Dynamics of bioactive follicle-stimulating hormone secretion in women with polycystic ovary syndrome: effects of estradiol and progesterone

Transcription

1 FERTILITY AND STERILITY VOL. 75, NO. 5, MAY 2001 Copyright 2001 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Dynamics of bioactive follicle-stimulating hormone secretion in women with polycystic ovary syndrome: effects of estradiol and progesterone Received July 12, 2000; revised and accepted November 20, Supported by grants HD (Dr. Beitins), HD (Drs. Kelch and Marshall), and HD (Dr. Padmambhan) from the National Institutes of Health and Clinical Research Center grant 5MO1-RR-42. Presented in part at the 72nd Annual Meeting of the Endocrine Society, Atlanta, Georgia, 1990, and the Ares Serono Symposium on Cell and Molecular Biology of the Testis, Rome, Italy, Reprint requests: Vasantha Padmanabhan, Ph.D., Department of Pediatrics, Box 0404, 300 N. Ingalls Building, University of Michigan, Ann Arbor, Michigan (FAX: ; a Department of Pediatrics, University of Michigan, Ann Arbor, Michigan. b Department of Obstetrics and Gynecology, University of Michigan, Ann Arbor, Michigan. c College of Medicine, University of Iowa, Iowa City, Iowa. d Department of Internal Medicine, University of Virginia, Health Science Center, Charlottesville, Virginia. e National Center for Research Resources, National Institutes of Health, Bethesda, Maryland /01/$20.00 PII S (01) Vasantha Padmanabhan, Ph.D., a Gregory M. Christman, M.D., b John F. Randolph, M.D., b Robert P. Kelch, M.D., c John C. Marshall, M.D., Ph.D., d and Inese Z. Beitins, M.D. e University of Michigan Medical Center, Ann Arbor, Michigan Objective: To determine the nature of bioactive FSH secretion in anovulatory women with polycystic ovary syndrome (PCOS) and its modulation by luteal levels of E 2 and P. Design: Interventional and observational study. Setting: Academic clinical research center. Patient(s): Five patients with PCOS. Intervention(s): Treatment for 21 days with luteal levels of E 2 and P. Main Outcome Measures: Serum levels of immunoreactive LH, immunoreactive FSH, bioreactive FSH, and the FSH isoform distribution pattern. Blood was sampled frequently and GnRH testing was done on day 0 (before treatment), days 10 and 20 (during treatment), and day 28 (7 days after treatment). Result(s): Treatment with E 2 and P suppressed circulating immunoreactive LH and immunoreactive FSH but not bioreactive FSH. Anovulatory women with PCOS showed a predominantly acidic pattern of FSH isoform distribution. Treatment with E 2 and P shifted the distribution profile of FSH isoforms to the less acidic. After cessation of E 2 P treatment, FSH reverted to its pretreatment pattern of distribution. Conclusion(s): Resumption of follicular growth after luteal replacement of E 2 and P in anovulatory women with PCOS may be related to the reduction in the elevated LH/FSH ratio and accompanying changes in the FSH signal. (Fertil Steril 2001;75: by American Society for Reproductive Medicine.) Key Words: Gonadotropins, bioactivity, FSH, hyperandrogenism, PCOS, anovulation Several lines of evidence point to the role of circulating FSH in the maturation of ovarian follicles; prevention of follicular atresia; proliferation of granulosa cells; and induction of aromatase, LH receptor, and FSH receptor (1 3). Research in the past two decades has established that FSH circulates as a mixture of heterogeneous isoforms of varying immunologic and biologic potencies (4). These molecular variations alter the metabolic clearance and actions of FSH. The proportion of various isoforms that reach the target cell may not only influence receptor binding but also signal transduction in various measurable biological end points. Existing evidence suggests that the endocrine milieu regulates the proportions of the FSH isoforms within the pituitary and circulation (4). In general, relatively high proportions of less acidic FSH isoforms or an increase in secretion of bioactive FSH, as assessed by using in vitro bioassays, accompany conditions in which circulating levels of E 2 are high (4). For instance, serum levels of bioreactive FSH, as measured by using the Sertoli cell bioassay (5), increase (6) during the late follicular phase of the menstrual cycle. This period is characterized by an increase in circulating levels of E 2. This increased FSH bioactivity is accompanied by the prevalence of less acidic FSH isoforms in the circulation (6, 7). In contrast, men have a more acidic mix of FSH isoforms (8, 9). Androgens appear to regulate the incorporation of sugar residues into the carbohydrate chains of pituitary FSH that favor the biosynthesis of complex-type oligosaccharide structures (10). 881

2 Women with polycystic ovary syndrome (PCOS) are hyperandrogenic and often anovulatory, and most have elevated serum levels of LH (11, 12). In a previous study, we demonstrated that administration of E 2 and P to anovulatory women with PCOS slowed GnRH pulse frequency, normalized the ratio of LH to FSH, and facilitated follicular growth (13). In this study, we sought to characterize the nature of bioactive FSH secretion in anovulatory women with PCOS and to determine the effects of luteal replacement of E 2 and P on FSH bioactivity and heterogeneity. MATERIALS AND METHODS To characterize circulating bioreactive FSH in women with PCOS, we obtained samples from a subset (n 5) of previously studied women with PCOS for whom serum was available (13). The study was approved by the institutional review board at the University of Michigan. The participant selection criteria and experimental protocol are reported elsewhere (13). All participants were nonobese (body mass index 25.0); had a history of oligomenorrhea with no menses for the preceding 90 days; and had clinical and laboratory features characteristic of PCOS, including elevated plasma levels of LH and T and follicular or anovulatory levels of P. Patients were taking no medications and had not received estrogen, P, progestins, or oral contraceptives within 90 days before the start of the study. All patients had normal levels of fasting glucose and insulin, prolactin, dehydroepiandrosterone sulfate, and 17-hydroxyprogesterone in response to ACTH stimulation testing and thyroid function tests. A normal immunoreactive FSH level and an elevated LH/FSH ratio 2.5 was recorded in all patients before the start of the study. Patients were treated for 21 days with luteal levels of E 2 (Estraderm transdermal patches; Ciba-Geigy, Summit, NJ), 0.1 mg/24 h, changed every 3 days, and P (vaginal suppositories, 50 mg in polyethlene glycol matrix, administered three times daily). In the absence of information documenting the ability of E 2 and P to slow LH pulse frequency in women with PCOS and the possibility that the slowing effect of E 2 and P may be diminished in women with PCOS, E 2 and P were provided for 21 days. Patients were studied on day 0 (before treatment), days 10 and 20 (during E 2 P replacement), and day 28 (7 days after treatment). The study points selected during E 2 P treatment corresponded to a physiologic period of luteal steroid exposure (day 10) and an extended period of exposure (day 20). The unique physiologic status and multiple endocrine alterations in patients with PCOS made selection of appropriate control groups difficult. The women with PCOS in this study exhibited no overt insulin resistance, a very rapid LH (GnRH) pulse frequency with a background of low follicular estradiol levels, and elevated levels of androgens. To appropriately control for the elevated GnRH pulse frequency, we needed to study women with normal cycles in the preovulatory period. These women, in contrast to PCOS women, have normal androgen levels but elevated E 2 levels. Therefore, to study the impact of physiologic luteal concentrations of E 2 and P on FSH heterogeneity in anovulatory PCOS women, we used each participant as her own control. This design limits our ability to extend our findings to normal women provided with luteal level replacement of E 2 and P or other subsets of women with PCOS who manifest other endocrine abnormalities, such as overt insulin resistance. Levels of bioreactive and immunoreactive LH were measured in serum samples drawn at hourly intervals from 12:00 15:00 and at 10-minute intervals from 15:00 18:00 during each of four study phases. The first hour of 10-minute sampling (15:00 16:00) corresponded to basal release (before GnRH stimulation testing), and the second two hours of 10-minute sampling corresponded to time points after administration of synthetic GnRH, 25 and 250 ng/kg, at 16:00 and 17:00, respectively (Beckman Co., Palo Alto, CA). Because of the long half-life of FSH (4, 5) and limited sample availability, serum levels of immunoreactive FSH were measured only at hourly intervals. Serum E 2 and P concentrations were determined at 12:00 (before the sampling period), 16:00 (before GnRH administration), and 18:00 (2 hours after the GnRH challenge). Radioimmunoassays Serum samples were stored at 20 C until assayed. Serum levels of immunoreactive FSH and immunoreactive LH were measured in duplicate by using double antibody radioimmunoassay. Values for circulating levels of immunoreactive LH and immunoreactive FSH from the 2nd International Reference Preparation have been reported (13). For comparison with bioreactive FSH, circulating levels of immunoreactive LH and immunoreactive FSH were remeasured by using National Institute of Diabetes and Digestive and Kidney Diseases hlh-3 and National Institute of Diabetes and Digestive and Kidney Diseases hfsh-3 (National Hormone and Pituitary Program, Bethesda, MD); we report these levels as ng/ml. For each hormone, all samples from each patient were assayed in the same assay. The intraassay and interassay coefficients of variation for the FSH assay averaged 5.6% 1.9% and 11.8% 1.2%, respectively, and the sensitivity averaged ng/ml (mean SE of 5 assays). The sensitivity and intraassay and interassay coefficients of variation of the LH assay on the basis of three quality control pools averaged ng/ml, 4.3% 1.4% and 9.5% 1.8% (mean SE), respectively. Serum E 2 and P concentrations were measured by using radioimmunoassay kits (Estradiol-Diagnostic Products Corp., Los Angeles, CA, and P-Radioassay Systems Laboratories, Inc., Carson, CA, respectively). 882 Padmanabhan et al. Bioactive FSH in PCOS Vol. 75, No. 5, May 2001

3 TABLE 1 Circulating levels of androgens, E 2, and P before and after E 2 P replacement in women with PCOS. Substance Day 0 Day 10 Day 20 Day 28 Testosterone (ng/ml) Androstenedione (ng/ml) E 2 (pg/ml) a a a P (ng/ml) a a Note: Values are means SE of samples taken hourly before GnRH administration. a P.05. Bioassay Serum levels of bioreactive FSH were measured by using the Sertoli cell aromatase induction assay (5) with hfsh-3 as the bioassay standard (the same as the immunoassay standard). Samples were measured in triplicate at volumes of 10 L. The sensitivity and intraassay and interassay coefficients of variation for the FSH bioassays (mean of 15 assays and based on 5 quality control pools) averaged ng/ml, 8%, and 18%, respectively. Levels of E 2 in the culture medium after Sertoli cell bioassay were measured by using a double-antibody radioimmunoassay. The sensitivity and intraassay and interassay coefficients of variation of this assay averaged 10 pg/ml, 5%, and 8%, respectively. In previous studies, we showed that long-term storage of serum samples and freeze-thaw do not affect bioreactive FSH measurements (14). Chromatofocusing Chromatofocusing separation was used to determine the isoform distribution pattern of FSH on days 0, 10, and 28 of E 2 P treatment. Low circulating levels of FSH precluded characterization of the FSH isoform distribution after 20 days of E 2 P treatment. Because limited quantities of serum were available, equal aliquots of serum were pooled from each woman for each time period tested. The details of the chromatofocusing method have been published elsewhere (6). For chromatofocusing fractionation, 9 10 ml of serum was used. Samples were dialyzed in M imidazole-hcl before application to a cm column packed with PBE-94 resin. A 1:10 dilution of polybuffer 74 (ph, 4) was applied for the collection of the first 80 fractions. Ten additional fractions were collected after application of 1MofNaClwash. The ph and immunoreactive FSH content of each fraction were determined. Data Analyses To account for heterogeneity in variance, hormonal data were transformed logarithmically before statistical analyses. Significant differences in mean serum gonadotropin concentrations were determined by analysis of variance for repeated measures and were compared by using Scheffe test statistics. A P value.05 was considered significant. RESULTS Effect of E 2 P Treatment on Circulating Levels of T, Androstenedione, E 2, and P Table 1 shows circulating steroid levels before, during, and after E 2 P treatment. As was reported for the larger cohort, treatment with E 2 and P increased circulating concentrations of E 2 to ranges seen during the mid-luteal phase (P.01). Levels of P were clearly in the luteal range but did not reach levels found at the peak mid-luteal period. Circulating levels of total T and androstenedione did not statistically differ between treatment periods. Total androgen/progesterone ratio was reduced by 96% and 97% after 10 and 20 days of E 2 P treatment, respectively (P.01); during the same period, the total androgen/estrogen ratio was reduced by 66% and 70%, respectively (P.01). Effect of E 2 P Treatment on Serum Levels of Immunoreactive LH, Immunoreactive FSH, and Bioreactive FSH Figure 1 shows the patterns of basal and GnRH-stimulated release of serum immunoreactive LH, immunoreactive FSH, bioreactive FSH, E 2, and P before, during, and after E 2 P treatment. Group mean levels of circulating E 2,P,and gonadotropins are shown in Table 2. Circulating levels of E 2 and P increased from early to mid-follicular range before treatment with E 2 P to the luteal range after treatment. Cessation of E 2 P treatment caused levels of E 2 and P to return to pretreatment values. Mean basal serum levels of LH as measured by National Institute of Diabetes and Digestive and Kidney Disorder hlh-3 were suppressed 61% and 80% after 10 and 20 days of E 2 P therapy, respectively. Serum immunoreactive LH concentrations remained suppressed even after cessation of E 2 P treatment. Mean basal serum levels of immunoreactive FSH, levels of bioreactive FSH as measured by NIDDK hfsh-3, and the ratio of bioreactive to immunoreactive FSH are shown in Table 2. Similar to immunoreactive LH, levels of immunoreactive FSH FERTILITY & STERILITY 883

4 FIGURE 1 Serum levels of immunoreactive LH, bioreactive FSH, immunoreactive FSH, E 2, and P in a woman with PCOS. Patients received E 2 and P on study days 1 21 and were studied on day 0, day 10, day 21, and day 28 (1 week after cessation of E 2 P treatment). On each day, serum immunoreactive LH and bioreactive FSH was measured hourly for the first 4 hours (12:00 to 16:00) and at 10 minute intervals for the next 3 hours. Values are the mean SE. were reduced 65% and 75% after 10 and 20 days of E 2 P treatment, respectively. In contrast, E 2 P replacement did not suppress circulating levels of bioreactive FSH; bioreactive FSH concentrations on day 0, 10, 20, and 28 of E 2 P treatment averaged , , , and ng/ml, respectively. After withdrawal of E 2 P treatment, levels of immunoreactive FSH recovered to 72% of pretreatment level, but levels of immunoreactive LH remained suppressed. TABLE 2 Mean circulating levels of immunoreactive LH, immunoreactive FSH, bioreactive FSH and the ratio of bioreactive to immunoreactive FSH in women with PCOS. Substance Day 0 Day 10 Day 20 Day 28 Immunoreactive LH (ng/ml) a a a Immunoreactive FSH (ng/ml) a a a Bioreactive FSH (ng/ml) Ratio of immunoreactive to bioreactive FSH a a Note: Values are means SE of samples taken hourly before GnRH administration. a P Padmanabhan et al. Bioactive FSH in PCOS Vol. 75, No. 5, May 2001

5 TABLE 3 Serum immunoreactive LH and bioreactive FSH responses GnRH. GnRH dose Day 0 Day 10 Day 20 Day ng Immunoreactive LH (ng/ml) a a Bioreactive FSH (ng/ml) a ng Immunoreactive LH (ng/ml) a a Bioreactive FSH (ng/ml) a Note: Values are means SE of change from baseline (peak baseline). a P.05. The reduction in immunoreactive FSH level and the lack of change in bioreactive FSH level produced an increase in the ratio of bioreactive to immunoreactive FSH from before treatment to and after 10 and 20 days of E 2 P, respectively. The ratio returned to pretreatment values after cessation of E 2 P treatment. Effect of E 2 P Treatment on GnRH-Induced Release of Immunoreactive LH, Immunoreactive FSH, and Bioreactive FSH Table 3 summarizes mean responses of immunoreactive LH and bioreactive FSH to intravenous administration of GnRH, 25 and 250 ng/kg. As expected, GnRH at both doses increased serum levels of immunoreactive LH, immunoreactive FSH (assessed from hourly samples; data not shown), and bioreactive FSH. At both doses of GnRH, replacement with luteal levels of E 2 and P in women with PCOS resulted in suppression of GnRH-induced release of immunoreactive LH after 20 days of E 2 P therapy. Withdrawal of E 2 P treatment for 7 days did not produce restoration of GnRH responses to pretreatment values, and GnRH-induced release of immunoreactive LH remained suppressed. Suppression of GnRH-induced release of bioreactive FSH after 20 days of E 2 P treatment was evident only at the 25-ng/kg dose and not at the 250-ng/kg dose. In contrast to the continued suppression of GnRH-induced release of LH after cessation of E 2 P therapy, the suppressive effects of E 2 P replacement on bioreactive FSH responses to the 25- ng/kg dose of GnRH were overcome after cessation of E 2 P therapy. Treatment with E 2 and P did not alter bioreactive FSH responses to the 250-ng/kg dose of GnRH; these responses were similar before, during, and after E 2 P therapy. Effect of E 2 P Therapy on the Distribution Pattern of Serum FSH Isoforms The quantity of serum was not sufficient to characterize the FSH isoform distribution patterns in individual patients. Distribution of FSH isoforms in pooled serum samples before, during (day 10 of E 2 P therapy), and after treatment (7 days after E 2 P therapy) are shown in Figure 2. Before E 2 P treatment, except for elution of FSH isoforms in the void volume (ph 7.0), the FSH distribution pattern was composed of predominantly acidic FSH isoforms that eluted in ph 4.0 ( 70% eluting in the salt peak). After 10 days of E 2 P treatment, relatively more FSH eluted in the less acidic fractions (ph 4.0). The pattern of FSH distribution returned to the pretreatment profile (more acidic) after cessation of E 2 P treatment. The relatively low concentrations of immunoreactive FSH precluded such characterization in the day-20 pool. Follicular development occurred in all 5 patients, and one patient experienced ovulation (13). DISCUSSION The role of gonadotropin heterogeneity in PCOS remains unresolved. Most earlier studies have focused on assessing levels of bioreactive LH and its regulation in women with PCOS. These studies have found that serum levels of bioreactive LH and immunoreactive LH are elevated in 95% and 75% of patients with PCOS, respectively (15). Intravenous administration of GnRH was shown to increase serum levels of bioreactive LH in a manner similar to immunoreactive LH and to maintain a relatively elevated but constant ratio of bioactive to immunreactive LH. Serum levels of bioreactive LH correlated well with serum androgens but not estrogen concentrations. After treatment with GnRH analogues, some investigators (16) have observed an increase in immunoreactive LH and poor bioactivity, whereas others (17) have found proportional decreases in levels of both bioreactive LH and immunoreactive LH, with levels of the former exceeding those of the latter at all times of suppression and recovery. Only two studies (18, 19) have characterized the endocrine regulation of bioreactive FSH in women with PCOS. Using a granulosa cell bioassay, Fauser et al. (18) and van Dessel et al. (19) found that bioreactive FSH levels in women with PCOS were within the normal range. Our findings also indicate that in anovulatory women with PCOS, serum levels of bioreactive FSH as measured by using the Sertoli cell bioassay (5) are within the reported range in adult men (9) and in cycling women during the early follicular FERTILITY & STERILITY 885

6 FIGURE 2 Serum distribution pattern of immunoreactive human (hfsh) after chromatofocusing before (day 0), during (10 and 20 days) and after (day 28) E 2 P therapy. phase of normal cycle (6). These findings are consistent with circulating levels of bioreactive FSH reported elsewhere for women with PCOS (18, 19). Surprisingly, whereas E 2 P treatment suppressed levels of immunoreactive LH and immunoreactive FSH (13), it failed to alter levels of circulating bioreactive FSH in anovulatory women with PCOS. This lack of suppression of serum bioreactive FSH in the presence of reduced levels of immunoreactive FSH levels and the associated dissociation between bioreactive FSH and immunoreactive FSH levels was analogous to the levels reported during the late follicular and mid-cycle phases of the menstrual cycle (6) and other high estrogenic states (4, 9) rather than the levels observed during the luteal phase. The lack of bioreactive FSH suppression during E 2 P replacement therapy in anovulatory women with PCOS does not appear to result from interference of increased serum levels of E 2 during this period with estimates of FSH bioactivity. Serum E 2 levels averaged 73, 141, and 16 pg/ml before (day 0), during (day 10), and after (day 28) E 2 P replacement, respectively. Because absolute levels of E 2 in serum ranged from pg in 10 L of serum (volume tested in the bioassay), lack of suppression of bioreactive FSH by E 2 P treatment does not appear to reflect interference from E 2. However, interference by E 2 was found to be a confounding factor in estimates of FSH bioactivity in samples obtained during late pregnancy (4, 9, 20), when circulating levels of E 2 are extremely elevated. The distribution pattern of FSH isoforms (higher release of less acidic FSH isoforms) during E 2 P treatment seems consistent with the higher ratio of bioreactive to immunoreactive FSH. To our surprise, the changes in FSH heterogeneity that followed E 2 P treatment were similar to those seen during the late follicular phase or mid-cycle and not the luteal phase of the human menstrual cycle (6, 7). In previous studies of women with normal menstrual cycles, we have found a preponderance of bioreactive FSH to immunoreactive FSH during the late follicular and preovulatory period compared with the ratios seen in the early follicular and luteal phases (6). This preponderance of bioreactive FSH was associated with a shift in FSH isoforms from acidic isoforms to more basic isoforms after chromatofocusing separation. The direction of the changes during the normal menstrual cycle are consistent with the high estrogenic status during the late follicular phase, since a similar pattern is also noted in women with gonadal dysgenesis after treatment with estrogenic compounds (6). In general, most of the evidence from studies in rodents and primates indicates that estrogen significantly alters the charge of the isoforms toward less acidic forms (4, 6, 7, 9). Before E 2 P treatment, women with PCOS appear to have a predominantly acidic pattern of circulating FSH. Pituitary FSH is invariably more acidic in men than in women (8, 9). Furthermore, administration of T to rats has been shown to shift the FSH distribution toward the more acidic (10). While investigations of the role of androgens in regulating FSH heterogeneity are not as extensive as for estrogen, the acidic distribution of FSH found in women with PCOS may be related to the elevated androgenic status of these patients. 886 Padmanabhan et al. Bioactive FSH in PCOS Vol. 75, No. 5, May 2001

7 Considering that the progestational (luteal) phase of the human menstrual cycle is characterized by a more acidic FSH distribution pattern (6, 7), the shift in FSH isoform distribution to the less acidic side after replacement with luteal levels of E 2 and P was somewhat surprising. In general, acidic FSH isoforms are found to be associated with progestogenic states (4, 6, 7). Serum bioreactive FSH concentrations as measured by using the Sertoli cell bioassay (5) are low during the luteal phase in normal-cycling women. This reduced bioactivity during the luteal phase of the human menstrual cycle is accompanied by the presence of predominantly acidic FSH isoforms (6). The quality of FSH seen during the luteal phase appears to be a function of the high progestational milieu. Earlier studies have also shown that P can counteract the effects of estrogen (21). The paradoxical increase in the ratio of bioreactive to immunoreactive FSH and less acidic FSH isoforms that followed E 2 P replacement may lie in the protective actions of P against androgens. Many animal studies have shown that P acts as an antiandrogen and can overcome the effects of androgens (22 24). It is therefore likely that P replacement leads to negation of the masculinizing effects of androgen and the consequent manifestations of increased estrogen action. This scenario would probably culminate in a preovulatory pattern of FSH distribution like that which we observed during E 2 P replacement in women with PCOS. In this context, it is of interest that the androgen/progesterone ratio in these patients is reduced about 25-fold during E 2 P treatment. Alternatively, the differences in the FSH distribution pattern seen in women with PCOS treated with luteal levels of E 2 P and that in normal women during the luteal phase may simply relate to an interaction with other ovarian feedback systems. Several regulatory proteins, such as inhibin, activin, and follistatin, have been shown to have regulatory effects on FSH (25). The gonadal protein milieu of women with PCOS may differ from that of normalcycling women. For instance, women with PCOS have been shown to have higher levels of inhibin B than normalcycling women (26). Very limited information is available, however, to document the involvement of inhibin and activin in modulating FSH heterogeneity. A control group of normal women to whom E 2 P is administered is required to properly interpret this paradoxical response of anovulatory women with PCOS to E 2 P treatment. As expected, cessation of E 2 P treatment shifted the FSH isoform distribution to a more acidic profile, a reversal back to the distribution pattern seen before treatment. Overall, these findings suggest that changes in the estrogen progesterone milieu may alter the FSH isoform profile, thus accounting for the differential changes in release of bioreactive and immunoreactive FSH. The differential regulation of bioreactive and immunoreactive FSH by E 2 P may be mediated through direct effects of E 2 and P at the pituitary level or through alterations in hypothalamic input and involve secretory alterations or peripheral modifications. Preliminary studies in sheep have shown that E 2 elicits alterations in secreted forms of FSH (27). Others have reported that such changes in FSH isoform distribution are accompanied by changes in pituitary -2,3-sialyltransferase activity (28). Regardless of the mode by which changes in FSH distribution pattern are produced, the altered mix of FSH isoforms that accompanied E 2 P therapy, along with reduction of LH levels, facilitate the initial progression of follicular development in anovulatory patients with PCOS. Continued follicular development (13) after cessation of E 2 P treatment may result from switch in dependence of the preovulatory follicles from FSH to LH, such as that which occurs during a normal menstrual cycle (1 3). Of note, our findings may apply only to nonobese anovulatory women with PCOS who have high LH levels and do not have overt insulin resistance. Changes in bioactive FSH in this population may not be representative of those in all women with PCOS because of the known heterogeneity of the disorder or the effects of obesity or insulin resistance. Acknowledgment: The authors thank the National Hormone and Pituitary Program for the generous supply of all the purified hormone standards used in this study. References 1. Adashi EY. The ovarian follicular apparatus. In: Adashi EY, Rock JA, Rosenwaks Z (eds). Reproductive endocrinology: surgery and technology. New York: Lippincott Raven, 1995: Findlay JK, Drummond AE. Control of follicular growth. In: Filicori M, Flamingi C (eds). The ovary: regulation, dysfunction and treatment. Amsterdam: Elsevier, 1996: Richards JS, Jahnsen T, Hedin L, Lifka J, Ratoosh S, Durica JM, et al. Ovarian follicular development: from physiology to molecular biology. Recent Prog Horm Res 1987;43: Ulloa-Aguirre A, Midgley AR Jr, Beitins IZ, Padmanabhan V. Follicle stimulating isohormones: characterization and physiological relevance. Endocr Rev 1995;16: Padmanabhan V, Chappel SC, Beitins IZ. An improved in vitro bioassay for follicle-stimulating hormone (FSH): suitable for measurement of FSH in unextracted human serum. Endocrinology 1987;121: Padmanabhan V, Lang LL, Sonstein J, Kelch RP, Beitins IZ. Modulation of serum follicle-stimulating hormone bioactivity and isoform distribution by estrogenic steroids in normal women and in gonadal dysgenesis. J Clin Endocrinol Metab 1988;67: Wide L, Bakos O. More basic forms of both human follicle-stimulating hormone and luteinizing hormone in serum at midcycle compared with the follicular and luteal phase. J Clin Endocrinol Metab 1993;76: Wide L. Follicle-stimulating hormones in anterior pituitary glands from children and adults differ in relation to sex and age. J Endocr 1989; 123: Beitins IZ, Padmanabhan V. Bioactivity of gonadotrophins. In: Styne D (ed). Endocrinology and metabolism clinics in North America. Philadelphia: WB Saunders Co., 1991: Rulli SB, Creus S, Pellizzari E, Cigorraga SB, Calandra RS, Campo S. Androgen regulation of immunological and biological activities of pituitary follicle-stimulating hormone isoforms in male rats. Neuroendocrinology. 1990;70: Ehrmann DA, Barnes RB, Rosenfield RL. Polycystic ovary syndrome as a form of functional ovarian hyperandrogenism due to dysregulation of androgen secretion. Endocr Rev 1995;16: Dunaif A. Insulin resistance and the polycystic ovarian syndrome: FERTILITY & STERILITY 887

8 mechanism and implications for pathogenesis. Endocr Rev 1997;18: Christman GM, Randolph JF, Kelch RP, Marshall JC. Reduction of gonadotropin releasing hormone pulse frequency is associated with subsequent selective follicle-stimulating hormone secretion in women with polycystic ovarian disease J Clin Endocrinol Metab 1991;72: Beitins IZ, Padmanabhan V, Kasa-vubu J, Kletter GB, Sizonenko PC. Serum bioactive follicle-stimulating hormone concentrations from prepuberty to adulthood: a cross-sectional study. J Clin Endocrinol Metab 1990;71: Lobo RA, Shoupe D, Chang SP, Campeau J. The control of bioactive luteinizing hormone secretion in women with polycystic ovary syndrome. Am J Obstet Gynecol 1984;148: Meldrum DR, Tsao Z, Monroe SE, Braunstein GD, Sladek J, Lee JKH, et al. Stimulation of LH fragments with reduced bioactivity following GnRH agonist administration in women. J Clin Endocrinol Metab 1984;58: Calogero AE, Macchi M, Montanini V, Mongioi A, Maugeri G, Vicari E et al. Dynamics of plasma gonadotropin and sex steroid release in polycystic ovarian disease after pituitary-ovarian inhibition with an analog of gonadotropin-releasing hormone. J Clin Endocrinol Metab 1987;64: Fauser BC, Pache TD, Lamberts SW, Hop WC, de Jong FH, Dahl KD. Serum bioactive and immunoreactive luteinizing hormone and folliclestimulating hormone levels in women with cycle abnormalities, with or without polycystic ovarian disease. J Clin Endocrinol Metab 1991;73: van Dessel HJ, Schoot BC, Schipper I, Dahl KD, Fauser BC. Circulating immunoreactive and bioactive follicle stimulating hormone concentrations in anovulatory infertile women and during gonadotrophin induction of ovulation using a decremental dose regimen. Hum Reprod 1996;11: Chappel SC. Heterogeneity of follicle stimulating hormone: control and physiological function. Hum Reprod Update 1995;1: Wide L, Naessen T, Eriksson K, Rune C. Time-related effects of a progestogen on the isoforms of serum gonadotropins in 17- -estradiol treated postmenopausal women. Clin Endocrinol 1996;44: Connolly PB, Handa RJ, Resko JA. Progesterone modulation of androgen receptors in the brain and pituitary of male guinea pigs. Endocrinology 1988;122: Botella J, Paris J, Lahlou B. The cellular mechanism of the antiandrogenic action of nomegestrol acetate, a new 19-nor progestogen on the rat prostate. Acta Endocrinologica 1987;115: Wright F, Kirchhoffer MO, Giacomini M. Natural anti-androgen progesterone. Ann Endocrinol (Paris) 1979;40(3 Suppl): Ying S. Inhibins, activins and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 1988;9: Anderson RA, Groome NP, Baird DT. Inhibin A and inhibin B in women with polycystic ovarian syndrome during treatment with FSH to induce mono-ovulation. Clin Endocrinol 1998;48: Lee JS, Manning JM, Foster DL, Padmanabhan V. Estrogen increases the secretion of less-acidic FSH isoforms in ovariectomized prepubertal and peripubertal lambs. In: Program and Abstracts of the 80th Annual Meeting of the Endocrine Society, New Orleans, Louisiana. 1998:348. [Abstract] 28. Damian-Matsumura P. Zaga V. Maldonado A. Sanchez-Hernandez C. Timossi C. Ulloa-Aguirre A. Oestrogens regulate pituitary alpha2,3- sialyltransferase messenger ribonucleic acid levels in the female rat. J Mol Endocrinol 1999;23: Padmanabhan et al. Bioactive FSH in PCOS Vol. 75, No. 5, May 2001