Hormone Synthesis and Responsiveness of Spontaneous Granulosa Cell Tumors in (SWR × SWXJ-9) F1 Mice

GYNECOLOGIC ONCOLOGY ARTICLE NO.

65, 143–148 (1997)

GO974635

Hormone Synthesis and Responsiveness of Spontaneous Granulosa Cell Tumors in (SWR 1 SWXJ-9) F1 Mice PETER M. GOCZE,*,† WESLEY G. BEAMER,‡ FRANK H.

DE

JONG,§

AND

DALE A. FREEMAN*,†

*Department of Internal Medicine and Department of Biochemistry and Molecular Biology, University of Oklahoma Health Sciences Center, and †Department of Veterans Affairs Medical Center, Oklahoma City, Oklahoma 73104; ‡Jackson Laboratory, Bar Harbor, Maine 04609; and §Department of Endocrinology and Reproduction, Erasmus University, Rotterdam, The Netherlands Received September 16, 1996

Granulosa cell tumors spontaneously occur in approximately 10–25% of female (SWR 1 SWXJ-9) F1 mice. The present studies were designed to test whether tumor-bearing mice produce a distinct hormonal profile by which they could be identified and determine whether cultured tumor cells are responsive to hormones and growth factors that regulate normal granulosa cells. Samples of female mouse blood taken from age 3 to 10 weeks allowed estimation of serum FSH, 17b-estradiol, and inhibin levels for normal mice and for mice destined to develop tumors. These studies indicated that FSH and 17b-estradiol values differed between normal and tumor-bearing animals, but overlapped sufficiently that such values could not accurately predict the tumor-bearing state. Inhibin concentrations did differentiate normal from tumorbearing animals in all cases. Increased levels of inhibin were observed coincident in time with visibly detectable tumors within the ovaries. Compared to inhibin synthesis in vivo, hormonal responsiveness in vitro was much more variable. Steroidogenesis was stimulated in all tumors by dibutyryl-cAMP and low-density lipoprotein (LDL). Some, but not all, tumors responded to IGF1 , EGF, FSH, and hCG. In about one-half of the tumors tested, FSH could induce hCG or dibutyryl-cAMP responsiveness. IGF1 pretreatment consistently increased the responsiveness of tumor cells stimulated by dibutyryl-cAMP and LDL. Production of inhibin by isolated tumor cells was detectable and decreased by EGF or dibutyrylcAMP treatments. We conclude that granulosa tumor cell secretion of inhibin may be under different control than secretion from normal granulosa cells and acts as an excellent marker for these tumors. q 1997 Academic Press

INTRODUCTION

Granulosa cell tumors make up about 2% of all ovarian tumors and are one of the most common hormone-producing tumors [1]. Although these tumors are comparatively common functional tumors, they are still rare in most clinical settings. For this reason, relatively little is known about the endocrine function of these tumors. These tumors usually appear in postmenopausal women and often cause uterine bleeding. Many patients show signs of estrogen excess such

as endometrial hyperplasia and in several patients estrogen levels are increased and in at least one case the tumor could synthesize estradiol de novo [2]. The peptide hormone, inhibin, has been found to be elevated in some patients with granulosa cell tumors [3, 4] and may act as a tumor marker for this type of tumor. One well-characterized granulosa cell tumor model exists in SWR inbred mice [5]. Recombinant inbred strains and hybrid F1 females derived from crosses of SWR with SJL mouse strains spontaneously develop tumors in 1–15% of females. Mice bearing these tumors show reduced levels of pituitary and serum LH and FSH, as well as reduced serum progesterone, dihydrotestosterone, and testosterone values [6]. Tumors do not take up 125I-hCG and do not have histologically identifiable 3b-hydroxysteroid dehydrogenase activity [6], but have a marked increase in numbers of receptors for EGF [7]. Tumor progression has been studied carefully [8]. By 4 weeks of age, preneoplastic follicles are visible to the naked eye, while by 5 to 6 weeks primary tumors are identified [8]. Far more is known about hormone responsiveness and synthesis by normal granulosa cells than tumors composed of these cells. Normal granulosa cells respond to FSH with induction of LH responsiveness [9], with induction of inhibin subunit synthesis [10] and stimulation of 17bestradiol synthesis [11]. Various growth factors, cytokines, and other peptides can modify granulosa cell function [12, 13, 14]. In the present studies mice genetically predisposed to development of pubertal-onset granulosa cell tumors were sequentially bled from the 3rd to the 10th week of life and analyzed for serum levels of FSH, estradiol, and inhibin. In addition, primary cultures of granulosa tumor cells were established and tested for steroid hormone responsiveness and for inhibin synthesis. The data showed that primary tumor cells shared some characteristics with normal granulosa cells, but varied markedly from tumor to tumor in hormone and growth factor responsiveness.

143

AID

Gyn 4635

/

6d1a$$$341

03-07-97 00:19:08

0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

goa

AP: GYN

144

GOCZE ET AL.

MATERIALS AND METHODS

Materials Hormones hCG (Cr 127) and oFSH (FSH-19-S1AFP, lot No. AFP4117A) were generous gifts of the National Hormone and Pituitary Distribution Program. EGF and IGF1 were purchased from Calbiochem (San Diego, CA). Crystalline steroids were obtained from Steraloids, Inc. (Wilton, NH). [1,2,6,7-3H]Progesterone, [1,2-3H]-20a-hydroxy-4-pregnen-3-one, and [7-3H]pregnenolone were purchased from DuPont – New England Nuclear (Boston, MA). Antibody to progesterone was from Wein Laboratories (Succasunna, NJ); antibody to 20a-dihydroprogesterone was from Accurate Chemicals (Westbury, NY); antibody to pregnenolone was from IgG Corp. (Nashville, TN); and antibody to inhibit was a gift from Dr. D. M. de Krefser of the Department of Anatomy, Monesh University (Clayton, Australia). Waymouth’s MB 751/1 medium, horse serum, and fetal bovine were from Gibco BRL/Life Technologies, Inc. (Baltimore, MD). Methods Animals. Mating pairs of SWR and SWXJ-9 mice were obtained from the research colony (WGB) at The Jackson Laboratory (Bar Harbor, ME). Mating pairs were housed together, fed Purina formula 5008 rodent chow, and maintained with 12 hr dark/light cycles in the Animal Research Facility of the Department of Veterans Affairs Medical Center in Oklahoma City. Mice were weaned at 21–24 days and housed with littermates of the same gender. Males were discarded and females were marked so that each pup in each litter could be consistently identified. Blood was taken from the tail vein under methoxyflurane anesthesia each week (from Week 3 until Week 9 or 10) at the same time of day. Altogether, 48 animals had blood drawn; blood from 5 animals with tumors as well as randomly chosen controls was analyzed for hormone content. At 9 or 10 weeks or, in one case, after an anesthetic death, the animals were necropsied and the ovaries were examined for tumors. Tumors measuring approximately 6–10 mm in diameter were excised using aseptic technique and placed into culture as described below. Granulosa tumor cell culture. Tumors were freed of connective tissue and any normal ovary. Tumor tissue was minced with scissors and scalpels and then digested as described previously [7]. Tumor cells were placed in culture in 24-well dishes containing 1 ml of Waymouth’s MB 752/ 1, pH 7.4, medium supplemented with 20 mM Hepes buffer, 1.12 g/liter NaHCO3 , 10% horse serum, and 5% fetal bovine serum. In some experiments after cells had attached, this medium was changed to defined medium Opti-MEM (Gibco BRL/Life Technologies, Baltimore, MD). Hormones or

AID

Gyn 4635

/

6d1a$$$342

03-07-97 00:19:08

growth factors were added from 100-fold concentrates in phosphate buffered saline pH 7.4, 0.1 mg/mL BSA. Steroid hormone assays. Radioimmunoassays for progesterone, 20a-dihydroprogesterone, and pregnenolone were as described previously [15]. For most experiments pregnenolone production was measured in cells blocked with 1006 M cyanoketone and 1005 M spironolactone. Pregnenolone was measured after we determined that the major steroid product of the tumors was not progesterone, 20a-dihydroprogesterone, 17-hydroxyprogesterone, androstenedione, testosterone, or estradiol. The assays for 17-hydroxyprogesterone, androstenedione, testosterone, and 17b-estradiol were kindly performed by Dr. M. Zavey and the Department of Obstetrics and Gynecology Core Laboratories, University of Oklahoma Health Sciences Center. Estradiol values of serum from the mouse bleedings was assayed using kits provided by Diagnostic Products Corp. (Los Angeles, CA). Peptide hormone assays. Assays of FSH and inhibin were performed as described previously [16]. Standards were NIADDK-rat-FSH-RP-3 and a bovine follicular fluid preparation [16]. The interassay variation of the inhibin assay was 12%, that for the estradiol assay was 5%, and that for the FSH assay was 5.1%. Statistical evaluation. The data were compared using Student’s t test. RESULTS

Inhibin levels increase in mice developing granulosa cell tumors. The data of Fig. 1 show values for FSH, estradiol, and inhibin in the serum of mice that either developed tumors or were tumor-free at the time of sacrifice. These data were obtained by sampling all female mice beginning on Week 3 or 4 of life and continuing until Week 9 or 10. One tumorbearing animal died during anesthesia at Week 8 so data for this animal ended at that point. The control data for each tumor-bearing animal are matched for litter. In each tumorbearing animal, there was a very significant rise in inhibin (Fig. 1A). Using this as an indicator of tumor occurrence, hormone data can be analyzed as before and after tumor occurrence. In Fig. 1B the average FSH value in control animals was 5.9 { 3.4 ng/mL (mean { SD) and in the tumorbearing animals it was 5.6 { 3.4 ng/mL before and 1.83 { 2.8 ng/mL after tumor occurrence. These mean values of control and preoccurrence tumor-bearing animals are the same and both are statistically different from tumor-bearing animals using Student’s t test at P õ 0.01. Overlap between the groups precludes using FSH values to predict the presence of tumor. In Fig. 1C estradiol values for control animals averaged 32.5 { 23.2, pmol/L; they averaged 32.3 { 23.5 pmol/L for animals before tumor occurrence, and in tumorbearing animals they averaged 89.5 { 62.4 pmol/L. The mean values of controls and animals pretumor occurrence

goa

AP: GYN

145

INHERITED GRANULOSA CELL TUMORS

were the same, while each was statistically different than tumor-bearing animals at P õ 0.0004; however, data from the two groups overlap to a great extent. Figure 1A shows that the inhibin levels in all tumor-bearing animals differed markedly from controls. The average inhibin concentration of control animals was 44.2 { 14.4 U/mL compared with

TABLE 1 Effects of FSH, Dibutyryl-cAMP, EGF, and LDL on Granulosa Cell Tumor Steroidogenesis a Pretreatment

Treatment

Pregnenolone (ng/48 h)

None None None None None

None FSH, 50 ng/ml EGF, 10 ng/ml 1 mM Bu2cAMP 50 mg LDL

141 187 205 587 281

{ { { { {

22 3a 17 a 75a 15a

None None FSH, 5 ng/ml FSH, 10 ng/ml FSH, 25 ng/ml FSH, 50 ng/ml

None hCG, hCG, hCG, hCG, hCG,

249 304 315 404 497 477

{ { { { { {

43 18b 19 b,c 75b,c 71b,c 76b,c

80 80 80 80 80

ng/ml ng/ml ng/ml ng/ml ng/ml

a

Granulosa cell tumors were grown for 72 hr in growth medium alone or containing the indicated pretreatments. After 72 hr the medium was replaced with medium containing the indicated additions as well as cyanoketone (1006 M) and spironolactone (1005 M). After 48 additional hours, the medium was removed and the content of pregnenolone was determined by radioimmunoassay. All data are the mean { SEM of triplicate determinations. b All data are statistically greater than unstimulated control (no pretreatment, no treatment) at P õ 0.05 or greater. c All data are statistically greater than cells stimulated with hCG, but not FSH pretreated.

FIG. 1. Plasma levels of inhibin (A), FSH (B), and 17b-estradiol (C) in normal and tumor-bearing mice. All female pups of individual litters were housed together and marked so that they could be definitely identified. From Weeks 3 to 10, mice were placed under anesthesia and blood was withdrawn from a tail vein incision. Tumor-bearing animals are indicated by l, m, ., j, and l, while normal controls from the same litters are indicated by s, n, ,, h, and L. Note, logarithmic scale in A.

AID

Gyn 4635

/

6d1a$$$342

03-07-97 00:19:08

31.8 { 11.9 U/mL before tumor occurrence and 1864.3 { 1755 U/mL for tumor-bearing animals. The difference between control values and pretumor values was not different but each of these mean values was statistically different than the values for tumor-bearing animals at P õ 0.0002. Inhibin levels in tumor-bearing animals were correlated to the FSH and estradiol data from the same animals. Inhibin values correlated negatively with FSH values (r Å 0.639, n Å 14 [n Å number of samples correlated] and P õ 0.01) and correlated positively with estradiol values (r Å 0.474, n Å 27, and P õ 0.01). Steroidogenic response of cultured granulosa tumor cells to hormones, growth factors, dibutyryl-cAMP, and LDL. Normal granulosa cells modify steroid hormone synthesis in response to a variety of hormones and growth factors. The data of Tables 1 and 2 show steroid responses to a variety of stimuli. Preliminary experiments with three tumors revealed that the cultured granulosa tumor cells produced only trace quantities of progesterone, 20a-dihydroprogesterone, 17-hydroxyprogesterone, androstenedione, testosterone, and 17b-estradiol. To quantitate effects on steroidogenesis without identifying a steroid end product, the 3b-hydroxysteroid dehydrogenase, D5 – D4-isomerase was blocked with cyanoketone and the 17-hydroxylase was blocked with spironolactone, thus preventing metabolism of pregnenolone to other steroids. The data of Table 1 show the responses of cells from two tumors. Data from one tumor demonstrate

goa

AP: GYN

146

GOCZE ET AL.

treatment responses to a variety of stimuli, while data from the second tumor show the effect of FSH preincubation on subsequent hCG responsiveness. In the first tumor cell population, treatment of cells with FSH, EGF, dibutyryl-cAMP, and LDL resulted in significant augmentation of pregnenolone synthesis. In this and other experiments with the mouse granulosa tumor cells, dibutyryl-cAMP elicited a greater steroid response than FSH, hCG, or hCG after FSH pretreatment. Tumor cells from the second tumor demonstrated that FSH pretreatment could induce responsiveness to hCG. Cells stimulated by hCG but not pretreated with FSH produced 304 { 18 ng pregnenolone per well in 48 hr, a value 1.2fold greater than unstimulated cells. Cells pretreated with a maximally effective dose of FSH (25 ng/mL) synthesized 497 { 7 ng pregnenolone in 48 hr, a value 2-fold greater than hCG-stimulated, but not FSH pretreated cells. FSH pretreatment at doses 10 ng/mL or higher all significantly augmented subsequent responsiveness to hCG. A striking finding of the steroid production experiments was how variable hormone and growth factor responses were. These data are summarized in Table 2. All tumors responded to dibutyryl-cAMP or to LDL. Sixty-three percent of tumors responded to FSH, 60% responded to EGF, 50% responded to IGF1 , and 30% responded to hCG. The lower responsiveness to hCG may indicate an FSH requirement TABLE 2 Effect of Various Treatment on Granulosa Tumor Cell Steroid Responsea Parameter

Positive/total tested

%

10/10 5/5 5/8 6/10 5/10 3/10

100 100 63 60 50 30

4/9 2/4

44 50

3/3

100

Steroidogenesis stimulated by b 1 mM Bu2cAMP LDL, 50 mg/ml FSH, 50 ng/ml EGF, 10 ng/ml IGF1 , 50 ng/ml hCG, 80 ng/ml FSH induction of (c) hCG responsiveness Bu2cAMP responsiveness IGF1 induction of (d) LDL receptor activity

a,b Individual granulosa cell tumors were cultured in 24-well culture dishes containing growth medium as well as cyanoketone and spironolactone. The hormones, cAMP analogues, growth factors, or lipoproteins were added to the incubations above and allowed to incubate for 48 hr at 377C. At this time medium was removed and the content of pregnenolone was determined by radioimmunoassay. Triplicate wells were tested for each tumor. A stimulatory factor caused more pregnenolone to be produced than the control incubations as determined by t test of the mean control and experimental incubations. Test wells were incubated for 72 hr with 50 mg/ ml FSH (c), with 50 ng/ml IGF1 (d), or with no additions. At that time the medium was removed and replaced with medium containing 80 ng/ml hCG (c) or 1 mM Bu2cAMP (c, d). After 48 hr medium was assayed for pregnenolone content.

AID

Gyn 4635

/

6d1a$$$342

03-07-97 00:19:08

TABLE 3 Inhibin Synthesis by Granulosa Cell Tumors in Culture a Treatment

Inhibin (IU/ml) Tumor 1

Control FSH, 50 mg/ml IGF1 , 50 ng/ml mEGF, 10 ng/ml

52.0 39.5 35.0 7.1

{ { { {

41.8 26.7 20.0 1.7*

11.5 7.5 2.8 6.2

{ { { {

4.7 2.4 0.3* 0.9*

Tumor 2 Control IGF1 , 50 ng/ml mEGF, 10 ng/ml Bu2cAMP, 0.5 mM

a Granulosa tumor cells were prepared and cultured in 24-well culture dishes as described under Materials and Methods. All additions were made into defined tissue culture medium Opti-MEM for 72 hr. At 72 hr the medium was removed and saved for radioimmunoassay of inhibin. The samples in Experiments 1 and 2 are from different tumors. The data are the mean { SD of triplicate wells. *P õ 0.05 compared with control data.

for hCG responsiveness. This seemed doubtful, however, since FSH was only able to induce hCG responsiveness in 44% of the tumors. Since FSH enhanced dibutyryl-cAMP responsiveness in half of the tumor cell populations tested, it seemed likely that FSH was also modifying some postLH receptor function. It has been shown that IGF1 stimulates HDL binding and HDL-supported progesterone synthesis in swine granulosa cells [17]. In all three cultures tested here, IGF1 pretreatment increased the amount of steroid synthesized in response to dibutyryl-cAMP and LDL. Cultures of granulosa tumor cells secrete inhibin: secretion of inhibin is decreased by EGF or Dibutyryl-cAMP treatment. The data of Table 3 show that each of the two tumor cell populations secreted detectable quantities of inhibin. Treatment of the tumor cells with EGF markedly decreased inhibin release in both experiments. The response of tumor cell inhibin secretion to dibutyryl-cAMP was only measured in one tumor and in this tumor the cAMP analogue significantly inhibited tumor inhibin secretion. Treatment of the tumor cells with IGF1 was without significant effect. Likewise, in the one tumor treated with FSH, no significant differences in inhibin secretion occurred. DISCUSSION

The data reported herein on spontaneous granulosa cell tumor-bearing hosts suggest that the tumors were associated with a characteristic endocrine syndrome. Tumor-bearing mice had very high inhibin concentrations, somewhat increased 17b-estradiol concentrations, and suppressed FSH values. Studies on dispersed granulosa tumor cells in vitro

goa

AP: GYN

147

INHERITED GRANULOSA CELL TUMORS

showed that responses in vitro were heterogenous. Individual tumors varied in their steroid responses to FSH, IGF1 , EGF, and hCG and only consistently responded to dibutyrylcAMP and LDL. Inhibin secretion in vitro showed that the tumor cells synthesized inhibin and that this secretion could be modified by EGF and dibutyryl-cAMP. Inhibin can act as a tumor marker for granulosa cell tumors [18, 19]. From the available literature, it would seem likely that most granulosa cell tumors produce measurable inhibin levels [18]. However, in some patients the inhibin levels must be provoked by FSH [19]. In the mice, inhibin concentrations were always elevated in animals bearing a tumor. Inhibin secretion was demonstrable in vitro as well. In both tumors tested, inhibin concentrations decreased in response to EGF. Granulosa cell tumors are classified as feminizing mesenchymomas [1] and often present with uterine bleeding in a postmenopausal patient. Several patients with such tumors have elevated estrogen levels in blood [2, 18–22] and at least one tumor produced estradiol [2]. The granulosa tumors in mice were also associated with 17b-estradiol levels about twofold greater than control. 17b-Estradiol levels positively correlated with inhibin levels. It would be tempting to attribute the elevated estradiol levels to tumor production directly; however, in vitro data did not demonstrate 17b-estradiol synthesis. Three tumors investigated for specific steroid products revealed no D4 steroid products. It is possible that steroid precursors synthesized by the tumor were converted by peripheral aromatase to estradiol or that different tumors had different steroid end products. Alternatively, the tumors may stimulate aromatase activity in the adjoining normal ovary, leading to estrogen synthesis from circulating androgen precursors [23, 24]. Normal granulosa cells are responsive to hormones (FSH, hCG) [9–11], to growth factors (IGF, EGF) [12–14], and to lipoproteins (LDL, HDL) [17, 22]. In addition, they respond to a variety of other cytokines and peptide stimuli. In the present studies, individual tumors responded to FSH, hCG, IGF1 , and EGF, all with increased steroid production. In some tumors FSH increased hCG responsiveness, but also increased responsiveness to dibutyryl-cAMP. This later finding may indicate an effect to increase the activity of the post-LH receptor pathway. When the tumors did respond to hormones, growth factors, or lipoproteins, it was in the way expected. Loss of hormone receptors and responsiveness is a common tumor characteristic [25] and may have been responsible for the unresponsiveness of tumors in vitro. No characteristic of tumor or host seemed to predict responsiveness to stimulation except possibly the age of the animal and tumor. It seemed that smaller tumors from younger animals were more likely to be responsive to FSH. Previous studies using these mice demonstrated receptors for EGF in the tumors [7]. In those studies EGF binding to tumor cell membranes ranged markedly between individual

AID

Gyn 4635

/

6d1a$$$342

03-07-97 00:19:08

tumors. The variability of steroid responsiveness to EGF may reflect this variable receptor complement. At least in the two tumors that were investigated in vitro, EGF markedly suppressed inhibin secretion. These data are consistent with data from normal granulosa cells where EGF also exerts a negative regulatory role [9]. Apparently, the native EGF levels are too low to effectively suppress inhibin output by tumor cells. The most surprising finding of the present studies was the heterogeneity of hormonal responsiveness in tumors of a common genetic background. Unlike the human tumors that occur at different times, the mouse tumors should have a constant genetic predisposition. One possible explanation is that the mutant genes leading to tumor growth affect cells early enough in development such that the cells are able to follow one of several developmental paths. From the data that were available, tumor inhibin responses seem to be more homogenous. ACKNOWLEDGMENTS The authors thank Apryl Webb and Marianna Timmerman for excellent technical support. This work was supported by the Department of Veterans Affairs, and CA62434 and ACS-BE170 awarded to W. G. Beamer.

REFERENCES 1. Young RH, Scully RE: Endocrine tumors of the ovary. Curr Top Pathol 85:113–164, 1992 2. Madden JD, MacDonald PC: Origin of estrogen in isosexual precocious pseudopuberty due to a granulosa-theca cell tumor. Obstet Gynecol 51:210–213, (1978) 3. Lappohn RE, Burger HG, Bouma J, et al: Inhibin as a marker for granulosa cell tumors. N Engl J Med 321:790–793, 1989 4. Witt BR, Wolf GC, Wainwright CJ, Thorneycroft IH: Endocrine function of granulosa cell tumors in vivo. Gynecol Obstet Invest 33:59– 64, 1992 5. Beamer WG, Hoppe PC, Whitten WK: Spontaneous malignant granulousa tumors in ovaries of young SWR mice. Cancer Res 45:5575– 5581, 1985 6. Beamer WG: Gonadotropin, steroid and thyroid hormone milieu of young SWR mice bearing spontaneous granulosa cell tumors. J Natl Cancer Inst 77:1117–1123, 1986 7. Tennent BJ, Beamer WG, Shultz LD, Adamson ED: Epidermal growth factor receptors in spontaneous ovarian granulosa cell tumors of SWRderived mice. Int J Cancer 44:477–482, 1989 8. Tennent BJ, Shultz KL, Sundberg JP, Beamer WG: Ovarian granulosa cell tumorigenesis in SWR-derived F1 hybrid mice: Preneoplastic follicular abnormality and malignant disease progression. Am J Obstet Gynecol 163:625–634, 1990 9. Hsueh AJ, Adashi EY, Jones PB, Welsh TH Jr: Hormonal regulation of the differentiation of cultured ovarian granulosa cells. Endocr Rev 5:76–127, 1984 10. LaPolt PS, Piquette GN, Soto D, Sincich C, Hsueh AJ: Regulation of inhibin subunit messenger ribonucleic acid levels by gonadotropins, growth factors, and gonadotropin releasing hormone in cultured rat granulosa cells. Endocrinology 127:823–831, 1990 11. Dorrington JH, Moon YS, Armstrong DT: Estradiol-17b biosynthesis

goa

AP: GYN

148

GOCZE ET AL.

in cultured granulosa cells from hypophysectomized immature rats: Stimulation by follicle-stimulating hormone. Endocrinology 97:1328– 1331, 1975 12. Adashi EY: Cytokine-mediated regulation of ovarian function: encounters of a third kind. Endocrinology 124:2043–2045, 1989 13. Adashi EY: Intraovarian peptides: Stimulators and inhibitors of follicular growth and differentiation. Endocrinol Metab Clin North Am 21:1– 17, 1992 14. Adashi EY: Intraovarian regulation: The proposed role of insulin-like growth factors. Ann NY Acad Sci 687:10–12, 1993 15. Freeman DA: Constitutive steroidogenesis in the R2C Leydig tumor cell line is maintained by the adenosine 3*,5*-cyclic monophosphateindependent production of a cycloheximide-sensitive factor that enhances mitochondrial pregnenolone biosynthesis. Endocrinology 120:124–132, 1987 16. Grootenhuis AJ, Van Sluijs FJ, Klaij IA, et al: Inhibin, gonadotrophins and sex steroids in dogs with Sertoli cell tumours. J Endocrinol 127:235–242, 1990 17. Veldhuis JD, Gwynne JT: Insulin-like growth factor type 1 (somatomedin-C) stimulates high density lipoprotein (HDL) metabolism and HDL-supported progesterone biosynthesis by swine granulosa cells in vitro. Endocrinology 124:3069–3076, 1989

AID

Gyn 4635

/

6d1a$$$343

03-07-97 00:19:08

18. Wentz C, McCranie M: Circulating hormone levels in a case of granulosa cell tumor. Fertil Steril 27:167–170, 1976 19. Adashi E, Rosenwaks Z, Lei P, Jones G, Migeon C: Endocrine features of an adrenal like tumor of the ovary. J Clin Endocrinol Metab 48:241– 245, 1979 20. Leiberman E, Glezerman M, Karny N, Levy J: Steroid receptors in a granulosa cell tumor (GCT) in an infant with precocious puberty. Med Pediatr Oncol 13:370–374, 1985 21. Friedman C, Lams JB, Schwizer FW, Kim MH: Hormonal screening of hyperestrogenemic elderly obese females for granulosa cell tumor. J Reprod Med 26:268–271, 1981 22. Tureck RW, Wilburn AB, Gwynne JT, Paavola LG, Strauss JF III: The role of lipoproteins in steroidogenesis by human luteinized granulosa cells in culture. J Steroid Biochem 19:1033–1038, 1983 23. Aiman J, Nalick RH, Jacobs A, et al: The origin of androgen and estrogen in a virilized postmenopausal woman with bilateral benign cystic teratomas. Obstet Gynecol 49:695–704, 1977 24. MacDonald PC, Grodin JM, Edman CD, Vellios F, Siiteri PK: Origin of estrogen in a postmenopausal woman with a nonendocrine tumor of the ovary and endometrial hyperplasia. Obstet Gynecol 47:644–650, 1976 25. Freeman DA: Steroid hormone-producing tumors in man. Endocrinol Rev 7:204–220, 1986

goa

AP: GYN