substituted for specific antiserum, and PRL and growth hormone (GH) antisera were preabsorbed for 1day by 25 pg/ml mouse PRL provided by Dr. K. Komoto (University of Tokyo) and 25-50 pg/ml rat GH (UCB Bioproducts, Belgium), respectively. xi Electron Microscopic Study The anterior pituitary glands from 10 mice (5 from each group) were used for electron microscopy. To obtain electron micrographs evenly from all parts of the anterior pituitary gland, a stereological morphometric method described previously was used (Sasaki, 1974). All pituitary glands were carefully excised from the base of the skull and cut in the midsagittal plane. The hemipituitary gland was discarded. To get good cytoplasmic preservation, the other half was diced into 6 small pieces, a s described previously (Sasaki, 1974). All of the blocks were singly fixed in OsO, for 1 h r a t 4°C. This fixation method gives better exocytotic figures than one in which glutaraldehyde was first used. The blocks wre dehydrated in a graded series of ethanol solutions and embedded in Epon for 1 day at 60°C. About 50 electron micrographs from each animal were obtained. Parenchymal cells exhibiting a nucleus in the micrograph totalled about 1,000 in each animal. Two serial semithin sections adjoining each electron micrograph were stained immunohistochemically by antibodies to PRL or GH a s described previously (Sasaki and Iwama, 1988). PRL-immunoreactive cells were classified into type I PRL cells (classical type), which contained irregularly shaped secretory granules, and type I1 PRL cells, which contained small (100-200 nm in diameter), round secretory granules. GH-immunoreactive cells included type I GH cells (classical type) which contained large (about 350 nm in diameter) round secretory granules, and type I1 GH cells containing small (about 100-200 nm in diameter) round secretory granules. Mammosomatotropes (Ms cells) were identified as those immunoreactive with both anti-PRL and anti-GH; they contained small (about 100-200 nm) round secretory granules, a s described previously (Sasaki and Iwama, 1988). Morphometry Cell counts were made of parenchymal, PRL, GH, and Ms cells on all electron micrographs described above. All cells were identified by immunohistochemistry. The population of PRL and GH cells was expressed as a percentage and a number. The number of parenchymal cells in each anterior pituitary gland was calculated a s reported previously (Sasaki and Sano, 1982). A cellophane sheet describing a 15 cm unit square (the value of 15 cm corresponds to 30 pm in an Fig. 1. The anterior pituitary gland of a control female (CF) mouse. A Type I PRL cells (nos. 3, 6-9, 12-15) contain irregularly shaped secretory granules. Type I GH cells (nos. 2 , 5 , 11) contain large (about 350 nm in diameter) round secretory granules. Secretory granules in exocytosis (arrow) are seen in a PRL cell. B: Stained with GH antiserum. C: Stained with PRL antiserum. D: A higher magnification of the area indicated by the arrow in A. A, x 5,000; B,C, x 1,200; D, x 20,000. Fig. 1. 106 A. YAMAJI ET AL. Fig. 2. The anterior pituitary gland in a n androgenized female (AF) mouse. A: Type I PRL cells (nos. 2, 3) in AF contain generally more abundant rough endoplasmic reticulum, Golgi apparatuses, and secretory granules compared with those of CF mice. Secretory granules in exocytosis (arrows) in a PRL cell are seen. Type I GH cell (no. 1) in the AF mouse shows no morphological differences compared with those in the CF mouse. B: Stained with GH antiserum. C: Stained with PRL antiserum. D,E: Higher magnification views of the areas indicated by the arrows in A. A, x 5,000; B,C, x 1,200;D,E, x 20,000. electron micrograph of x 5,000) was superimposed onto all electron microgrlaphs obtained. For each animal, the mean number (N) of parenchymal cells with a nucleus in this square (900 em2) was counted and transformed into the number (N3”) of cells in the unit cube (27,000 I J . ~ with ~ ) the sides of 900 pm2. The number of parenchymal cells (T) was calculated by substituting these values- into the formula of Sasaki and Sano (1982): T = N3’2 x V/27,000. Numbers of GH and PRL cells were calculated by multiplying T by the respective percentage of the types of cells obtained above. To measure the sizes of type I PRL and GH cells, 50 cells of each type per animal showing a nucleus and with the entire profile of the cell on the electron micrographs were chosen a t random. Each cellular area was measured with a photopattern analyzer. The computer automatically read the areas of individual cells. Areas for each mouse were averaged, and the data reported here are averages of data from 5 mice. To compare the PRL synthetic function in PRL cells between CF and AF animals, the areas of rough endoplasmic reticulum (RER) and of cytoplasm, except the nucleus, in 30 cells/mouse pituitary were measured a t the final magnification of x 10,000 by a point-counting ~ Fig. 3. The anterior pituitary gland in an androgenized mouse. A. Type I PRL cells (nos. 1-8) contain generally more rough endoplasmic reticulum, Golgi apparatuses, and secretory granules compared with those in the CF mouse. Secretory granules in exocytosis (arrows) occur more frequently than in the CF mouse. B-H: Higher magnification views of areas indicated by arrows in A. A, ~ 5 , 0 0 0 ;B-H, x 20,000. PRL AND GH CELLS IN ANDROGENIZED MICE Fig. 3. 107 108 A. YAMAJI ET AL. method, using a square lattice of points with a spacing of 1 cm (in this case, the number of points corresponds to the real number in pm2). RER is a good index for expressing the PRL synthetic function morphologically in PRL cells. The data were the averages of the data from 5 mice. The data were expressed as the absolute area (pm2)of RER or the relative area (%) of RER per the area of the cytoplasm in a PRL cell. Statistics Student's t test and Mann-Whitney's U test were used to compare the data from the two groups. RESULTS Light Microscopic Observation of Ovaries Ovaries in CF mice contained many follicles and corpora lutea, while those in AF mice had numerous follicles of various sizes but no corpora lutea. Morphometry Body weight and volume of anterior pituitary gland No difference in the body weight was found between these two groups [CF: 25.5 2 2.2 g (mean t S.D.); AF: 25.1 t 2.0 gl. The volume of the anterior pituitary gland, however, in AF mice [1.3 +- 0.2 mm3 (mean S.D.)] was significantly larger (P < 0.01) than that of CF mice (0.8 c 0.1 mm3). * PRL, GH, and Ms cells Table 1 shows the percentages of PRL, GH, and Ms cells. In CF mice, almost all of the PRL-immunoreactive cells (about 43% of all parenchymal cells) were type I. Almost all of the GH-immunoreactive cells (about 30%of all parenchymal cells) were type I. Only a few type I1 PRL cells (about 0.4%) and type I1 GH cells (about 2.5%) were found. No differences in percentage were found when PRL and GH cells in AF mice were compared with those in CF. Ms cells were fewer than 1% of all parenchymal cells in both groups of mice. The numbers of parenchymal, PRL, and GH cells are presented in Table 2. In AF mice, the numbers of parenchymal (P < 0.001), of PRL (P < 0.011, and of GH cells ( P < 0.001) were markedly increased when compared with those in CF mice. The type I PRL cells were larger (P < 0.01) in AF mice (107.7 t 2.0 pm2) than in CF mice (99.5 ? 2.4 pm2). There was no difference in the size of type I GH cells between AF (118.8 t 9.5 pm2) and CF mice (124.9 10.1 pm2). Type I PRL cells in AF mice also contained larger (P < 0.01) areas of RER than those in CF mice: 27.4 5 2.9 pm2 vs. 17.8 ? 3.6 pm2 in absolute area, and 38.0 t 4.6% vs. 27.2 2 3.0% in the percentage of RER in the cytoplasm of a PRL cell. * Ultrastructural Changes In AF mice (Figs. 2, 3), type I PRL cells contained more abundant secretory granules and rough endoplasmic reticulum and a better-developed Golgi complex than those in CF mice (Fig. 1).The type I PRL cells also included more secretory granules in exocytosis in AF mice (Figs. 2,3) than those in a CF mouse (Fig. 1).Type I GH cells in both AF and CF mice contained a number of large (about 350 n m in diameter) and round granules. Although some difference was observed in the content of small (100-200 nm in diameter) round secretory granules and of cell organelles among individual cells for each group, no clear morphological differences were found in type I1 PRL cells, type I and I1 GH cells, or Ms cells between AF and CF mice. DISCUSSION This morphological study showed that synthesis and secretion in PRL cells of the adenohypophyses in androgenized female mice increased significantly compared with that in controls. Thus, PRL cells in AF increased in number and size, in the frequency of secretory granules in exocytosis, and in the development of cell organelles as compared with those in CF. Radioimmunoassay data also has demonstrated that PRL synthesis and secretion in the adenohypophysis are greater in AF than in CF mice (Mallampati and Johnson, 1973, 1974; Cheng and Johnson, 1973174; Harlan and Gorski, 1977; Nakamura et al., 1988). In this study, the numbers of GH a s well as PRL cells in AF mice increased significantly over control levels. Within spontaneous pituitary tumors in aging female mice, the predominant cell types were PRL and GH cells (Schechter et al., 1981). Thus, the adenohypophysis in AF resembled that of this pituitary tumor in the increase in the number of PRL and GH cells. It has been shown that the anterior pituitary gland differentiates into the female type or male type due to the absence or presence of a testis (or testosterone) during the neonatal period and then due to the presence or absence of a n ovary (or estrogen andlor progesterone) after puberty. A clear sex difference is observed in the population and size of PRL cells and GH cells in the anterior pituitary gland of adult mice; PRL cells are more abundant and larger in female mice than in males, and GH cells are more numerous in males (Sasaki and Sano, 1980, 1982, 1983). By these criteria, the anterior pituitary gland in AF mice is of the female type. It has been well documented that the sex differentiation in the release of gonadotropin-releasing hormones (GnRH) in the hypothalamus occurs during that neonatal period named the critical period, due to the presence or absence of a testis or testosterone (Harris, 1964; Barraclough, 1967; Gorski, 1968; MacLusky and Naftolin, 1981). In fact, serum levels of testosterone in male r a t pups during this period were significantly higher than those in female pups (Pang et al., 1979; Slob et al., 1980; Pang and Tang, 1984). In control female mice, since the hypothalamus becomes the female type, it secretes GnRH cyclically after puberty; then, the adenohypophysis can cyclically secrete gonadotropins (Gn). Thus, the female ovaries cyclically produce follicles and corpora lutea (Gorski, 1968). The estrogen and progesterone secreted cyclically from these ovaries may lead their anterior pituitaries to the condition of the normal female type by stimulating the glands. On the other hand, in AF mice, the administration of testosterone during the critical period leads the hypothalamus to secrete GnRH tonically after puberty; then, the adenohypophysis secretes Gn tonically. Thus, the ovaries of these AF mice contain only follicles but no corpora lutea (Gorski, 1968). Estrogen secreted con- PRL AND GH CELLS IN ANDROGENIZED MICE tinuously from these follicles may lead their adenohypophyses to develop a prolactinoma by stimulating PRL cells in the adenohypophysis. The presence or absence of neonatal testosterone also causes a sexual difference within the hypothalamus that may account for sexually specific PRL release (Demarest et al., 1981; Gunnet and Freeman, 1982). This mechanism may be the same as the one mentioned above. It has been reported that the administration of estrogen induces hypertrophy of the pituitary gland (Wolfe and Wright, 1938; Ratner et al., 1963; Ramiretz and McCann, 1964; Gersten and Baker, 19701, hyperplasia (Hymer et al., 1961; Shimazaki et al., 1962; Gersten and Baker, 1970; Lloyd et al., 1973) and marked development of the endoplasmic reticulum and Golgi complex of PRL cells (Hymer et al., 1961; Shimazaki et al., 1962; Pantie and GenbaEev, 1969; Watari and Tsukagoshi, 1969; Zambrano and Deis, 1970; Stratmann et al., 1974; Coates et al., 1975; Shiino and Rennels, 1976), and hyperactive secretion reflected by an increased blood level of PRL (Ramiretz and McCann, 1964; Shinha et al., 1972; Yamamoto et al., 1975; Gudelsky et al., 1981). On the other hand, progesterone partially inhibits the production (Haug and Gautvik, 1976) and secretion of PRL (Chen and Meites, 1970; Caligaris et al., 1974). Since estrogen and progesterone are secreted alternatively from ovaries in control adult females, progesterone may inhibit their pituitary glands to cause a prolactinoma. In AF mice, PRL cells may originally increase in size and number due to the stimulation of estrogen persistently secreted from polycystic ovaries. GH cells might secondarily increase in number to maintain the anterior pituitary glands as the female type in the percentage of PRL and GH cells, as shown in this study. ACKNOWLEDGMENTS The authors thank Dr. K. Kohmoto (University of Tokyo) for providing mouse PRL antiserum and mouse PRL, and Dr. H. Seo (Nagoya University) for providing GH antiserum. LITERATURE CITED Barraclough, C.A. 1967 Modifications in reproductive function after exposure to hormones during the prenatal and early postnatal period. In Neuroendocrinology.L. Martini and W. F. Ganong, eds. Academic Press, New York, pp. 61-99. Barraclough, C.A., and J.H. Leathem 1954 Infertility induced in mice by a single injection of testosterone propionate. Proc. SOC. Exp. Biol. Med., 85,673-674. Caligaris, L., J.J. Astrada, and S. Taleisnik 1974 Oestrogen and progesterone influuence on the release of prolactin in ovariectomized rats. J. Endocrinol., 60:205-215. Chen, C.L., and J. Meites 1970 Effects of estrogen and progesterone on serum and pituitary prolactin levels in ovariectomized rats. Endocrinology, 86.503-505. Cheng, H.C., and D.C. Johnson 1973174 Serum estrogens and gonadotropins in developing androgenized and normal female rats. Neuroendocrinology, 13:357-365. Coates, P.W., C.A. Blake, D.S. Maxwell, and C.H. Sawyer 1975 Physiological and morphological correlates of increased prolactin secretion in ovariectomized rats treated with estrogen and progesterone. In: Electron Microscopic Concepts of Secretion: Ultrastructure of Endocrine and Reproductive Organs. M. Hess, ed. John Wiley and Sons, New York, pp. 299-315. Demarest, K.T., D.W. McKay, G.D. Riegle, and K.E. Moore 1981 Sexual differences in tuberoinfundibular dopamine nerve activity induced by neonatal androgen exposure. Neuroendocrinology, 32; 108-113. 109 Gersten, B.E., and B.L. Baker 1970 Local action of intrahypophyseal implants of estrogen as revealed by staining with peroxidaselabeled antibody. Am. J . Anat., 128:l-20. Gorski, R.A. 1968 The nueral control of ovulation. In: Biology of Gestation. N.S. Assali, ed. Academic Press, New York, pp. 1-66. Gudelsky, G.A., D.D. Nansel, and J.C. Porter 1981 Role of estrogen in the dopaminergic control of prolactin secretion. Endocrinology, 108:440-444. Gunnet, J.W., and M.E. Freeman 1982 Sexual differences in regulation of prolactin secretion by two hypothalamic areas. Endocrinology, 110:697-702. Harlan, R.E., and R.A. Gorski 1977 Correlations between ovarian sensitivity, vaginal cyclicity and luteinizing hormone and prolactin secretion in lightly androgenized rats. Endocrinology, 101: 750-759. Harris, G.W. 1964 Sex hormones, brain development and brain function: The Upjohn Lecture of the Endocrine Society. Endocrinology, 75:627-648. Haug, E., and K.M. Gautvik 1976 Effects of sex steroids on prolactin secreting rat pituitary cells in culture. Endocrinology, 99.14821489. Hymer, W.C., W.H. McShan, and R.G. Christiansen 1961 Electron microscopic studies of anterior pituitary glands from lactating and estrogen-treated rats. Endocrinology, 69.81-90. Lloyd, H.M., J.D. Meares, and J. Jacobi 1973 Early effects of stilbestrol on growth hormone and prolactin secretion and on pituitary mitotic activity in the male rat. J . Endocrinol., 58:227-231. MacLusky, N.J., and F. Naftolin 1981 Sexual differentiation of the central nervous system. Science, 212,1294-1303. Mallampati, R.S., and D.C. Johnson 1973 Serum and pituitary prolactin, LH, and FSH in androgenized female and normal male rats treated with various doses of estradiol benzoate. Neuroendocrinology, 11:46-56. Mallampati, R.S., and D.C. Johnson 1974 Gonadotropins in female rats androgenized by various treatments: Prolactin as an index to hypothalamic damage. Neuroendocrinology, 15:255-266. Nakamura, S., R. Demura, H. Komatsu, T. Suzuki, K. Jibiki, E. Odagiri, and H. Demura 1988 Ovarian inhibin activity in rats with persistent estrus. Folia Endocrinol., 64,1102-1114 (in Japanese with English summary). Pang, S.F., and F. Tang 1984 Sex differences in the serum concentrations of testosterone in mice and hamsters during their critical periods of neural sexual differentiation. J. Endocrinol., 100t7-11. Pang, S.F., A.R. Caggiula, V.L. Gay, R.L. Goodman, and C.S.F. Pang 1979 Serum concentrations of testosterone, oestrogens, luteinizing hormone and follicle-stimulating hormone in male and female rats during the critical period of neural sexual differentiation. J. Endocrinol., 80,103-110. Pantic, V., and 0. GenbaEev 1969 Ultrastructure of pituitary lactotropic cells of oestrogen treated male rats. Z. Zellforsch., 95:280289. Pfeiffer, C.A. 1936 Sexual differences in the hypophyses and their determination by the gonads. Am. J. Anat., 58:195-225. Ramiretz, V.D., and S.M. McCann 1964 Induction of prolactin secretion by implants of estrogen into hypothalamo-hypophyseal region of female rats. Endocrinology, 75t206-214. Ratner, A., P.K. Talwalker, and J. Meites 1963 Effect of estrogen administration in uiuo on prolactin release by pituitary in uitro. Proc. SOC.Exp. Biol. Med., 112;12-15. Sasaki, F. 1974 Quantitative studies by electron microscopy on the sex-difference and the change during the oestrous cycle in the mouse anterior pituitary. Arch. Histol. Jpn., 37:41-57. Sasaki, F., and Y. Iwama 1988 Sex difference in prolactin and growth hormone cells in mouse adenohypophysis: Stereological, morphometric, and immunohistochemical studies by light and electron microscopy. Endocrinology, 123,905-912, Sasaki, F., and M. Sano 1980 Role of the ovary in the sexual differentiation of prolactin and growth hormone cells in the mouse adenohypophysis during postnatal development: A stereological morphometric study by electron microscopy. J. Endocrinol., 85: 283-289. Sasaki, F., and M. Sano 1982 Role of the ovary in the sexual differentiation of prolactin and growth hormone cells in the mouse adenohypophysis: A stereological morphometric study by electron microscopy. J . Endocrinol., 93.117-121. Sasaki, F., and M. Sano 1983 Role of the ovary in sexual differentiation of lactotrophs and somatotrophs in the mouse adenohypophysis: A stereological morphometric study by electron microscopy. J. Endocrinol., 99:355-360. Schechter, J.E., L.S. Felicio, J.F. Nelson, and C.E. Finch 1981 Pitu- 110 A. YAMAJI ET AL. itary tumorigenesis in aging female C57BLi6J mice: A light and electron microscopic study. Anat. Rec., 199:423-432. Shiino, M., and E.G. Rennels 1976 Recovery of rat prolactin cells following cessation of estrogen treatment. Anat. Rec., 185:31-48. Shimazaki, M., G. Ueda, M. Ito, H. Mukobayashi, and J . Shirakawa 1962 Electron microscopic studies of the estrogen-induced pituitary tumors. Wakayama Med. Reports, 7:l-11. Shinha, Y.N., F.W. Selby, U.J. Lewis, and W.P. Vanderlaan 1972 Studies of prolactin secretion in mice by a homologous radioimmunoassay. Endocrinology, 91:1045-1053. Slob, A.K., M.P. Ooms, and J.T.M. Vreeburg 1980 Prenatal and early postnatal sex differences in plasma and gonadal testosterone and plasma luteinizing hormone in female and male rats. J . Endocrinol., 87:81-87. Stratmann, I.E., C. Ezrin, and E.A. Sellers 1974 Estrogen-induced transformation of somatotrophs into mammotrophs in the rat. Cell Tiss. Res., 152:229-238. Watari, N., and N. Tsukagoshi 1969 Electron microscopic observa- tions on the estrogen-induced pituitary tumor. Gunma Symposia Endocrinol., 6:297-313. Wolfe, J.M., and A.W. Wright 1938 Histologic effects induced in the anterior pituitary of the rat by prolonged injection of estrin with particular reference to the production of pituitary adenoma. Endocrinology, 23200-210. Yamamoto, K., K. Kasai, and T. Ieiri 1975 Control of pituitary functions of synthesis and release of prolactin and growth hormone by gonadal steroids in female and male rats. Jpn. d. Physiol., 25: 645-658. Yazaki, I. 1968 The occurrence of the functional sexual development of the hypothalamo-hypophyseal system. In: Function of Brain and Reproduction. M. Kawakami, ed. Kyodo Press, Tokyo, pp. 41-53 (in Japanese). Zambrano, D., and R.P. Deis 1970 The adenohypophysis of female rats after hypothalamic oestradiol implants: An electron microscopic study. J . Endocrinol., 47:101-110.