Activin A increases the number of follicle-stimulating hormone cells in anterior pituitary cultures

179

Molecular and Cellular Endocrinology, 69 (1990) 179-185

Elsevier Scientific Publishers Ireland, Ltd. MOLCEL 02245

Activin A increases the number of follicle-stimulating in anterior pituitary cultures Tetsuro

Katayama,

hormone cells

Kunio Shiota and Michio Takahashi

Department of Veterinary Physiology, Veterinary Medical Science, University of Tokyo, I -I -I Yayoi, Bunkyo-ku, Tokyo I1 3, Japan

(Received 1 June 1989; accepted 6 December 1989)

Key words: Activin A; Anterior pituitary; Cell number; Cell density

Summary The mode of action of activin A on the anterior pituitary gland (AP) was investigated using primary cultured cells. The AP cells were cultured with activin A (1 ng/ml) at various cell densities for 24-96 h, and further incubated for 3 h with activin A-free medium. When the cells were pretreated with activin A for 48 h, follicle-stimulating hormone (FSH) secretion during the following 3-h incubation was increased mm ‘, but not at 2 X lo5 or 4 X lo5 cells/200 only at a density of 1 X lo5 cells/200 mm2. A longer pretreatment (96 h) was required in order to induce this response at a density of 2 x lo5 cells/200 mm2. Luteinizing hormone (LH) secretion was not affected by activin A. Thus, the FSH secretory activity of the primary culture of AP was stimulated by activin A in a cell density-dependent manner. Furthermore, it was found that treatment with activin A (10 ng/ml) for 72 h increased the number of immunoreactive FSH cells by 25-418, and that these newly induced FSH cells were low responders to gonadotropin-releasing hormone stimulation. The proportions of immunoreactive LH, thyroid-stimulating hormone, prolactin or growth hormone cells were not affected. From these results, we conclude that activin A increases FSH secretion by changing the cell population of pituitary gonadotropes.

Introduction Activins (A, AB) have been purified to homogeneity and their amino-acid sequences have been determined from the nucleotide sequences of their cDNAs (Ling et al., 1986a, b; Vale et al., 1986).

Address for correspondence: Professor Michio Takahashi, Department of Veterinary Physiology, Veterinary Medical Science, University of Tokyo, l-l-l Yayoi, Bunkyo-ku, Tokyo 113. Japan. This work was supported in part by a Grant-in-Aid for Scientific Research (01304025) from the Ministry of Education, Science and Culture of Japan. 0303-7207/90/$03.50

Activin A, which is a homodimer of PA subunits of the structurally related protein, inhibin A (Mason et al., 1985, 1986; Forage et al., 1986; Mayo et al., 1986; Stewart et al., 1986; Esch et al., 1987) was shown to be identical to erythroid differentiation factor (EDF) isolated from the leukemia cell line THP-1 (Eto et al., 1987; Murata et al., 1988). In contrast to the follicle-stimulating hormone (FSH)-suppressing action of inhibins, activins have been shown to stimulate FSH secretion in cultures of anterior pituitary cells. Therefore, such glycoproteins as inhibins and activins have been thought to play roles in regulating the anterior pituitary gland as endocrine factors. These glycoproteins

0 1990 Elsevier Scientific Publishers Ireland, Ltd.

are also reported to play roles in the regulation of steroidogenesis as paracrine signal substances in the gonads (Chari et al., 1985; Hsueh et al., 1987). The activities of inhibins and activins have been assayed using systems involving primary cultures of rat anterior pituitary cells (Steinberger and Steinberger, 1976; de Jong et al., 1979; Eddie et al., 1979; Scott et al., 1980; Ling et al., 1986a, b; Vale et al., 1986), in which chronic treatment (i.e., 72 h; Vale et al., 1986) of the cells with active fractions effectively altered FSH secretion, whereas short-term treatment had no effect (Ling et al, 1986a, b; Kitaoka et al., 1987). Thus, the mode of action is very different from the rapid action of hypothalamic factors, such as gonadotropin-releasing hormone (GnRH) (Denef and Andries, 1983) and thyrotropin-releasing hormone (TRH) (Shiota et al., 1984). As to the mechanism of action of inhibins, Franchimont et al. (1983) showed that the amount of intracellular CAMP was reduced after addition of inhibin. As for activins, the stimulatory effect of activin A on FSH secretion was shown to be blocked by actinomycin D (Schwa11 et al., 1988). indicating that activin A stimulates FSH biosynthesis as well as FSH release. However, it cannot be explained why inhibins and activins require long-term treatment in order to manifest their effects. Since gonadotropes in the pituitary gland are heterogeneous in their hormone storage (Nakane, 1970; Moriarty, 1975; Childs et al., 1982; Childs, 1983) and in their response to GnRH stimulation (Denef et al., 1978; Lloyd and Childs, 1988), we considered it possible that activins might change the population of gonadotropes. In this study, the effects of activin A on both FSH secretion and the number of immunoreactive FSH cells were examined using rat anterior pituitary cells in primary culture. Materials and methods Animals Male Wistar-Imamichi rats (lo-15 weeks old) obtained from the Imamichi Institute for Animal Reproduction (Ohmiya, Japan) were maintained for at least 1 week in a controlled environment at 23 1. 1°C with a 14-h li~t/lO-h dark cycle (lights on at 05:OO h). The animals had free access to water and standard laboratory chow.

R6?ffpt.7 Activin A (EDF) was a gift from the Central Research Laboratory of Ajinomoto Co. (Kawasaki, Japan). Other materials were obtained from the following sources: Dulbecco’s modified Eagle medium (DMEM), fetal calf serum, deoxyribonuclease and pancreatin from Grand Island Biological Co. (Grand Island, NY, U.S.A.); N-IL-hydroxyethylpiperazine-N ‘-2-ethanesulfonic acid (Hepes), bovine serum albumin (BSA) (fatty acidfree), penicillin and streptomycin from Sigma Chemical Co. (St. Louis, MO, U.S.A.); collagenase (type II) from Worthington Biochemical Co. (Freehold, NJ, U.S.A.): GnRH from Takeda Chemical Industries (Osaka, Japan). Culture method Primary culture of the cells was performed by a modification of the method reported elsewhere (Shiota et al., 1984). Briefly, rat anterior pituitary glands were minced and washed with Hepes buffer (137 mM NaCl, 5 mM KCl, 0.7 mM Na2HP0, and 25 mM Hepes, pH 7.3). The tissue was dispersed by incubation first in Hepes buffer containing 0.4% collagenase, 0.4% BSA, 10 pg/tnl DNase and 0.2% glucose, and then with 10% pancreatin. Average cell viability was more than 95%. DMEM supplemented with 10% fetal calf serum was added to bring the cell concentration to 1 X lo’, 2 X lo5 or 4 X lo5 cells/ml. The cells were plated in sterile tissue culture dishes with 1.0 ml (Nunc 24-well dish, 16 mm in diameter) or 160 yl (Costar 96-well dish, 6.4 mm in diameter) of DMEM and incubated under a saturated atmosphere of 5% CO,-95% air in a CO, incubator at 37°C. 18 h later, the medium was exchanged for a fresh one before the experiments. Activirl A treatments The cells were cultured in DMEM containing 10% fetal calf serum at a density of 2 X lo5 cells/200 mm’ (24-well dish). To examine the influence of cell density on the action of activin A, densities of 1 X lo’, 2 X lo5 or 4 X lo5 cells/200 mm’ were employed. The concentration of activin A was 1 ng/ml/weIl unless otherwise stated. The cells were incubated with activin A for 24-96 h depending on the experiments. In some experiments, after preincubation with activin A, the cells

181 were further incubated in fresh medium without fetal calf serum for 3 h in the absence or presence of GnRH (10 nM). Collection and storage of samples At the end of the incubation, the collected medium was centrifuged (480 X g for 6 min) and the supernate was stored at - 20’ C until assay of FSH and luteinizing hormone (LH). For assay of the intracellular FSH content, the cells were washed with 50 mM phosphate-buffered saline (PBS, pH 7) and scraped off with a rubber spatula into a tube containing 1.0 ml of PBS with 0.1% Triton X-100. The cells were homogenized with a Teflon-glass homogenizer, followed by freezingthawing, and the lysate after centrifugation was stored at - 20 o C until assay. Immunocytochemistry For immunocytochemistry, the cells cultured at a density of 1 x lo5 cells/200 mm* (96-well dish) were incubated with activin A (1.6 ng/well) for 72 h. After washing with DMEM without fetal calf serum, the cells were fixed in Bouin’s fluid (without acetic acid) for 10 min at room temperature and then stained immunocytochemically using peroxidase-antiperoxidase (PAP) reagents (Bio Genex Laboratories, San Ramon, CA, U.S.A.). The primary antibodies and their dilutions were as follows: 1 : 1000 for anti-FSH (NIADDK-antirFSH-S-11), 1 : 1200 for anti-LH (NIDDK-antirLH-S-10) 1 : 200 for anti-thyroid-stimulating hormone (TSH) (NIDDK-anti-rTSH-S-5) 1 : 10,000 for anti-prolactin (PRL) (NIADDK-anti-rPRLIC-3) and 1: 5000 for anti-growth hormone (GH) (i538/002-Ab to rat GH, UCB Bioproducts, Brussels, Belgium). The immunostainings were carried out according to the procedure described in the explanatory note accompanying the kit. Briefly, the sequences and reaction times were strictly fixed as follows: 3% hydrogen peroxide for 10 min; normal goat serum for 5 min; primary antibody (not supplied in the kit) for 1 h; goat antirabbit IgG for 5 min; PAP complex for 5 min; peroxidase substrate (3-amino-9-ethyl-carbazole) for 10 min. All the steps were run at 37°C. The reaction time with the primary antibodies (1 h) was shown in a preliminary examination to be sufficient for obtaining the maximal staining in-

tensity. In the experiments to evaluate the specificity of the staining, normal rabbit serum or the antibody for FSH preabsorbed with 25 pg/ml homologous antigen (NIADDK-rFSH-RP-2) at 4°C for 1 h was used instead of the primary antibodies. More than 200 cells from five independent areas in each of three wells were counted using a light microscope, and the visibly stained cells were considered to be hormone-positive cells. The data were presented as averages for each experimental group. Radioimmunoassay Amounts of FSH and LH were measured in duplicate by double-antibody radioimmunoassay using kits provided by the NIADDK. The data were expressed in terms of NIADDK-rat FSHRP-2 and NIADDK-rat LH-RP-2. Statistics The data are presented as means t_ SE for three different trials each and were evaluated by Student’s t-test. P values of less than 0.05 were considered significant unless otherwise described. Results FSH and LH secretion in the presence of activin A When the cells were cultured with activin A (1 ng/well) at a density of 2 X lo5 cells/200 mm2 for 24 or 48 h, there was no significant difference in FSH secretion between the activin A-treated and the control cells (P > 0.1). A 36% increase was observed after 72 h of incubation with activin A, though it was not significant at P < 0.05 but was significant at P < 0.1 (Fig. la). LH secretion was not changed by activin A throughout the culture period (P > 0.1) (Fig. lb). Thus, there was a tendency of increase in FSH secretion after long-term incubation with activin A. FSH and LH secretion of cells pretreated with activin A FSH and LH secretion into the fresh medium after activin A pretreatment (48, 72 or 96 h) was measured. The cell density in this experiment was 2 X lo5 cells/well. FSH secretion by the cells pretreated with activin A for 48 or 72 h did not exceed that by the control cells. However, when

182 (bl

(0) Control

ti 15-

1

(b)

(a)

1 ~CControl

~Control 0

@ Activln

@Actkin

I

Control

mAct,vm

I 9

4 /’

;

i Cell density (xlOScells/well)

24 Culture

time lhl

Fig. 1. FSH and LH secretion in the presence of activin A. The cells (2 ~lO~/well) were cultured with or without activin A (1 ng/ml/well) for 24-72 h. Amounts of FSH (a), and LH (h) in the medium are expressed as means+SE for three different trials.

the pretreatment period was extended to 96 h, FSH secretion exceeded that of the control cells (Fig. 2~). On the other hand, LH secretion was not significantly affected by activin A (Fig. 2h). Thus, pretreatment was effective for manifestation of the action of activin A and the effect was specific on FSH secretion.

(b)

(a)

@Actkin

rf

I

0

48

72

Pretreatment

96 time ihi

0

48 72 96 Pretreatment time lhl

Fig. 2. FSH and LH secretion by cells pretreated with activin A. The cells (2x105/well) were cultured with or without activin A (1 ng/ml/well) for 48-96 h, and then further incubated in fresh medium without activin A for 3 h. Amounts of FSH ( CI), and LH (6) secreted into the fresh medium during 3-h incubation were assayed. The data are expressed as means k SE for three different trials ( * P < 0.05).

/

4 2 Cell densrty (xlO’cells/well)

Fig. 3. Influence of cell density on activin A action. The cells were cultured at cell densities of 1 x 105. 2 X lo5 or 4X 105/well with or without activin A (1 ng/ml/well) for 48 h. Then. the cells were further incubated in fresh medium without activin A for 3 h. Amounts of FSH in the fresh medium (a). and sums of secreted FSH and intracellular FSH (h) are expressed as means f SE for three different trials ( * P <: 0.05).

Influence of ceil density on activin A action The influence of cell density on the action of activin A was examined using a 48-h culture period, which had been shown to be insufficient for manifesting the effect of activin A at a cell density of 2 X 10’ cells/well. When the cell density was reduced to 1 X lo5 cells/well, FSH secretion by the activin A-pretreated cells was higher than that by the control cells, but not at 2 x lo5 or 4 X lo5 cells/200 mm2 (Fig. 3~). Similarly, total FSH, i.e. the sum of secreted and intracellular FSH, was higher in the activin A-treated cells only at 1 X 10’ cells/200 mm2 (Fig. 3h). Influence of activin A treutment on the number of FSH cells The effect of activin A on the numbers of hormone-containing cells was examined by immunocytochemistry (Table 1). The percentage of FSH-positive cells in the cells treated with activin A for 72 h was higher than that in the control cells by 41%, 37% and 25%, respectively, in three independent experiments (P < 0.01). The percentage of total immunoreactive gonadotropes detected by simultaneous staining with anti-FSH and anti-LH was also 32% higher in the cells treated with activin A (P < O.Ol), but the ratio of LH-positive cells was not affected. The activin A treatment did not affect the ratios of TSH-, PRL- or GH-positive cells. No staining was observed when normal

183

TABLE

Response of activin A-treated cells to GnRH stimulation

1

PERCENTAGES

OF IMMUNOSTAINED

CELLS

The cells were treated with or without activin A for 72 h and then stained by the peroxidase-antiperoxidase (PAP) complex technique. The data are expressed as means+ SE for three different trials. Control Expr. I a Normal rabbit serum Anti-FSH

Activin A-treated

0

0

0

% increased

12.0

41

0 7.8kO.3 0 11.5Iko.4

0 10.7 f 0.1 * 0 15.2k0.2 *

0 37

0 8.3kO.l 9.8kO.2 9.5 *0.5 25.7 f 0.7 26.9rt1.4

10.4 f 0.4 * 9.8 * 1.1 9.5 f 0.7 26.0 + 0.7 26.8k1.4

8.5

Expt. 2

Normal rabbit serum Anti-FSH Anti-FSH + FSH b Anti-FSH + anti-LH

32

Expr. 3

Normal rabbit serum Anti-FSH Anti-LH Anti-TSH Anti-PRL Anti-GH

0

0 25 0 0 1.2 -0.4

’ Means for two different trials. b Anti-serum was preabsorbed with FSH at 4OC for 1 h. * P < 0.01 vs. control.

serum or anti-FSH serum preabsorbed with FSH was used. Thus, the increase in the number of hormone-containing cells by activin A was specific for FSH-producing cells. rabbit

I3 Control

a:c(pcoos) b:d(p>oos)

10~ QActivin 2.4-fold

”

0

10

1.5_lold

0

The response to GnRH stimulation of cells pretreated with activin A for 96 h at a cell density of 2 x lo5 cells/well was examined (Fig. 4). In the control cells, GnRH stimulation during 3 h of incubation increased FSH secretion by 2.4-fold, from 2.36 + 0.23 ng/well to 5.54 k 0.81 ng/well. Activin A pretreatment itself enhanced FSH secretion. FSH secretion by the activin A-pretreated cells was further increased by GnRH stimulation, but the amount did not exceed that by the GnRH-stimulated control cells. In the activin Apretreated cells, GnRH stimulation increased FSH secretion by only 1.5-fold.

10

GnRH (nM)

Fig. 4. Response of activin A-treated cells to GnRH stimulation. The cells (2x105/well) were cultured with or without activin A (1 ng/ml/well) for 96 h and then further incubated in fresh medium without activin A for 3 h either in the presence or absence of GnRH. Amounts of FSH in the fresh medium are expressed as means f SE for three different trials.

Discussion With regard to the mode of action of activin A, we confirmed that long-term activin A treatment was required in order to manifest its effect (Ling et al., 1986a, b; Vale et al., 1986; Kitaoka et al., 1987). Activins have been thought to stimulate biosynthesis and secretion of FSH in individual FSH cells, which results in increased FSH biosynthesis and secretion as a whole (Ling et al., 1986a, b; Vale et al., 1986; Kitaoka et al., 1987). However, there has been no rationale to explain why long-term treatment is required in order for activins to manifest their effects on pituitary cells. Our results showed that the effect of activin A was retained in the cells after removal of activin A from the medium. This mode of action is apparently different from those of secretagogues such as GnRH (Denef and Andries, 1983) or TRH (Shiota et al., 1984), whose effects disappear within a few minutes after their removal. The action of activin A on pituitary cells was not transient but long-lasting, suggesting that activin A acted on pituitary cells in the same way as a growth- or differentiation-promoting factor. This concept is further supported by the finding that activin A increased the number of immunoreactive FSH cells, and that the action of activin A on FSH secretion and synthesis was cell density-dependent. Since cell growth and differentiation are strictly controlled by cell-to-cell interactions in various tissues (Houssaint, 1980; Smith et al.,

1x4

1986), they are not independent of the influence of cell density. Pituitary gonadotropes are heterogeneous (Nakane, 1970) and are thought to be composed of cells in various stages of differentiation (Childs et al., 1981). although the mechanism responsible for regulating cell differentiation and proliferation is far from clear. Begeot et al. (1983) showed that pituitary primordia from fetal rats differentiated into lactotropes upon GnRH stimulation in vitro. In addition, Khar et al. (1978) demonstrated that proliferation could take place in cultured anterior pituitary cells. Since some of the cells in adult rats are thought to be stem cells (Childs et al., 1981; Naor et al., 1982), it is possible that activin A may induce differentiation or/and proliferation in such stem cells. It was demonstrated that activin A treatment increased the number of immunoreactive FSH cells, although the response of the activin A-treated cells to GnRH stimulation was lower than that of normal cells. Thus, immunoreactive FSH cells newly induced after the activin A treatment would be low responders to GnRH stimulation. Monohormonal gonadotropes are known to possess a lower ability to respond to GnRH stimulation (Denef et al., 1978; Lloyd and Childs, 1988), and to be at a lower differentiation stage than multihormonal cells (Naor et al., 1982). The possibility that the increase in the number of immunoreactive FSH cells was due to the conversion of other hormonal cell types into FSH cells is unlikely, because the proportions of immunoreactive LH, TSH, PRL or GH cells were not affected by the activin A treatment. Lloyd and Childs (1988) showed that monohormonal gonadotropes could be driven by GnRH to synthesize other gonadotropins and therefore could be changed to multihormonal gonadotropes. However, since the proportion of immunoreactive gonadotropes (immunoreactive FSH and LH cells) was also increased, the increase in FSH-positive cells was not dependent on the conversion of LH cells to multihormonal cells, but these cells seemed to originate from hormonally negative cells. There may be other alternative explanations of the data or some criticisms in our interpretation. The strict definition of ‘FSH cells’ is difficult because it depends on the sensitivity of the assay

technique employed. Cells with an insufficient amount of FSH might not be expressed as FSH cells by the imnlunostaining methods used in the present study. Therefore, it is possible to assume that some of these cells might produce and store larger amounts of FSH under activin A stimulation. and thus be considered de novo FSH cells. However, it is still difficult to explain the long-term requirement for activin A action based on this assumption. solely in terms of the effect on FSH synthesis in already differentiated FSH cells, because the increase in FSH synthesis would be observed much earlier, within several hours (Apfelbaum and Taleisnik, 1976). It would be preferable that the increase in the amount of FSH did not depend on a quantitative change in the activity of FSH synthesis but rather on a qualitative change in the cells. Thus, it is of great interest which cell type(s) is the target for the action of activin A. Further studies are required, and investigations are now in progress in our laboratory using specific cell type-enriched populations (separatcd, for example, by centrifugal elutriation). In summary, we found that (1) pretreatment of pituitary cells with activin A enhanced their FSH secretion. (2) the expression of this activin A action depended on cell density, and (3) activin A caused an increase in the number of immunoreactive FSH cells. Acknowledgements The authors thank Dr. T. Imamichi, Imamichi Institute for Animal Reproduction, for donation of the animals, the Central Research Laboratory of Ajinomoto Co. for the gift of activin A (EDF), and NIADDK and NIDDK for providing the materials with which to perform radioimmunoassays and immunocytochemistry. We also thank Dr. D.B. Douglas for proofreading the manuscript. References Apfelbaum, M.E. and Taleisnik, S. (1976) J. Endocrinol. 68. 127-136. Begeot. M.. Hemming. F.J.. Martinat, N.. Dubois, M.P. and Dubois. P.M. (1983) Endocrinology 112. 2224-2226. Chari, S.. Daume, E.. Sturm, G.. Vaupel. H. and Schuler. I. (1985) Mol. Cell. Endocrinol. 41, 137-145. Childs, G.V. (1983) Stain Technol. 5X. 2X1--289.

185 Childs, G., Ellison, D., Foster, L. and Ramaley, J.A. (1981) Endocrinology 109, 1683-1692. Childs, G.V., Ellison, D.G., Lorenzen, J.R., Collins, T.J. and Schwartz, N.B. (1982) Endocrinology 111, 13181328. de Jong, F.H., Smith, SD. and van der Molen, H.J. (1979) J. Endocrinol. 80, 91-102. Denef, C. and Andries, M. (1983) Endocrinology 112.813-822. Denef, C., Hautekeete, E. and Dewals, R. (1978) Endocrinology 103. 736-747. Eddie, L.W., Baker, H.W.G.. Higginson, R.E. and Hudson, B. (1979) J. Endocrinol. 81, 49-60. Esch, F.S., Shimasaki, S. Cooksey, K., Mercado, M., Mason, A.J., Ying, S.-Y., Ueno, N. and Ling, N. (1987) Mol. Endocrinol. 1, 388-396. Eto, Y., Tsuji, T., Takezawa, M., Takano, S., Yokogawa, Y. and Shibai, H. (1987) Biochem. Biophys. Res. Commun. 142, 1095-1103. Forage, R.G., Ring, J.M., Brown, R.W., McInerney, B.V., Cobon, G.S., Gregson, R.P., Robertson, D.M., Morgan, F.J., Hearn, M.T.W., Findlay, J.K., Wettunhall, R.E.H.. Burger, H.G. and de Jong, D.M. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 3091-3095. Franchimont, P., Lecomte-Yerna, M.-J., Henderson, K., Verhoeven, G., Hazee-Hagelstein, M.-T., Jaspar, J.-M., Charlet-Renard, C. and Demoulin, A. (1983) in Role of Peptides and Proteins in Control of Reproduction (McCann, SM. and Dhindsa, D.S., eds.), pp. 237-255, Elsevier, New York. Houssaint. E. (1980) Cell Differ. 9, 269-279. Hsueh, A.J.W., Dahl, K.D., Vaughan, J., Tucker, E., Rivier, J., Bardin, C.W. and Vale, W. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 5082-5086. Khar, A., Debeljuk, L. and Jutisz, M. (1978) Proc. Sot. Exp. Biol. Med. 158, 471-474. Kitaoka, M., Yamashita, N., Eto, Y., Shibai, H. and Ogata, E. (1987) Biochem. Biophys. Res. Commun. 146. 1382-1385. Ling, N., Ying, S.-Y., Ueno, N., Shimasaki, S., Esch, F., Hotta, M. and Guillemin, R. (1986a) Nature 321, 779-782.

Ling, N., Ying, S.-Y., Ueno, N., Shimasaki, S., Esch, F., Hotta, M. and Guillemin, R. (1986b) B&hem. Biophys. Res. Commun. 138, 1129-1137. Lloyd, J.M. and Childs. G.V. (1988) Endocrinology 122, 1282-1290. Mason, A.J., Hayflick, J.S., Ling, N., Esch, F., Ueno, N., Ying, S.-Y., Guillemin, R., Niall, H. and Seeburg, P.H. (1985) Nature 318, 659-663. Mason, A.J., Niall, H.P. and Seeburg, P.H. (1986) B&hem. Biophys. Res. Commun. 135, 957-964. Mayo, K.E., Cerelli, G.M., Spiess, J., Rivier, J., Rosenfeid, M.G., Evans, R.M. and Vale, W. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 5849-5853. Moriarty, G.C. (1975) Endocrinology 97, 1215-1225. Murata, M., Eto, Y., Shibai, M. and Muramatsu, M. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 2434-2438. Nakane, P.K. (1970) J. Histochem. Cytochem. 18, 9-20. Naor, Z.. Childs, G.V., Leifer, A.M., Clayton, R.N., Amsterdam, A. and Catt, K.J. (1982) Mol. Cell. Endocrinol. 25, 85-97. Schwall, R.H.. Nikolics, K., Szonyi, E., Gorman, C. and Mason, A.J. (1988) Mol. Endocrinol. 2, 1237-1242. Scott, R.S., Burger, H.G. and Quigg, H. (1980) Endocrinology 107, 153661542. Shiota, K., Yoshida, K., Masaki, T., Kawase, M.. Nakayama, R. and Sudo, K. (1984) Endocrinol. Jpn. 31, 165-175. Smith, B.T., Floras, J. and Post, M. (1986) in Cellular Endocrinology; Hormonal Control of Embryonic Cellular Differentiation (Serrero, G. and Hayashi, J., eds.), pp. 141-146, Alan R. Liss, New York. Steinberger, A. and Steinberger, E. (1976) Endocrinology 99, 918-921. Stewart, A.G., Milborrow, H.M., Ring, J.M., Crowther, C.E. and Forage, R.G. (1986) FEBS Lett. 206, 329-334. Vale, W.. Rivier, J., Vaughan, J., McClintock, R., Carrigan, A., Woo, W., Karr, D. and Spiess, J. (1986) Nature 321, 776-779.