REPRODUCTIVE BIOLOGY. University of Oulu, Oulu, Finland

REPRODUCTIVE BIOLOGY FERTILITY AND STERILITY VOL. 78, NO. 1, JULY 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Effects of follicle-stimulating hormone (FSH) and human chorionic gonadotropin in individuals with an inactivating mutation of the FSH receptor Tommi E. Vaskivuo, M.D., a,b Kristiina Aittomäki, M.D., Ph.D., c Mikko Anttonen, M.D., d Aimo Ruokonen, M.D., Ph.D., e Riitta Herva, M.D., Ph.D., b Yoshio Osawa, M.D., f Markku Heikinheimo, M.D., Ph.D., d Ilpo Huhtaniemi, M.D., Ph.D., g and Juha S. Tapanainen, M.D., Ph.D. a,b University of Oulu, Oulu, Finland Received August 15, 2001; revised and accepted December 27, 2001. Supported by the Sigrid Jusélius Foundation, the Academy of Finland, the Emil Aaltonen Foundation, and Oulu University Hospital and Helsinki University Central Hospital research funds. Reprint requests: Juha Tapanainen, M.D., Ph.D., Department of Obstetrics and Gynecology, Oulu University Hospital, P.O. Box 5000, FIN-90014 University of Oulu, Finland (FAX: 358-8-315-4310; E-mail: juha.tapanainen@ oulu.fi). a Department of Obstetrics and Gynecology. b Department of Pathology. c Department of Medical Genetics, University of Helsinki, Helsinki, Finland. d Children s Hospital, University of Helsinki, Helsinki, Finland. e Clinic of Clinical Chemistry. f Medical Foundation of Buffalo Research Institute, Buffalo, New York. g Department of Physiology, University of Turku, Turku, Finland. 0015-0282/02/$22.00 PII S0015-0282(02)03148-5 Objective: To study the gonadal steroid responses to FSH and hcg in individuals with the inherited Finnish-type inactivating Ala189Val mutation of the FSH receptor gene. Design: Prospective clinical and descriptive study. Setting: University hospital. Patient(s): Two women and one man homozygous for the Ala189Val mutation of the FSH receptor gene, and ovarian biopsies from four affected and four healthy women, and four normal fetuses. Intervention(s): Individuals were treated with increasing doses of recombinant FSH (300 IU/day start, 900 IU/day final) and/or a single dose of hcg (5000 IU). Ovarian biopsies were used in immunohistochemical analyses for detection of aromatase cytochrome P450 and transcription factor GATA-4. In situ 3 -end labeling analyses were used for detection of apoptosis. Main Outcome Measure(s): Measurements of serum concentrations of follicle-stimulating hormone, leuteinizing hormone, inhibin A and B, estradiol, testosterone (T), androstenedione, and prolactin, immunostaining for ovarian aromatase, GATA-4, and apoptosis. Result(s): Administration of FSH had no effect on production of the steroids. Similarly, human chorionic gonadotropin (hcg) treatment, alone or after FSH administration, failed to raise serum steroid concentrations. Ovarian apoptosis was absent, and the expression of transcription factor GATA-4 and aromatase was negligible in the ovarian biopsies from Ala189Val homozygous individuals. Conclusion(s): The Ala189Val mutation of the FSH receptor gene results in a complete block of FSH action in vivo. Furthermore, the failure of hcg to increase both ovarian estradiol and testosterone secretion emphasizes the possible contribution of FSH in regulating ovarian androgen synthesis, and supports the concept that both gonadotropins are necessary for appropriate ovarian steroidogenesis in humans. (Fertil Steril 2002;78:108 13. 2002 by American Society for Reproductive Medicine.) Key Words: FSH, FSH receptor mutation, ovary, testis, aromatase, apoptosis, GATA-4 Follicle-stimulating hormone (FSH), together with leuteinizing hormone (LH), is responsible for normal ovarian follicular development and spermatogenesis. Its action is transmitted into target cells via a membranebound receptor. The FSH receptor belongs to a family of G-protein-coupled receptors; the LH, thyroid-stimulating hormone (TSH), and FSH receptors form a subclass within this large gene family, distinguished by a large extracellular domain. Follicle-stimulating hormone stimulates follicular maturation and granulosa cell estrogen production in the ovary and Sertoli cell functions in the testis. We earlier discovered a recessively inherited inactivating point mutation of the FSH receptor gene that causes C to T transition at position 566 of exon 7, predicting an Ala to Val substitution at residue 189 in the extracellular domain of the receptor molecule (1). In vitro functional testing of the Ala189Val FSH receptor mutation in transfected myeloid suppressor cells, MSC-1, dem- 108

TABLE 1 Clinical characteristics of patients with Ala189Val FSH receptor mutation. Patient Sex Age Height (cm) Weight (kg) BMI (kg/m 2 ) A Female 33 157 60 24 B Female 29 168 65 23 C Male 33 178 68 21 Vaskivuo. Gonadal steroidal response in FSH receptor mutation. Fertil Steril 2002. onstrated a markedly reduced binding capacity and complete inhibition of signal transduction (1). In two studies, 22 Finnish women (1) and 5 men (2) were found to be heterozygous for the mutation. In the women, the mutation resulted in poor development of secondary sex characteristics, as well as primary amenorrhea and infertility. Serum FSH and LH concentrations were elevated to postmenopausal levels and E 2 concentrations exceeded postmenopausal levels but were low compared with those in ovulating women. Furthermore, the ovaries of female homozygotes have been found to contain high numbers of primordial and primary follicles but only occasional signs of further follicular development (3). It is unknown whether the low amounts of circulating estrogens are derived from peripheral tissues or from the ovaries. Male homozygotes for the Ala189Val mutation have been found to be capable of producing sperm, and two out of five men studied have fathered children, although they had oligoasthenospermia. They had normal serum testosterone (T) levels and were normally masculinized but had low-normal to clearly low testicular volume (2). In addition to the first mutation of FSH receptor described by our group, six other inactivating FSH receptor mutations have been described in humans, all of them in women (4). All the studies describing women with inactivating mutations demonstrate variable disturbances in ovarian function, such as follicular arrest at the early antral or antral stage and lowered E 2 and inhibin production (5, 6). These findings are supported by similar observations in animal models: female FSH and FSH receptor knockout mice are infertile as a result of arrested follicular development; males have low testicular volume and weight, and lowered production and quality of sperm (7 9). In the cycling ovary, FSH is responsible for the recruitment of growing follicles and granulosa cell proliferation. Recruited follicles either ovulate or, more likely, degenerate through the mechanism of apoptosis (10). One of the genes regulated by FSH in the ovary is that for the transcription factor GATA-4. It is present in the human ovary in primary follicles onward (11), and its expression in mouse ovary has been shown to be FSH dependent (12). Another factor reflecting FSH action in the ovary is the enzyme aromatase cytochrome P450, which is located in the granulosa cells of growing follicles and which regulates the conversion of androgens to estrogens (13). As the findings of our previous studies on the Ala189Val mutation have significantly changed the concept of the role of FSH in reproductive function, especially male reproduction, we have now conducted in vivo stimulation studies to back up our previous in vitro findings and to gain further insight into the role of FSH in ovarian steroidogenesis. To this end, we treated two women and one man, homozygotes for the Ala189Val mutation, with high doses of recombinant human FSH and examined serum inhibin and steroid hormone responses and follicular status. Furthermore, we examined GATA-4 and aromatase expression and the extent of apoptosis in ovarian biopsies obtained from women with Ala189Val FSH receptor mutation and compared the findings with results from normal adult and fetal ovaries. MATERIALS AND METHODS Patients Two women and one man with previously diagnosed Ala189Val FSH receptor mutation (1, 2) were included in the study. The clinical characteristics are shown in Table 1. Both of the women had elevated serum FSH and LH concentrations (woman A: 58.5 and 40.1 IU/L; woman B: 48.0 and 23.8 IU/L, respectively), lowered serum E 2 concentrations, and primary amenorrhea. Both women were infertile. These clinical findings are typical of the mutation. Both of the women had been on hormone replacement therapy for years, which was discontinued 1 month before the study. Possible ovarian follicular development was followed by using transvaginal ultrasonography. The man also had elevated serum FSH and LH concentrations (47.3 and 13.5 IU/L, respectively), a normal serum T concentration, lownormal testicular size (8.6/6.0 ml), and lowered sperm motility and quality. He does not have children, and he and his partner have been treated for infertility. Ovarian biopsy specimens had been obtained from four women at the time of primary diagnosis of ovarian dysgenesis. They all had a typical phenotype and were later confirmed to be homozygotes for the Ala189Val FSH receptor mutation. For comparison, normal ovarian tissue was obtained from four women (22 to 45 years old) who were undergoing ovariectomy as a primary treatment of endometrial cancer. In addition, ovaries from aborted fetuses (n 4, fetal age 16 to 33 weeks) were studied. The study was approved by the ethics committee of Oulu University. FSH and hcg Stimulation The study participants were treated with daily subcutaneous injections of recombinant human FSH (Ares-Serono, Geneva, Switzerland) for 9 days. The starting dose was 300 IU of FSH/day. On the third day of treatment, the dose was FERTILITY & STERILITY 109

TABLE 2 Effect of treatment with hcg and/or FSH on serum hormone concentrations in patients with Ala189Val FSH receptor gene mutation: serum concentrations of inhibin B, testosterone, and estradiol. Day Patient Sex Treatment Hormone 0 1 2 4 6 7 8 9 11 14 A Female hcg a Inhibin B (ng/l) 10.00 10.00 10.0 10.00 10.00 T (nmol/l) 1.90 3.00 1.70 2.20 2.00 E 2 (nmol/l) 0.05 0.05 0.04 0.04 0.06 FSH b Inhibin B (ng/l) 10.0 10.00 10.00 10.00 10.0 10.0 10.00 10.00 T (nmol/l) 1.60 1.90 2.80 2.50 2.40 2.20 3.10 1.00 E 2 (nmol/l) 0.07 0.02 0.01 0.06 0.04 0.02 0.02 0.05 B Female FSH b hcg c Inhibin B (ng/l) 10.0 10.0 10.0 10.00 10.0 10.00 10.00 10.00 T (nmol/l) 0.80 0.90 0.80 0.80 0.90 c 1.10 0.70 0.90 E 2 (nmol/l) 0.03 0.03 0.04 0.02 0.03 0.04 0.04 0.03 C Male FSH b Inhibin B (ng/l) 41.40 38.90 33.30 24.70 T (nmol/l) 8.80 9.20 8.40 9.30 E 2 (nmol/l) 0.06 0.06 0.07 0.07 a hcg dose: day 0, 5000 IU im. b FSH dose: days 0 1, 300 IU; days 2 3, 600 IU; days 4 8, 900 IU. c hcg dose: day 7, 5000 IU im. Vaskivuo. Gonadal steroidal response in FSH receptor mutation. Fertil Steril 2002. increased to 600 IU/day; on the fifth day, it was increased to 900 IU/day. Higher doses were not used because the injection volume would have become too large and uncomfortable for the individuals. To assess the possible ovarian follicular response to FSH, transvaginal ultrasonography was performed at the beginning of the study, and thereafter at 1 and 2 weeks of treatment. On another occasion, woman A was given a single injection of hcg only (5000 IU, Pregnyl Organon, Oss, The Netherlands). Her serum steroid concentrations were followed for 8 days. Woman B was given a single injection of 5000 IU of hcg after FSH treatment for 6 days (Table 2). Hormone Measurements Serum inhibin A and B dimer concentrations were separately determined by commercial enzyme-linked immunosorbent assays (ELISAs), using monoclonal antibodies to the common subunit and the specific A and B subunits of inhibin (Serotec Limited, Oxford, UK). Serum concentrations of T and prolactin (PRL) were determined by using a chemiluminescence system (ACS-L80, Medfield, MA). Serum concentrations of LH and FSH were determined by fluoroimmunoassays (Wallac Ltd., Turku, Finland), as were those of E 2 (Orion Diagnostica, Oulunsalo, Finland) and androstenedione (A) (Diagnostic Products Corporation, Los Angeles, CA), following the instructions of the manufacturers. All hormones were assayed in one batch. The intra-assay variation was 5.5% for inhibin A, 5.2% for inhibin B, 4.0% for T, 3.8% for PRL, 4.9% for LH, 3.8% for FSH, 5.7% for E 2, and 5.0% for A. Immunohistochemistry For aromatase cytochrome P450 immunohistochemistry, a specific rabbit antibody against immunoaffinity-purified human placental aromatase was used. The purified aromatase was more than 97% homogeneous, with a molecular weight of 55,000 kd, and it showed a 300-fold to 400-fold increase in aromatase activity (14). Transcription factor GATA-4 immunohistochemistry was studied by the use of goat polyclonal anti-mouse GATA-4 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA). Immunohistochemical analyses of aromatase and GATA-4 were carried out following normal procedures. Paraffin sections were deparaffinized in xylene and hydrated gradually through graded alcohols. Endogenous peroxidase activity was prevented by incubating the samples in 225 ml of methanol with 25 ml of 30% hydrogen peroxidase. Primary antibody was applied to the sample and it was incubated overnight. Further steps for aromatase and GATA-4 immunostaining were carried out using Vectastain goat antirabbit and Vectastain rabbit anti-goat ABC kits (Vectastain Elite ABC Kit, Vector Laboratories, Burlingame, CA), respectively. The sections were incubated with biotinylated secondary antibody for 1 hour, and the results were visualized with 0.02% diaminobenzene in phosphate-buffered saline. Counterstaining was carried out with hematoxylin. In Situ 3 -End Labeling of Ovarian Tissue The 3 -end labeling of apoptotic DNA was carried out as previously described (15), using an ApopTag in situ apoptosis detection kit (Oncor, Gaithersburg, MD). Endogenous 110 Vaskivuo et al. Gonadal steroid response in FSH receptor mutation Vol. 78, No. 1, July 2002

peroxidase activity was quenched with 2% hydrogen peroxidase in phosphate-buffered saline (ph 7.2). Diaminobenzene was used to develop the color reaction and the specimens were counterstained with hematoxylin. RESULTS Functional Study of C566T FSH Receptor Mutation In Vivo Recombinant human FSH treatment effectively raised serum FSH concentrations in the women from 48.0 58.5 IU/L to 98.8 116.6 IU/L at 9 to 11 days of treatment measured 24 hours after the injection. In the man, the increase of FSH was less marked; from 47.4 to 65.3 IU/L at 5 days. However, the increased FSH concentrations did not result in gonadal stimulation, as measured by serum inhibin A or B concentrations. All of the individuals had undetectable serum inhibin A concentrations ( 3 ng/l) throughout the FSH treatment. Similarly, the women had undetectable quantities of circulating inhibin B ( 10.0 ng/l) during the FSH treatment. In the man, FSH treatment did not increase serum inhibin B concentrations. Furthermore, there were no changes in serum E 2, T, A, or PRL concentrations in any of the individuals during FSH treatment (see Table 2). No ovarian follicular development was observed by transvaginal ultrasonography during FSH therapy. A single injection of hcg (5000 IU, woman A) resulted in a slight increase of serum T, but E 2 levels did not change. In woman B, after 6 days of FSH treatment the administration of hcg did not affect the serum concentrations of T and E 2 (see Table 2). Aromatase Histologically, the ovarian architecture of the women with Ala189Val FSH receptor gene mutation resembled that of fetal tissue more than that of an adult ovary: a large number of primordial follicles were present in the ovaries of the women with the FSH receptor mutation, as well as in fetal ovaries; only a few follicles showed further development. Immunohistochemical staining of aromatase in the normal adult ovaries was observed in the granulosa cells of growing follicles; staining was absent or negligible in the ovaries of individuals with FSH receptor mutation. Similarly, fetal ovaries showed only minimal aromatase staining (Fig. 1). GATA-4 In normal adult and fetal ovarian tissue, GATA-4 was expressed in follicular granulosa cells, as was also observed in our previous study (11). In contrast, ovarian tissue from individuals with FSH receptor mutation showed very low or negligible GATA-4 expression (see Fig. 1). Apoptosis In ovarian tissue from women with the FSH receptor mutation, the numerous primordial follicles did not contain any apoptotic cells. Only two follicles were observed to have developed beyond the primary follicle stage, but even they were negative as regards apoptosis. In contrast, significant apoptosis was observed in the oocytes and granulosa cells of fetal ovarian tissue as early as at the primordial follicle stage, and it was also observed in growing follicles of adult ovaries (see Fig. 1), as we have previously described (11). DISCUSSION Our findings demonstrate that Ala189Val mutation of the FSH receptor gene results in complete blockade of FSH action in vivo. Treatment with high doses of recombinant human FSH did not affect the serum levels of inhibin, T, or E 2, and ultrasonographic examination did not reveal any signs of ovarian follicular growth. Supporting this finding, ovarian biopsies from women with the Ala189Val FSH receptor mutation showed only negligible GATA-4 expression, which has been shown to be FSH dependent (12). Normal cycling ovaries have an increased granulosa cell turnover rate in the growing follicles, and significant apoptosis is observed in granulosa cells (11, 16). The Ala189Val FSH receptor mutation resulted in stagnation of follicular growth and in the absence of follicular apoptosis. Furthermore, the Ala189Val FSH receptor mutation resulted in decreased aromatase expression in the ovary compared with normal ovaries, indicating low ovarian estrogen synthesis. Administration of a single dose of hcg, alone or after FSH pretreatment, did not affect serum E 2 concentrations, although levels of the E 2 precursor T increased slightly in woman A. This emphasizes the importance of FSH in regulating aromatase activity and estrogen biosynthesis. According to the two-cell-two-gonadotropin theory, hcg should have provided sufficient stimulus to induce theca cell production of estrogen precursors. Thus, in the absence of the paracrine FSH-mediated priming of theca cells, no steroidogenic response to hcg was observed. Recently, Barnes et al. (17) reported new evidence against a strict two-cell-two-gonadotropin mode of action of ovarian steroidogenesis after treating a woman with FSH deficiency with hcg alone, or after priming the ovaries with FSH. Administration of FSH prior to the hcg dose resulted in greater serum androgen concentrations than hcg alone, suggesting that paracrine factors stimulated by FSH are required for ovarian steroid production. Our results support these findings and provide further evidence for the concept that in humans both gonadotropins (FSH and LH) are necessary for proper ovarian steroidogenesis. An interesting question is why women with the Ala189Val FSH receptor gene mutation have variable but often normal breast development despite their low circulating estrogen levels. This is in accordance with findings in FSH receptor knockout female mice, which have low circulating estrogen concentrations but increased T levels (8). Similarly, in our study, serum T concentrations in woman A FERTILITY & STERILITY 111

FIGURE 1 Aromatase cytochrome P450 and GATA-4 expression, and apoptosis in ovarian tissue from individuals with the Ala189Val FSH receptor gene mutation and from normal adult and fetal ovarian tissue. (A), In ovaries from women with the FSH receptor mutation, only negligible aromatase immunostaining was observed. (B), In normal adult ovaries strong aromatase immunostaining was found in growing follicles. (C), Only very weak aromatase staining was observed in fetal ovaries. (D), Women with the FSH receptor mutation showed only negligible ovarian GATA-4 expression. (E), In normal adult ovaries, GATA-4 immunostaining was observed in granulosa cells of growing follicles. (F), In normal fetal ovaries, GATA-4 staining was observed in follicular granulosa cells as well as in stromal cells. (G), Ovarian apoptosis was not detected in women with the FSH receptor mutation. (H), In normal adult ovarian tissue, apoptosis was localized to the follicular granulosa cells. (I), In normal fetal ovarian tissue, apoptosis was mainly localized to the oocytes, although apoptotic granulosa cells were also observed. Scale bars 5 m. Vaskivuo. Gonadal steroidal response in FSH receptor mutation. Fertil Steril 2002. were high-normal, which may be due to a relative increase of ovarian stroma, high serum LH levels, and/or decreased ovarian conversion of T to estrogens. This may in turn result in increased tissue E 2 concentrations through peripheral aromatization of T to E 2, which is supported by the finding that woman B had about 50% lower LH levels and consequently lower T levels compared with woman A. On the other hand, a compensatory increase of adrenal androgen secretion in these individuals cannot be excluded. The lack of FSH action in stimulation studies was confirmed by the results of ovarian immunohistochemical studies and by in situ 3 -end labeling of apoptotic DNA. In mouse ovarian granulosa cells, GATA-4 expression has been 112 Vaskivuo et al. Gonadal steroid response in FSH receptor mutation Vol. 78, No. 1, July 2002

shown to be induced by FSH (12). Furthermore, previous studies in humans have suggested that ovarian apoptosis occurs mainly in growing follicles, that is, follicles under FSH control (11, 18). The absence of apoptotic cells in the ovaries of individuals with FSH receptor mutation supports this concept. In conclusion, the present evidence indicates that the Ala189Val FSH receptor gene mutation causes a complete blockade of FSH action in vivo, as we have previously shown in cell culture models. Furthermore, the absence of a steroidogenic response to FSH and hcg emphasizes the importance of both gonadotropins (FSH and LH) in estrogen biosynthesis, and suggests the possible role of paracrine factors regulated by FSH in ovarian androgen synthesis. Acknowledgments: The authors thank Mirja Ahvensalmi, laboratory nurse, for her skilled technical assistance. References 1. Aittomäki K, Lucena JL, Pakarinen P, Sistonen P, Tapanainen J, Gromoll J, et al. Mutation in the follicle-stimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 1995; 82:959 68. 2. Tapanainen JS, Aittomäki K, Min J, Vaskivuo T, Huhtaniemi IT. Men homozygous for an inactivating mutation of the follicle-stimulating hormone (FSH) receptor gene present variable suppression of spermatogenesis and fertility. Nat Genet 1997;15:205 6. 3. Aittomäki K, Herva R, Stenman UH, Juntunen K, Ylöstalo P, Hovatta O, et al. Clinical features of primary ovarian failure caused by a point mutation in the follicle-stimulating hormone receptor gene. J Clin Endocrinol Metab 1996;81:3722 6. 4. Themmen APN, Huhtaniemi IT. Mutations of gonadotropins and gonadotropin receptors: elucidating the physiology and pathophysiology of pituitary-gonadal function. Endocr Rev 2000;21:551 83. 5. Beau I, Touraine P, Meduri G, Gougeon A, Desroches A, Matuchansky C, et al. A novel phenotype related to partial loss of function mutations of the follicle stimulating hormone receptor. J Clin Invest 1998;102: 1352 9. 6. Touraine P, Beau I, Gougeon A, Meduri G, Desroches A, Pichard C, et al. New natural inactivating mutations of the follicle-stimulating hormone receptor: correlations between receptor function and phenotype. Mol Endocrinol 1999;13:1844 54. 7. Kumar TR, Wang Y, Lu N, Matzuk MM. Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility. Nat Genet 1997;15:201 4. 8. Danilovich N, Babu PS, Xing W, Gerdes M, Krishnamurthy H, Sairam MR. Estrogen deficiency, obesity, and skeletal abnormalities in folliclestimulating hormone receptor knockout (FORKO) female mice. Endocrinology 2000;141:4295 308. 9. Krishnamurthy H, Danilovich N, Morales CR, Sairam MR. Qualitative and quantitative decline in spermatogenesis of the follicle-stimulating hormone receptor knockout (FORKO) mouse. Biol Reprod 2000;62: 1146 59. 10. McGee EA, Hsueh AJ. Initial and cyclic recruitment of ovarian follicles. Endocr Rev 2000;21:200 14. 11. Vaskivuo TE, Anttonen M, Herva R, Billig H, Dorland M, te Velde ER, et al. Survival of human ovarian follicles from fetal to adult life: apoptosis, apoptosis-related proteins, and transcription factor gata-4. J Clin Endocrinol Metab 2001;86:3421 9. 12. Heikinheimo M, Ermolaeva M, Bielinska M, Rahman NA, Narita N, Huhtaniemi IT, et al. Expression and hormonal regulation of transcription factors GATA-4 and GATA-6 in the mouse ovary. Endocrinology 1997;138:3505 14. 13. Tamura T, Kitawaki J, Yamamoto T, Osawa Y, Kominami S, Takemori S, et al. Immunohistochemical localization of 17 alpha-hydroxylase/ C17-20 lyase and aromatase cytochrome P-450 in the human ovary during the menstrual cycle. J Endocrinol 1992;135:589 95. 14. Yoshida N, Osawa Y. Purification of human placental aromatase cytochrome P-450 with monoclonal antibody and its characterization. Biochemistry 1991;30:3003 10. 15. Vaskivuo TE, Stenbäck F, Karhumaa P, Risteli J, Dunkel L, Tapanainen JS. Apoptosis and apoptosis-related proteins in human endometrium. Mol Cell Endocrinol 2000;165:75 83. 16. Morita Y, Tilly JL. Oocyte apoptosis: like sand through an hourglass. Dev Biol 1999;213:1 17. 17. Barnes RB, Rosenfield RL, Namnoum A, Layman LC. Effect of follicle-stimulating hormone on ovarian androgen production in a woman with isolated follicle-stimulating hormone deficiency. N Engl J Med 2000;343:1197 8. 18. Yuan W, Giudice LC. Programmed cell death in human ovary is a function of follicle and corpus luteum status. J Clin Endocrinol Metab 1997;82:3148 55. FERTILITY & STERILITY 113