Screening of the mutations in luteinizing hormone p-subunit in patients with menstrual disorders*

Transcription

1 FERTILITY AND STERILITY Copyright e 1995 American Society for Reproductive Medicine Printed on acid-free paper in U. s. A. Screening of the mutations in luteinizing hormone p-subunit in patients with menstrual disorders* Nobuhiko Suganuma, M.D.t Kenji Furui, M.D. Madoka Furuhashi, M.D. Yoshimasa Asada, M.D.:j: Fumitaka Kikkawa, M.D. Yutaka Tomoda, M.D. Department of Obstetrics and Gynecology, Nagoya University School of Medicine, Nagoya, Japan Objective: To evaluate the clinical significance of the LH consisting of a mutant,b-subunit (Trp8 to Ar~ and Ile 15 to Thr 15 ). Design: Clinical and biochemical studies. Setting: Fertility center at the University Hospital and its research laboratory. Patients: Fifty-one patients with menstrual disorders and three homozygote cases and two heterozygote cases of the mutant LH who were reported previously. Interventions: Nucleotide mutations of the LH,B gene in patients with menstrual disorders were screened using techniques of the polymerase chain reaction and restriction fragment length polymorphism. Immunologic and biologic activities of the mutant LH and endocrinologic profiles in the affected women were evaluated. Main Outcome Measures: Serum LH levels measured with different immunoassay kits; serum FSH and LH on the GnRH test; serum thyroid-stimulating hormone, PRL, T, and androstenedione; ultrasound examination of the ovaries; clinical hyperandrogenic symptoms; and biologic activity of LH. Results: Two cases of homozygotes and four of heterozygotes affected by the LH,B gene mutations were discovered in the current study through screening of patients with menstrual disorders. Serum LH levels in the homozygote cases were undetectable using a LH immunoassay kit, whereas levels in the heterozygote cases showed reduced detectability with the kit. However, the ratio of the mutant LH values in the bioassay to those in the immunoassay was higher in the homozygote group than that in the control subjects. Response patterns of serum gonadotropins to GnRH in the homozygote were similar to those in patients with polycystic ovary syndrome. Conclusion: The mutations of LH,B-subunit might be related to menstrual disorder in some patients. Fertil Steril1995;63: Key Words: LH,,B-subunit, gene mutation, menstrual disorder, polycystic ovary syndrome, bioassay and immunoassay of LH Received April 28, 1994; revised and accepted December 5, * Supported in part by a scientific grant (no ) from the Ministry of Education, Culture and Science, Tokyo, Japan. t Reprint requests: Nobuhiko Suganuma, M.D., Department of Obstetrics and Gynecology, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466, Japan (FAX: ). :j: Present address: The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Norfolk, Virginia. Luteinizing hormone is a glycoprotein hormone that is synthesized and secreted in the anterior pituitary. It plays important roles in the regulation of gonadal functions. In women, LH promotes ovulation and luteinization of the ovarian follicle and enhances steroid production in the ovaries. From a clinical perspective, reduced LH secretion induces anovulation or luteal insufficiency, leading to menstrual disorders and infertility. Measuring serum LH levels to clarify the endocrine status is essential in diagnosing patients with menstrual disorders. Ultrasensitive and specific immunoassays for human LH have been developed (1, 2) using monoclonal antibodies (3, 4). However, because these monoclonal antibodies may be very specific to indigenous LH, some studies have noted restricted reactivity by the monoclonal antibodies in some populations (5, 6). In an earlier study (7), we developed a polymerase Suganuma et a1. LHf3 mutations in menstrual disorders 989

2 chain reaction (PCR) procedure that allows specific amplification of the LH,B gene. Subsequent nucleotide sequencing revealed two point mutations in the gene coding of the LH,B-subunit, indicating two amino acid replacements: Trp8 crgg) to Ar~ (QGG) and Ile 15 (ATC) to Thr 15 (AQC) among three women with immunologically anomalous LH. Further examination by restriction fragment length polymorphism (RFLP) revealed an existence of homozygotes and heterozygotes of the mutant LH,B gene in their family members, which indicates the genetic traits of the abnormal LH molecule. Although the homozygotes were of various grades of ovarian dysfunction, the clinical significance of the LH variant and the effects of the mutant LH on gonadal function remained unclear. To better evaluate these factors, mass screening of the LH mutations in patients with menstrual disorders was conducted using our previously developed PCR procedure followed by RFLP. Through these analyses, we found two new homozygotes and four new heterozygotes that are affected by the LH,B mutations. The immunologic and biologic activities of the mutant LH and hormonal profiles of these women were examined also in conjunction with the cases reported previously (7). Subjects MATERIALS AND METHODS Fifty-one patients with menstrual disorders participated in this study. Ages varied from 16 to 36 years (26.1 ± 5.1 years; mean ± SD). These patients exhibited symptoms of primary or secondary amenorrhea (no menstruation for >6 months), menstrual irregularity such as polymenorrhea (intervals < 18 days) or oligomenorrhea (intervals> 45 days), or anovulatory cycles determined by BBT records, measurements of serum E2 and P levels, and serial sonography of their ovaries. Clinical features such as the body mass index, grade of hirsutism, and symptoms specific to each disease were recorded. Fifty women ranging from 22 to 37 years (28.7 ± 5.5 years) with regular menstrual cycles (intervals between 23 and 39 days) were used as control subjects. Ovulation was confirmed by BBT and clinical and laboratory examinations such as serum E2 and P increases observed during the ovulatory and midluteal phases, respectively, and serial ultrasound imaging of their ovaries to observe follicle growth and disappearance of the follicle after ovulation. From 54 patients with menstrual disorders, including 3 cases of homozygotes for the mutant LH,B-subunit (cases 1,2, and 3) as reported previously (7), and from 52 women with regular menstrual cy- 990 Suganuma et ai. LH(3 mutations in menstrual disorders cles, including 2 age-matched cases ofheterozygotes (sisters of case 2 [31 years] and case 3 [22 years]) reported previously (7), peripheral blood was taken before and 30 and 60 minutes after an IV injection of 100 J.lg GnRH (GnRH test) on the 2nd to 4th day of menstruation or withdrawal bleeding with estrogen and P. Blood samples were collected to screen for the LH,B gene mutations as described below, and serum samples were separated and kept at -20 C until assaying for various hormones. Based on the clinical features and laboratory examinations of the 54 patients, 8 women were diagnosed with polycystic ovary syndrome (PCOS), 8 with hyperprolactinemia, 4 with Turner's syndrome, 3 with congenital GH deficiency, and one each with congenital panhypopituitarism, premature ovarian failure, and amenorrhea induced by severe weight loss. All other patients with no specific clinical causes for their menstrual disorder were classified as idiopathic hypothalamic, pituitary, or ovarian failure as determined by serum gonadotropin levels on the GnRH test. Patients with high serum FSH levels (> 15 miu/ml; conversion factor to SI unit, 1.00) were categorized as ovarian failure. Patients with low or normal FSH levels (s15 miu/ml) were diagnosed with hypothalamic failure if the peak LH level was ~7 miu/ml (conversion factor to SI unit, 1.00) and the peak:basal LH ratio was ~2 on the GnRH test; all others were classified as pituitary failure. Clinical PCOS was evaluated according to the criteria for PCOS proposed by The Committee for Reproduction and Endocrine in Japan Society of Obstetrics and Gynecology (8). The criteria involve menstrual disorder, high LH (~7 miu/ml) and normal FSH (4 to 14 miu/ml) in basal serum levels and ultrasound detection of multiple ovarian cysts (see Discussion). All procedures were performed after the subjects provided informed consent. Specific Amplification of LHP Gene and RFLP Analysis The nuclear DNA fractions were obtained from peripheral blood lymphocytes (9) and were used directly as a template DNA for the PCR as described previously (7). In brief, the LH,B genes of the patients and control subjects were amplified with primers of F1 (5'-GAAGCAGTGTCCTTGTCCCA-3') and R1 (5'-GAAGAGGAGGCCTGAGAGTT-3'), which span exon 2, intron 2, and exon 3 of the LH,B gene. Total genomic DNA was subjected to 25 cycles of PCR using a thermalcycler (Perkin Elmer-Cetus, Norwalk, CT) in a 100-J.lL reaction mixture containing U/mL of Taq polymerase (Perkin Elmer-Cetus), 50 mmolll KCI; 10 mmolll Tris-CI, ph 8.3; 1.5 mmolll Fertility and Sterility

3 MgCI 2 ; 0.01% (wt/vol) gelatin; 200 JLmol/L dntp (50 JLmol/L each of datp, dctp, dgtp, and dttp); and 1.0 JLmol/L primers. Protocol for the reaction was set as follows: 1 minute at 94 C, 2 minutes at 65 C, and 3 minutes at 72 C. These methods already have been proven to amplify only the LH,B gene without contaminating of the hcg,b genes (7). Restriction fragment length polymorphism analysis was conducted to screen for the mutations of the LH,B gene in both the menstrual disorder patients and the control subjects. Because the mutations at Trp8 to Arg 8 (CCAQGG) and at Ile l5 to Thr l5 (NNN NNNNNNNNNNCAQCC) destroyed the recognition sites of Ncol (CCATGG) and Fokl (NNNNNNNNN NNNNCA1CC), respectively, LH,B gene fragments amplified by PCR using primers of F1 and R1 were digested along with these restriction enzymes, which were purchased from Takara Co. (Tokyo, Japan). An aliquot of the reaction mixture was examined by 1% agarose gel electrophoresis with 3% Nusieve GTG agarose (FMC BioProduct, Rockland, ME) added. The mutant LH,B-subunit gene in the homozygotes detected by RFLP analysis were sequenced to confirm the substitutions of the nucleotides (data not shown). Immunoassay of Serum LH and Other Hormones Two commercial immunoassay kits were used for serum LH determination. One was the SPAC-S LH Kit (Spac) purchased from Daiichi Radioisotope Laboratory (Tokyo, Japan), which uses an immunoradiometric assay method (1) and two different monoclonal antibodies that react with the,b-subunit and intact dimer, respectively. The other was an immunofluorometric assay kit, DELFIA LHspec (Delfia), made by Wallac (Turku, Finland) (2); the recognition sites for the two monoclonal antibodies in this kit are both located in the,b-subunit. World Health Organization First International Reference Preparation-LH 68/40 was supplied as a reference for standard LH in the immunoassay kits. A ratio of the Spac:Delfia findings was calculated using basal LH measured?7 mlu/ml by the Delfia kit. When the basal LH level was <7 mlu/ml by Delfia, the peak LH values on the GnRH test were used for the calculation because more critical evaluation could be done with higher hormonal levels. Serum FSH, PRL, thyroid-stimulating hormone, T, and androstenedione were measured using RIA kits specific for each substance at Ohtsuka Assay Laboratory in Tokushima, Japan. Bioassay of Serum LH The bioactivities of serum LH in the 5 cases of homozygotes and the 6 cases ofheterozygotes as well as in the women with normal LH,B gene (6 cases of PCOS and 50 control subjects) were evaluated by in vitro T production using mouse interstitial cells, as described by Van Damme et al. (10) with some subsequent modifications (11). In brief, decapsulated testicles from 7- to 9-week-old male Swiss-Webster mice were minced in medium (Dulbecco's Modified Eagle's Medium containing 0.5% heat-inactivated calf serum). Tissue was gently stirred, filtered through 70 to 80 mm mesh to obtain the interstitial cells, and then incubated in the medium at 34 C for 1 hour in 95% O2-5% CO2 After preincubation, the cells were washed and diluted, and then were placed into 48- well plates (100 JLL; 6 to 7 X 10 4 cells) containing standards (human LH, #L5020; Sigma, St. Louis, MO), control serum, or test serum at a final test volume of 100 JLL. All samples and standards were run in triplicate. Plates were incubated for 2 hours at 37 C in a humidified incubator with 5% CO 2-95% air. The plates were used for RIA of T. The lower limit of detection was approximately 5.0 lull LH. The intraassay and interassay coefficients of variation were 5% and 15%, respectively. Statistical Analysis Data are expressed as the mean ± SD. The unpaired Student's t-test was used for statistical analysis. P < 0.05 was considered significant. RESULTS Screening of the Mutations in the LH p-subunit Gene in Patients with Menstrual Disorders Luteinizing hormone,b-subunit genes in the 51 patients with menstrual disorders were analyzed using the PCR procedure and subsequent RFLP analysis. Two homozygotes (cases 4 and 5; lanes 1 and 35 in Fig. 1) and four heterozygotes (lanes 24, 38, 46, and 47 in Fig. 1) for the mutant LH,B gene were found through Fokl (Fig. 1) and Ncol (data not shown) digestion. In the 50 control subjects, however, no anomalous LH,B gene was detected (data not shown). Comparing the Values of LH with Different Immunoassay Kits Ratios of the LH concentrations were calculated based on findings by Spac and Delfia in serum samples collected from 54 women with menstrual disorders and 52 women with normal ovarian function. Serum LH in the five homo zygotes was undetectable «0.1 mlu/ml) with Spac, whereas the six heterozygotes showed reduced detectability. The ratio of Spac-to-Delfia measurements in the heterozygotes ranged from 0.17 to 0.51 (0040 ± 0.13) whereas, in Suganuma et a1. LHf3 mutations in menstrual disorders 991

4 H H M H M H M H M H bp 433 ~390 -"-176 obesity, or serum androgen (T or androstenedione) increases were observed. The other three cases could not be diagnosed as PCOS because their ovaries showed no formation of multiple cysts. They also did not suffer from obesity, hirsutism, or serum androgen increases. Although case 3, who was diagnosed as idiopathic pituitary failure, had a high serum LH level, a normal response pattern of serum LH to GnRH was observed. Cases 2 and 5, who were categorized as having idiopathic hypothalamic failure and in whom serum basal LH levels were normal, showed LH responses similar to those observed for PCOS patients on the GnRH test. Among the six patients with PCOS, who were not affected by the mutations in the LH,6-subunit, five showed hyperandrogenemia (increased levels of serum T and/or androstenedione) but no obesity or hirsutism except for one case. The four heterozygotes were discovered in one patient with hyperprolactinemia, two with hypothalamic failure, and one with pituitary failure. The other two heterozygotes, who are sisters of the homozygotes, have regular menstrual cycles. None of the heterozygotes showed hyper-response of serum LH to the GnRH injection (data not shown). Figure 1 Restriction fragment length polymorphism analysis of the patients with menstrual disorders. Genomic DNA was extracted from 51 patients, and PCR amplification was performed using primers offl and Rl. The amplified LH,B gene was digested with FokI to detect mutation at Ile 15 (ATC) to Thr 15 (ACC). FokI digested the fragment into three parts-(433, 176, and 51 base pairs [bpj) when the mutation was present but into four parts (390, 176, 51, and 43 bp) when the mutation was absent. Fragments in the heterozygotes included five parts by the digestion. These fragments were separated on 1% agarose gel with 3% Nusieve GTG agarose. Through these analyses, two homozygotes (lanes 1 [case 4] and 35 [case 5J) and four heterozygotes (lanes 24, 38, 46, and 47) were detected. The DNA molecular weight marker (lane M; q,x174 RF DNAlHaeIII fragments) and the digestion pattern ofthe LH,B gene from a heterozygote (lane H; mother of case 1 [reference 7J) also are indicated on each gel. patients and control subjects with no mutant LH,6 gene, the ratio ranged from 0.66 to 1.63 (1.10 ± 0.21). Characteristics of the Hormonal Profiles in the Homozygotes for the Mutant LH The serum LH response patterns to GnRH and clinical and endocrinologic features in the homozygotes are outlined in Figure 2 and Tables 1 and 2, respectively. Cases 1 and 4 showed high LH and normal FSH in serum and a hyper-response of LH on the GnRH test, similar to those in patients with PCOS. These cases were diagnosed as clinical PCOS because multiple ovarian cysts also were detected by ultrasound imaging. However, no signs of hirsutism, 992 Suganuma et al. LH,B mutations in menstrual disorders Bioassay of Serum LH in the Homozygotes and Heterozygotes In the homozygotes (2.7 ± 0.8), the ratios of LH values in the bioassay to those in the immunoassay using the Delfia kit were significantly higher than those in the control subjects with normal LH (1.5 ± 0.3) (Fig. 3). The heterozygotes (1.8 ± 0.4) and PCOS patients not affected by the mutant LH (1.9 ± 0.3) showed a somewhat higher bioassay:immuno- 60..J... : ::c 4..J 2 E III minutes after GnRH injection Figure 2 Responses in serum LH to GnRH stimulation in the homozygotes. Changes in the serum LH levels determined by Delfia kit in cases 1 through 5 are drawn as a function of the time course after the GnRH injection. Shaded area (1It\l) and dotted area (1ITl), respectively, indicate the regions of the mean:!: SD in the 8 patients with pcas and the 50 control subjects in this series. Fertility and Sterility

5 Table 1 Clinical Profiles of the Homozygotes for the Mutant LH Menstrual Body Body Case Age disorder Diagnosis height weight Body mass Ovarian polycysts Ovulation index Hirsutism* by ultrasound without hcg y em kg 1 31 Amenorrhea PCOS Anovulatory cycle Hypothalamic Oligomenorrhea Pituitary Amenorrhea PCOS Anovulatory cycle Hypothalamic No Yes No 23 No No Yes 20 No No Yes 21 No Yes Yes 18 No No Yes * No indicates <8 by the score of Ferriman and Gallwey (12). assay ratio when compared with the control subjects, but the difference was not significant. DISCUSSION As a follow-up to our previous report in which two amino acid replacements in the LH,B (Trp8 to Arg 8 and Ile l5 to Thr l5 ) were detected in patients with immunologically anomalous LH (7), we screened for the mutations of the LH,B-subunit gene in patients with menstrual disorders in this study. In 51 patients randomly screened, 2 cases of homozygotes (3.9%) and 4 cases of heterozygotes (7.8%) for the LH,B mutations were found. However, none of the control subjects with regular menstrual cycles and ovulation were affected by the mutations in the LH,B-subunit. To evaluate the clinical and pathological significance of the mutant LH, the endocrinologic status of women affected by the mutations in the LH,Bsubunit gene were analyzed. Two women (cases 1 and 4) in the homozygote patients were diagnosed as peos based on clinical and laboratory examinations whereas the other three cases of homozygotes showed oligomenorrhea with high LH in the serum basal level (case 3) or anovulatory cycles with a hyper-response of LH on the GnRH test (cases 2 and 5). Polycystic ovary syndrome is a heterogeneous disorder containing no simple definition. This disease often is characterized by hyperandrogenism, menstrual abnormalities, and obesity in women with enlarged polycystic ovaries. Although obesity and hirsutism are the primary clinical symptoms in peos patients (13), only a respective 20% and 23% of the peos patients in Japan suffer from obesity and hyperandrogenic symptoms (8). Elevated serum androgen levels are observed frequently in peos patients; however, the percentage of patients with increased androgen secretion (50% in T and 35% in androstenedione) is lower among Japanese than other races. The Japanese Society of Obstetrics and Gynecology proposed diagnostic criteria for clinical peos in Japan (8); menstrual disorder, high LH and normal FSH in serum basal levels, and multiple-cyst formations detected by ultrasound examination are indispensable. Other observations such as virilism, obesity, or infertility as clinical symptoms, a hyperresponse of LH and a normoresponse of FSH on the GnRH test, a high ratio of estrone:e 2 or a high serum T or androstenedione level as laboratory findings, and macroscopic or histologic changes in the ovaries as pathological findings are consultative. Using the criteria for clinical peos to evaluate the gonadotropin levels and the response pattern on the GnRH test, the homozygotes for the mutant LH were similar to findings in patients with peos. The mutant LH with a higher bioactivity than native LH in the present study and the hereditary trait of the mutations in the LH,B-subunit observed by pedigree Table 2 Endocrinologic Profiles of the Homozygotes for the Mutant LH LH* Case Spac Delfia mlu/ml 1 Undetectable Undetectable Undetectable Undetectable Undetectable 3.5 * Conversion factors to SI unit, t Conversion factor to SI unit, FSH* miu/ml TSH* PRL* Tt Androstenedione:!: I'U/mL ng/ml ng/ml ng/dl :!: Conversion factor to SI unit, Suganuma et al. LHf3 mutations in menstrual disorders 993

6 4 f ~ 3 c ~ E.5... > o :a = 1 * O---L--r----r----r----r- Figure 3 Ratios of LH values in the bioassay to those in the immunoassay in the homozygote (e), the heterozygote (_), and the PC OS patient without mutant LH (0). Dotted area (El) indicates the mean:+: SD region for the ratio derived from 50 control subjects. Serum LH levels were measured immunologically using a Delfia kit and biologically by in vitro T production in mouse interstitial cell. *P < analysis of the three family members (cases 1, 2, and 3) in the previous study (7) also are compatible with findings reported for PCOS patients by Lobo et al. (14) and Cooper et al. (15). However, we could not establish completely the relationship between the mutations in the LH,B-subunit and PCOS because only a limited number of patients were evaluated. It should be noted, however, that sensitive immunoassay systems have proven the existence of microheterogeneity of LH in large populations even in physiological situations (5, 16-18). Pettersson et al. (6) also reported one "healthy" woman in whom one of the monoclonal antibodies did not detect LH. Although it is not clear whether anomalous immunoreactivity of the LH depends on mutations of LH,Bsubunit, because the gene structure of the LH,B-subunit in their woman was not analyzed, this report indicates the possibility that the microheterogeneity oflh,b may not affect the physiological roles of LH. On the contrary, because no women with the mutations in the LH,B-subunit were found among the control subjects in the current study, we may speculate that the mutant LH affects gonadal function and that the LH microheterogeneity may be related at least in part to menstrual irregularity. As reported previously (7), serum LH in the homozygotes was undetectable and that in the heterozygotes showed reduced detectability using the Spac 994 Suganuma et at. LH(3 mutations in menstrual disorders kit. Because the Spac:Delfia ratios did not overlap among the homozygote cases, the heterozygote cases, or the women with indigenous LH, we may conclude that measuring serum LH with specific immunoassay systems can identify affected women completely from normal subjects. This screening test can be useful in detecting the mutant LH,B gene, and further analysis with much larger samples from women with menstrual disorders, including PCOS patients, will serve to further clarify the clinical significance of the mutant LH. Moreover, future studies using the purified mutant LH expressed in cells transfected with the mutant,b-subunit gene should be performed to estimate the biologic effects of the mutant LH more directly. Acknowledgments. We thank Ms. Susan Wilson and Ms. Catherine Boyd, Eastern Virginia Medical School, Norfolk, Virginia, for the bioassay of LH, and Ms. Megumi Ono for her technical assistance. We are indebted to members of the Department of Obstetrics and Gynecology at Nagoya University School of Medicine and its Branch Hospital, for their support in obtaining blood samples from patients with menstrual disorders. REFERENCES 1. Odell WD, Griffin J. Two-monoclonal-antibody "sandwich" type assay of human lutropin, with no cross reaction with choriogonadotropin. Clin Chern 1987;33: Haavisto AM, Dunkel L, Pettersson K, Huhtaniemi I. LH measurements by in vitro bioassay and a highly sensitive immunofiuorometric assay improve the distinction between boys with constitutional delay of puberty and hypogonadotropic hypogonadism. Pediatr Res 1990;27: Wide L, Bennich H, Johansson SGO. Diagnosis of allergy by an in-vitro test for allergen antibodies. Lancet 1967;2: Kohler G, Milstein C. Continuous cultures off used cells secreting antibody of predefined specificity. Nature 1975;256: Pettersson KS, Soderholm JR. Individual differences in lutropin immunoreactivity revealed by monoclonal antibodies. Clin Chern 1991;37: Pettersson K, Ding Y-Q, Huhtaniemi I. An immunologically anomalous luteinizing hormone variant in a healthy woman. J Clin Endocrinol Metab 1992;74: Furui K, Suganuma N, Tsukahara S, Asada Y, Kikkawa F, Tanaka M, et al. Identification of two point mutations in the gene coding luteinizing hormone (LH) (3-subunit, associated with immunologically anomalous LH variants. J Clin Endocrinol Metab 1994;78: The Committee for Reproduction and Endocrine in Japan Society of Obstetrics and Gynecology. Annual report ( ) for the determination of diagnostic criteria for polycystic ovary syndrome (Japanese). Acta Obstet Gynaecol Jpn 1993;45: Gross-Bellerd M, Oudet P, Chambon P. Isolation of high molecular weight DNA from mammalian cells. Eur J Biochem 1973;36: Van Damme M-P, Robertson DM, Diczfalusy E. An improved in vitro bioassay method for measuring luteinizing hormone (LH) activity using mouse Leydig cell preparations. Acta Endocrinol (Copenh) 1974;77: Fertility and Sterility

7 11. Gordon K, Williams RF, Danforth DR, Hodgen GD. Antideinduced suppression of pituitary gonadotropin and ovarian steroid secretion in cynomolgus monkeys: premature luteolysis and prolonged inhibition of folliculogenesis following single treatment. BioI Reprod 1991;44: Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21: Goldzieher JW, Axelrod LR. Clinical and biochemical features of polycystic ovarian disease. Fertil Steril 1963; 14: Lobo RA, Kletzky OA, DiZerega GS. Elevated serum bioactive LH concentrations in women with PCO. Fertil Steril 1982; 37: Cooper HE, Spellacy WN, Prem KA, Cohen WD. Hereditary factors in the Stein-Leventhal syndrome. Am J Obstet Gynecol 1968; 100: Reader SC, Robertson WR, Diczfalusy E. Microheterogeneity of luteinizing hormone in pituitary glands from women of pre- and postmenopausal age. Clin Endocrinol (Ox ) 1983; 19: Apter D, Cacciatore B, Alfthan H, Stenman UH. Serum luteinizing hormone concentrations increase 100-fold in females from 7 years to adulthood, as measured by time-resolved immunofluorometric assay. J Clin Endocrinol Metab 1989; 68: Jaakkola T, Ding YQ, Kellokumpu LP, Valavaara R, Martikainen H, Tapanainen J, et al. The ratios of serum bioactivel immunoreactive luteinizing hormone and follicle-stimulating hormone in various clinical conditions with increased and decreased gonadotropin secretion: reevaluation by a highly sensitive immunometric assay. J Clin Endocrinol Metab 1990; 70: Suganuma et al. LHf3 mutations in menstrual disorders 995