GONADOTROPIN BINDING SITES IN HUMAN POSTMENOPAUSAL OVARIES*

Transcription

1 FERTILITY AND STERILITY Copyright 1976 The American Fertility Society Vol. 27, No.7, July 1976 Printed in U.S.A. GONADOTROPIN BINDING SITES IN HUMAN POSTMENOPAUSAL OVARIES* JOHN J. PELUSO, PH.D.,t RICHARD W. STEGER, PH.D.,t S. JASZCZAK, M.D.,t AND E. S. E. HAFEZ, PH.D. Reproductive Physiology Laboratories, C. S. Matt Center for Human Growth and Development, Wayne State University School of Medicine, Detroit, Michigan Glucose-6-phosphate dehydrogenase (G6PD) activity and gonadotropin binding sites were localized within seven postmenopausal ovaries. G6PD was localized in the cells of cortical stroma and hilus using a histochemical technique for the reduction of the tetrazolium salt, Nitro-Blue tetrazolium. Gonadotropin binding sites were localized by autoradiography following incubation of ovarian sections with 125 /-labeled gonadotropins. The binding sites for both labeled luteinizing hormone ( hLH) and 125 /-labeled folliclestimulating hormone ( hFSH) were identified in the cortical stroma and hilus cells. Since these cells contain G6PD and other enzymes necessary for steroidogenesis and also have the capacity to bind both hlh and hfsh, steroidogenesis in postmenopausal ovaries appears to be controlled by circulatting gonadotropins. Blood vessels within postmenopausal ovaries also bound both gonadotropins, but hFSH binding was often more intense than hLH binding. The endocrine profile of the postmenopausal women is characterized by elevated gonadotropin levels and reduced serum estrogen levels. 1 2 Although there is considerable variation among women, ovariectomy often results in a further reduction of serum estrogen levels, indicating that the postmenopausal ovary still has limited steroidogenic activity. 2 Histochemical studies have shown that the cortical stroma of the postmenopausal ovary possesses several of the enzymes necessary for steroid synthesis. 3 4 Judd and co-workers 5 have shown that the postmenopausal ovary actively synthesizes predominately androgens but also produces limited amounts of estrogens. Accepted March 5, *Presented at the Thirty-Second Annual Meeting of The American Fertility Society, April 7 to 9, 1976, Las Vegas, Nev. tford Foundation Postdoctoral Fellow. 789 In addition, there is a high correlation between serum estrogen levels and ovarian glucose-6-phosphate dehydrogenase activity. 4 Although luteinizing hormone and follicle-stimulating hormone levels are elevated in the postmenopausal woman, the ability of these gonadotropins to regulate ovarian function is still in question. Therefore, it is of interest (1) to determine whether the postmenopausal ovary can bind 125 I-labeled gonadotropins, (2) to identify the ovarian structures to which the gonadotropins bind, and (3) to correlate the gonadotropin binding sites with steroid-synthesizing cells as indicated by high glucose-6-phosphate dehydrogenase activity. MATERIALS AND METHODS Samples of ovarian tissue were obtained from patients undergoing hysterectomy and bilateral salpingo-oophorectomy. Immediately after surgery, each

2 790 PELUSO ET AL. July 1976 ovarian sample was divided into two pieces. One piece was fixed in Bouin's solution for histologic examination. The other piece was quick-frozen in a Dry Ice-ethanol bath and stored at - 20o C. The Bouin's-fixed tissue was embedded in paraffin, sectioned at 10 JLm, and stained with hematoxylin and eosin or Masson's trichrome. The quick-frozen tissue was sectioned at 10 JLm in a cryostat, placed on glass slides, and stored at 4 C in a moist chamber until a sufficient number of sections were cut. Alternate sections were then used for determining the localization of gonadotropin binding sites, glucose-6-phosphate dehydrogenase (G6PD) activity, or for hematoxylin-eosin staining. Localization of Gonadotropin Binding Sites. Purified human FSH (hfsh) and LH (hlh) (Calbiochem, La Jolla, Calif.) were used in this experiment. The hlh had a biologic potency of 2000 IU/mg, as judged by the ovarian ascorbic acid depletion bioassay, and 4 IU of FSH activity in the human chorionic gonadotropin augmentation bioassay. 125 I-hLH did not bind to an FSH antiserum which had been preabsorbed with hlh. The human hfsh preparation had a biologic potency of 3500 IU/mg when tested in the human chorionic gonadotropin augmentation bioassay. The LH contamination of this FSH preparation was less than 5% when tested in both the ovarian ascorbic acid depletion bioassay and against preabsorbed LH antisera. Both hlh and hfsh were iodinated with 125 I, using the chloramine T method.6 After iodination, the reaction mixture was passed through a Bio-Gel P-60 column (Bio-Rad, Richmond, Calif.) in order to separate the 125 I-labeled gonadotropin from the free 125!. Approximate values for specific activity were 26 and 50 JLCil JLg of protein for hlh and hfsh, respectively. In order to identify the cellular sites to which 125 I-labeled gonadotropins bind, sections were incubated for 1 hour at 37 C with either 125 I-hLH (2.5 x 105 cpm/50 JLD or 125 I-hFSH (2.3 x 10 5 cpm/ 50 JLD. After incubation the sections were washed for 10 minutes in cold phosphate-buffered saline (ph 7.6), rinsed in distilled water for 5 minutes, and airdried. The sections were then dipped in Kodak NTB-3 nuclear track emulsion, exposed for 15 days at 4 C, and developed. 7 Selected sections were incubated with 125 I-hLH or 125 I-hFSH in the presence of a 100-fold excess of the respective unlabeled gonadotropin in order to determine binding specificity. Localization of G6PD Activity. G6PD activity was localized in ovarian sections by using a histochemical technique involving the reduction of the tetrazolium salt, Nitro-Blue tetrazolium. The sections were incubated for 30 to 40 minutes at 37" C with the incubation media to identify the cells that possessed G6PD activity. 8 Age of patient yr TABLE 1. Clinical Data on Seven Postmenopausal Patients Time after menopause mo " " Ovarian morphology Corrugated surface Small, some adhesions Small, atrophic Pathology Uterine fibroid Uterine fibroid Adenoma to us hyperplasia of endometrium Adenomatous hyperplasia of endometrium Fibroid, cervical carcinoma in situ Uterine fibroid, vaginal carcinoma Adenoma to us hyperplasia of endometrium "Time of menopause was difficult to establish because of prolonged, irregular, uterine bleeding associated with uterine fibromyomata.

3 Vol. 27, No.7 GONADOTROPIN BINDING TO HUMAN OVARY 791 FIGs. 1 TO 4. Alternate sections from a postmenopausal ovary (x 90). FIG. 1. A section showing cortical stromal cells (S), corpora albicantia (A), and blood vessels (arrowhead) (hematoxylin and eosin). FIG. 2. A section demonstrating G6PD activity within the cortical stromal cells. FIGs. 3 AND 4. Autoradiographs showing '"I-hLH (Fig. 3) and 125!-hFSH (Fig. 4) binding to the cortical stroma. The corpora albicantia do not bind either '2 '!-labeled gonadotropin.

4 792 PELUSO ET AL. July 1976 FIGS. 5 TO 7. Hilus cells found within the postmenopausal ovary (Fig. 5) have G6PD activity (Fig. 6) and hLH binding sites (Fig. 7) (x 360).

5 Vol. 27, No. 7 GONADOTROPIN BINDING TO HUMAN OVARY 793 FIGS. 8 AND 9. The blood vessels within the ovary bind both hLH (Fig. 8) and hFSH (Fig. 9). The gonadotropins occasionally bind only to the inner and outer surfaces of the ovarian arteries (arrow) (x 270). RESULTS Evaluation of Clinical Material. The ovaries selected for this investigation were obtained from women 50 to 68 years of age undergoing hysterectomy and bilateral salpingo-oophorectomy. These patients were 2 months to 13 years postmenopausal. All ovaries were free of any gross pathology. Postoperative examination of the uteri ofthe seven women revealed four cases of uterine fibroids and three cases of endometrial hyperplasia. One patient had cervical carcinoma in situ and another had vaginal carcinoma (Table 1). In two patients, the time of menopause was difficult to establish because of prolonged, irregular, uterine bleeding associated with fibromyomata, but these women had not experienced uterine bleeding for 2 to 3 months prior to surgery. Ovarian Histology. The ovaries did not contain follicles in any stage of development, but corpora albicantia in various stages of degeneration were often observed (Fig. 1). Stromal cells were noted in the ovarian cortex. In some ovaries, stromal cells were aggregrated in dense areas and arranged in a pattern resembling cortical stromal hyperplasia. Other dense clusters of stromal cells were identified as cortical stromal fibrosis by their dominant collagen staining. Hilus cells and blood vessels were observed throughout the medulla of the postmenopausal ovary. G6PD Activity and Gonadotropin Binding. Preliminary autoradiographic studies have shown that 2 to 3 x 10 5 cpm/50 ILl of either hlh or hfsh were required to visualize specific binding sites within the ovary. When sections were incubated with both labeled and unlabeled gonadotropin, binding to the cortical stroma, hilus, and blood vessels was substantially reduced, indicating binding specificity.

6 794 PELUSO ET AL. July 1976 In all ovaries studied, G6PD activity was found in both the cortical stromal cells and hilus cells (Figs. 2 and 6). These cell types also bound both 125 I labeled gonadotropins (Figs. 3, 4, and 7). In contrast, the cells of the corpora albicantia did not demonstrate either G6PD activity or the ability to bind 125 I-hLH or 125 I-hFSH (Figs. 2 to 4). The arterioles and venules found in postmenopausal ovaries bound both labeled gonadotropins, but 125 I-hFSH binding was often more intense than 125 I-hLH binding (Figs. 8 and 9). The pattern of gonadotropin binding to the arterioles varied within the same ovary. In some cases, the labeled gonadotropin bound uniformly to the entire wall of the arteriole, whereas binding to other arterioles was confined to the inner and outer surfaces of the arterial wall. Both gonadotropins bound uniformly to ovarian venules (Figs. 8 and 9). DISCUSSION The postmenopausal ovary does not have follicles capable of ovulating but retains a limited endocrine function. Numerous reports have shown that postmenopausal women have higher serum and urine steroid levels than do women who have undergone bilateral oophorectomy. However, there is extensive individual variation in the percentage of postmenopausal women whose ovaries secrete estrogens. The lipid-rich luteinized cells of postmenopausal ovarian stromal hyperplasia resemble the theca interna cells of the follicle and contain several enzymes, including 3,8-hydroxysteroid dehydrogenase, which are necessary for steroid synthesis: 1 9-lt It is assumed that these cells are the sites of steroid synthesis in the postmenopausal ovary. 12 Other studies have shown a positive correlation between the number of G6PD-active cells in the stroma of the postmenopausal ovary and the degree of estrogenic stimulation of the endometrium.3 The present study demonstrates that the cortical stromal and hilus cells of the postmenopausal ovary have both G6PD activity and the ability to bind labeled LH and FSH. In contrast, the cells of the corpora albicantia do not possess either G6PD activity or gonadotropin binding sites. These observations suggest that cells of the stroma and hilus in the postmenopausal ovary can bind circulating gonadotropins. Since these cells do possess gonadotropin binding sites, their ability to synthesize steroids may be regulated by circulating LH and FSH. In postmenopausal women the concentrations of testosterone, androstenedione, 17,8-estradiol, and estrone are higher in ovarian venous blood than in peripheral blood. 5 The differences between ovarian and peripheral levels are 15-fold for testosterone, 4-fold for androstenedione, and 2-fold for both 17,8- estradiol and estrone. Since LH promotes testosterone synthesis in ovarian tissue, 13 LH binding to the stromal and hilus cells could stimulate androgen biosynthesis. In the rat ovary, FSH stimulates the ovarian aromatizing enzyme system that converts androgens into estrogens. 14 Since a small amount of estrogen is produced by the postmenopausal ovary, 5 the presence of FSH binding sites and G6PD activity suggest that the stromal cells may also be a site of estrogen synthesis. The hilus cells of the postmenopausal ovary also have binding sites for hlh and hfsh as well as G6PD activity, suggesting that these cells have steroidogenic activity. Histochemically, the hilus cells are similar to the interstitial cells of the testes and are generally assumed to produce and~gens. 3 The presence of gonadotropin binding sites suggests that gonadotropins may regulate androgen biosynthesis in the hilus cells. Ovarian arteries and veins. also bind

7 Vol. 27, No.7 GONADOTROPIN BINDING TO HUMAN OVARY 795 both labeled gonadotropins. The functional significance of this phenomenon is not fully understood. Preliminary studies have shown that the blood vessels within the rat ovary also bind 125 I-labeled gonadotropin. 15 In rat and rabbit ovaries, both LH and FSH can regulate ovarian blood flow It is not known whether gonadotropins affect ovarian blood flow by a direct action on the blood vessels or indirectly through increased steroid synthesis. However, the ability of the ovarian blood vessels to bind gonadotropins suggests that gonadotropins could regulate blood flow by a direct effect on the wall of ovarian blood vessels. Acknowledgments. The authors are indebted to Drs. M. E. Azzam and T. N. Evans for providing the ovarian tissue and to Mr. Bruce Borin for providing the iodinated hormones. REFERENCES 1. Riley GM: Endocrinology of the climacteric. Clin Obstet Gynecol 7:432, Paulsen CA, Leach RB, Sandberg H, Sheinfield S, Maddock WO: Function of the postmenopausal ovary. Comparison of urinary estrogen and gonadotropin excretion and response to administration of FSH in postmenopausal and ovariectomized women. JAm Geriatr Soc 6:803, Novak ER, Goldberg B, Jones GS, O'Toole RV: Enzyme histochemistry of the menopausal ovary associated with normal and abnormal endometrium. Am J Obstet Gynecol 93:669, Brandau H, Brandau L, Mestwerdt W: Endocrine activity in postmenopausal ovaries. Eur J Obstet Gynecol Reprod Bioi [Suppl 1] 4:S187, Judd HL, Judd GE, Lucas WE, Yen SSC: Endocrine function of the postmenopausal ovary: concentration of androgens and estrogens in ovarian and peripheral vein blood. J Clin Endocrinol Metab 39:1020, Greenwood FC, Hunter WM, Glover JS: The preparation of 131 I-labeled human growth hormone of high specific radio-activity. Biochem J 89:114, Midgley AR: Autoradiographic analysis of gonadotropin binding to rat ovarian tissue sections. In Receptors for Reproductive Hormones, Edited by BW O'Malley, AR Means. New York, Plenum Press, 1972, p Pearse AGE: Histochemistry-Theoretical and Applied. London, Churchill Livingstone, Scully RE, Cohen RB: Oxidative-enzyme activity in normal and pathologic human ovaries. Obstet Gynecol 24:667, Fienberg R: The stromal theca cell and postmenopausal endometrial adenocarcinoma. Cancer 24:32, Pudsen S, Ikonen M, Procope BJ, Saure A: Androst-5-ene-3,8, 17,8-diol in ovary of postmenopausal hyperoestrogenic women. Acta Endocrinol (Kbh) 58:364, Wall E, Hertig AT, Smith GVS, Johnson LT: The ovary in endometrial carcinoma. Am J Obstet Gynecol 56:617, Villee C, Channing CP, Eckstein B, Sulovic V: Effect of gonadotropins on the metabolism of luteinized and follicular ovarian tissue. In The Gonads, Edited by KW McKems. New York, Meredith, 1969, p Moon YS, Darrington JH, Armstrong DT: Stimulatory action of follicle-stimulating hormone on estradiol-17,b secretion by hypophysectomized rat ovaries in organ culture. Endocrinology 97:244, Peluso JJ: Unpublished data 16. Janson PO: Effects of luteinizing hormone on blood flow in the follicular rabbit ovary, as measured by radioactive microspheres. Acta Endocrinol (Kbh) 79:122, Cons JM, Kragt CL: Changes in ovarian circulation following FSH administration to HCGprimed adult rats. Bioi Reprod 11:440, 1974