| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Reproductive Endocrine Unit, Department of Medicine, Massachusetts General Hospital (C.K.W., A.E.T., J.M.A., A.L.S.), Boston, Massachusetts 02114; Department of Obstetrics and Gynecology, Brigham and Women’s Hospital (J.F.), Boston, Massachusetts 02115; and Department of Pathology and Laboratory Medicine, Women and Infants Hospital (G.M.M.), Providence, Rhode Island 02903
Address all correspondence and requests for reprints to: Dr. Corrine K. Welt, Reproductive Endocrine, BHX 511, Massachusetts General Hospital, 55 Fruit Street, Boston, Massachusetts 02114. E-mail: cwelt{at}partners.org.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of this study was to test the hypothesis that inhibin A and inhibin B protein concentrations are also decreased in PCOS follicles.
Design: The design was a parallel study.
Setting: The study was performed at an in vitro fertilization suite.
Participants: We studied women with regular cycles (n = 36) and women with PCOS (n = 8).
Interventions: Follicular fluid was aspirated from the follicles of women with PCOS (n = 14 follicles) and from women with regular cycles at various times during the follicular phase (n = 50 follicles).
Main Outcome Measure: Inhibin A and B concentrations from PCOS follicles were compared with those in size-matched follicles, dominant follicles (
10 mm), and subordinate follicles from regularly cycling women.
Results: Inhibin A (220 ± 38 vs. 400 ± 72 IU/ml; P < 0.05) and inhibin B (75.4 ± 10.4 vs. 139 ± 26 ng/ml; P < 0.05) concentrations were lower in the follicular fluid of PCOS follicles compared with those of size-matched follicles from regularly cycling women. Inhibin A was also lower in the follicular fluid of PCOS compared with subordinate follicles from normal women (577 ± 166 IU/ml; P < 0.05). Inhibin A concentrations increased with increasing follicle size, resulting in significantly higher follicular fluid concentrations in dominant follicles from normal women compared with PCOS follicles (2298 ± 228 IU/ml; P < 0.05).
Conclusions: These data demonstrate that inhibin A and inhibin B concentrations are significantly reduced in the follicular fluid of women with PCOS compared with those in the follicular fluid of size-matched follicles from normal women, consistent with the decreased inhibin subunit mRNA expression in previous studies. These findings point to the potential importance of inhibins in normal follicle development and suggest that inhibin deficiency may play a role in the follicle arrest associated with PCOS.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In normal women, cyclic follicle development is dependent on circulating FSH and LH levels as well as locally produced growth factors and hormones (7, 8). Because FSH levels are normal and should be sufficient to permit ovulatory follicle development in women with PCOS (9), and PCOS granulosa cells respond robustly to FSH in culture (6), investigators have begun to assess whether the locally produced ovarian products could contribute to the follicular arrest. Indeed, previous studies have suggested that ovarian inhibin, activin, and/or follistatin (FS) have roles in modulating normal follicle development (10), raising the possibility that disorders of synthesis or secretion of these proteins could contribute to abnormal follicle development in PCOS.
We previously demonstrated that in follicular fluid from normal women, linear increases in follicular inhibin A and estradiol (E2) concentrations characterize antral follicle development (11). Moreover, inhibin A was detectable in follicles as small as 6.5 mm, before the reported time of aromatase induction and the consequent increase in E2 (11). This observation suggests that the production of inhibin A might be important for normal follicular development even before follicles are selected and dominance established. In support of this concept, we recently found that granulosa cell inhibin 
- and ßA-subunit mRNA levels were lower in arrested follicles from women with PCOS compared with size-matched normal follicles (12). In the present study we hypothesized that intrafollicular inhibin and/or activin protein concentrations would be deficient in arrested follicles from women with PCOS compared with those in size-matched follicles from regularly cycling women, alluding to the importance of inhibins in normal follicle development. To test this hypothesis, sex steroids, inhibin A, inhibin B, activin A, and FS were measured in follicular fluid aspirated transvaginally from women with PCOS and from well-characterized normal women at various times during the follicular phase. The findings demonstrate that inhibin A and B concentrations are deficient in PCOS follicular fluid compared with those in follicular fluid from size-matched normal follicles, supporting the concept that these hormones are important for normal follicle development and selection.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All subjects were between 18 and 42 yr of age, had been taking no medications for at least 3 months, were in generally good health, had no history of pelvic adhesions, and had normal TSH and prolactin levels, platelet counts, and bleeding times. The Partners human research committee at Massachusetts General and Brigham and Women’s Hospitals approved the protocol, and all subjects gave written informed consent.
Normal women (n = 36) had a history of regular, 25- to 35-d menstrual cycles and had normal physical examinations. Results from a subset of these subjects (n = 26) were reported previously (12).
Women with PCOS (n = 12) met the following criteria: 1) fewer than nine menses per year and 2) biochemical (elevated testosterone, androstenedione, or dehydroepiandrosterone sulfate) and/or clinical [hirsutism (Ferriman-Gallwey score, >8) or acne] evidence of hyperandrogenism. A subset of these subjects (n = 4) was previously reported (12).
Protocol
Daily blood samples were obtained from the normal women across two menstrual cycles, beginning with menses in cycle 1 and continuing until the day of transvaginal follicle aspiration in cycle 2, as previously reported (11). Transvaginal pelvic ultrasounds were performed every 2–3 d in the follicular phase of both cycles in normal subjects to document follicle size and growth. In normal subjects, follicles were aspirated at different times spanning the follicular phase to obtain a broad range of follicle sizes. Women with PCOS began the study without regard to their previous menstrual periods. PCOS subjects provided daily blood samples, underwent at least two ultrasounds over a 10-d period to document the absence of dominant follicle growth, then underwent transvaginal follicle aspiration. In all subjects, the maximum follicle diameter was recorded. Daily blood samples were analyzed for LH, FSH, E2, progesterone (P4), androstenedione, testosterone, inhibin A, and inhibin B to determine the timing of ovulation in normal women, the absence of follicle development in PCOS women, and the associated androgen changes in both groups.
Follicle aspirations were performed in the in vitro fertilization suite at Brigham and Women’s Hospital. Subjects were offered an oral sedative 1 h before the procedure, but no other general or local anesthesia was used. After sterile preparation, the largest follicle on either ovary, referred to herein as the lead follicle in normal subjects, was measured as the average of two cross-sectional diameters. The follicle was aspirated, and the contents of the sterile needle and tubing were flushed with 2.0 ml Dulbecco’s PBS containing 20 IU heparin/ml (Sage Biopharma, Bedminster, NJ) (11). The difference between the recovered volume and the 2-ml flush volume was considered to be the follicle volume, and a dilution factor was calculated to correct later hormone analyses.
In normal subjects, follicular fluid was obtained from a total of 36 follicles in 41 attempts at follicle aspiration in 36 subjects. Five normal subjects underwent follicle aspirations on two separate occasions. Follicular fluid was not obtained from one subject, who was in the luteal phase based on an increased serum P4 level in retrospective analysis; from two subjects in whom only blood was obtained; and in three subjects in whom follicular fluid was determined to be inadequate based on the absence of E2 in the final sample. A second, smaller follicle was aspirated from the same ovary as the first follicle in 15 normal subjects (total of 50 follicular fluid samples). These follicles are referred to as subordinate if the follicle was less than 11 mm (n = 9) and the first follicle aspirated was more than 11 mm to indicate that they were not the dominant, selected follicle. If both the first and second follicles aspirated were less than 11 mm (n = 6), both were counted as lead follicles in the analysis, because it was not known whether follicle selection had taken place and which one was the dominant follicle. In PCOS subjects, follicular fluid was obtained from a total of 14 follicles in 12 attempts at follicle aspiration in 12 subjects. Five PCOS subjects had two (n = 4) or three (n = 1) follicles aspirated during the procedure, with each treated as an individual sample. No follicular fluid was obtained from four subjects: one who decided not to go through with the aspiration in the operating room, one in whom only blood was obtained, and two in whom follicular fluid was determined to be inadequate based on the absence of E2 in the final sample. Follicular fluid aspirates were analyzed for E2, P4, androstenedione, inhibin A, inhibin B, activin A, and free FS.
Assays
LH, FSH, and E2 in serum and E2 in follicular fluid were analyzed using a two-site monoclonal nonisotopic system (Axsym, Abbott Laboratories, Inc., Abbott Park, IL), and P4 was analyzed in serum and follicular fluid using a sequential competitive immunoassay (Immulite, Diagnostic Products Corp., Los Angeles, CA) (13, 14). Gonadotropin levels are expressed as international units per liter, as equivalents of the Second International Reference Preparation 71/223 of human menopausal gonadotropins. For LH, the interassay coefficients of variation (CVs) were 5.3, 5.5, and 7.4% for quality control sera containing 5.6, 26.2, and 69.0 IU/liter, respectively. For FSH, the interassay CVs were 6.9, 7.1, and 6.3% for quality control sera containing 4.3, 35.4, and 79.5 IU/liter, respectively. For E2, the interassay CVs were 10.2, 6.5, and 8.2% for quality control sera containing 81, 284, and 683 pg/ml (297, 1042 and 2507 pmol/liter), respectively. For P4, the interassay CVs were 14.4, 10.6, and 10.8% for quality control sera containing 1.5, 3.2, and 14.3 ng/ml (4.8, 10.2, and 45.5 nmol/liter), respectively. Total testosterone in serum and androstenedione in serum and follicular fluid were measured by RIA (Coat-a-Count, Diagnostic Products Corp.), as previously described (15, 16). For total testosterone, the interassay CVs were 7.3, 8.1, and 6.0% for quality control sera containing 188, 454, and 948 ng/dl (8.4, 20.3, and 42.3 nmol/liter). For androstenedione, the interassay CVs were 8.7 and 6.2% for quality control sera containing 1.04 and 4.80 ng/ml (3.63 and 16.8 nmol/liter).
Inhibin A, inhibin B, and activin A in serum and follicular fluid were measured by ELISA (Serotec, Oxford, UK), as previously described (17). The inhibin A assay uses a lyophilized human follicular fluid calibrator standardized as equivalents of the World Health Organization recombinant human inhibin A preparation 91/624, and values are reported as international units per milliliter. The limit of detection for the dimeric inhibin A assay was 0.6 IU/ml, and the interassay CVs were 11.7 and 10.3% for quality control sera containing 5.01 and 10.01 IU/ml, respectively. For inhibin B, the limit of detection of the assay was 15.6 pg/ml, the intraassay CVs were 4–6%, and the interassay CVs were 15–18% for quality control sera containing 121, 250, and 723 pg/ml, respectively. For activin A, the intra- and interassay CVs were less than 7%, and the assay sensitivity was 80 pg/ml.
Free FS in follicular fluid was determined using a two-monoclonal solid-phase immunochemiluminescent assay immunoassay in which FS bound to activin was undetectable (18). The limit of detection was 2 ng/ml; the intra- and interassay CVs were less than 6% and less than 17% respectively. The FS assay standard was recombinant FS288 obtained from the National Hormone and Pituitary Program.
All samples were analyzed in duplicate, except for inhibin B, which was analyzed once, and all samples from an individual were analyzed in the same assay.
Statistical analysis
All follicular fluid hormone concentrations were tested for normal distribution. The concentrations of E2, P4, and the androstenedione/E2 ratio (A:E ratio) in follicular fluid were not normally distributed and were log-transformed for analysis.
The follicular fluid hormone levels in PCOS follicles were compared with those in normal lead follicles and normal subordinate follicles, with maximum follicle diameters within the same range and with the same mean (Table 2
) as those in PCOS follicles (4–8.5 mm). Follicular fluid hormone concentrations in PCOS follicles were compared with levels in these normal, size-matched, lead (range,4.5–8.5 mm) and subordinate follicles (range, 6–9.5 mm) and were compared with concentrations in normal, lead, dominant follicles (10 mm; range, 10.5–23.5 mm) (19) using two-tailed t tests. The three normal follicle groups were mutually exclusive. To adjust for multiple testing, a value of P < 0.01 was considered significant.
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. As expected, women with PCOS had higher body mass index (BMI), Ferriman-Gallwey score, LH, LH/FSH ratio, and androstenedione level compared with normal subjects.
|
As previously described in a smaller subset of normal subjects (11), follicular fluid E2 (r = 0.794; P < 0.001), P4 (r = 0.761; P < 0.001), and inhibin A concentrations (r = 0.798; P < 0.001) increased with increasing follicle size, whereas the A:E ratio (r = –0.702; P < 0.001) decreased (Table 2
). There was no relationship between follicular fluid inhibin B, activin A, and free FS concentrations and follicle size. In follicular fluid from subordinate follicles (the smaller follicles aspirated from the same ovary as the lead follicle), the A:E ratio decreased (r = –0.855; P = 0.006) with increasing follicle size. There was no relationship between any hormone concentration and follicle size in PCOS follicles.
Follicular fluid hormone concentrations in PCOS subjects were compared with concentrations in normal 1) size-matched, lead follicles (i.e. largest follicle aspirated from any ovary or both follicles if two follicles <11 mm were aspirated; n = 13); 2) size-matched, subordinate follicles (i.e. second, smaller follicle aspirated from the same ovary as a follicle 11 mm; n = 7); and 3) dominant, lead follicles (10 mm; n = 23; Table 2). Compared with normal, size-matched, lead follicles, PCOS follicles had 40–50% lower P4, inhibin A, and inhibin B concentrations (Table 2 and Fig. 1), and these differences were independent of BMI. No differences in E2, androstenedione, activin A, or free FS concentrations or the A:E ratio were observed (Table 2). Compared with normal, size-matched, subordinate follicles, in which E2, P4, and inhibin A concentrations were lower and the A:E ratio higher than in the dominant follicle from the same ovary (11), PCOS follicles had 50% lower P4 and inhibin A concentrations and a 2-fold increase in androstenedione (Table 2 and Fig. 1). As expected, compared with dominant, lead follicles, PCOS follicles had lower E2, inhibin A, inhibin B, and P4 concentrations and a higher androstenedione concentration and A:E ratio, but no difference in activin A or FS concentrations (Table 2 and Fig. 1). These relationships were independent of BMI.
|
In PCOS subjects, serum LH, E2, androstenedione, and testosterone concentrations remained stable for 10 d before follicle aspiration (Fig. 2). The FSH concentration increased over the 10 d, with the level on d 4 significantly higher than that on d 1. Serum LH, FSH, and androstenedione concentrations were higher, and E2 levels were lower in PCOS subjects compared with normal subjects over 10 d in the follicular phase.
|
). The FSH concentration changed significantly in PCOS subjects over the 10 d, with no change in ovarian size, E2, inhibin A, or inhibin B concentrations. Serum and follicular fluid E2 concentrations were correlated (r = 0.744; P < 0.001), as were serum and follicular fluid inhibin A concentrations (r = 0.490; P = 0.008) in normal subjects. Serum FSH correlated inversely with follicular fluid E2 (r = –0.639; P < 0.001), inhibin A (r = –0.627; P < 0.001), and P4 (r = –0.629; P < 0.001) in normal subjects. There were no additional relationships between serum and follicular fluid hormone concentrations in normal subjects, and there were no significant relationships for these parameters in PCOS subjects.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
- and ßA-subunit mRNA expression is decreased in granulosa cells from the follicles of PCOS subjects and indicate that there is an overall deficiency in inhibin A and B biosynthesis. Indeed, the relative follicular fluid inhibin A deficiency in PCOS compared with normal follicles is not rescued by treatment with exogenous FSH stimulation during ovarian hyperstimulation (15). Whether the deficiency in these factors is mechanistically related to follicle arrest in PCOS or is secondary to an as yet unknown overriding defect remains to be determined. Nevertheless, it is clear that reduced inhibin A, inhibin B, and P4 concentrations are characteristic of arrested follicles in PCOS. Although the defect in inhibin biosynthesis in the follicular fluid of women with PCOS is consistent with mRNA studies, a previous study of the follicular fluid from follicles of women with PCOS failed to demonstrate a difference in inhibin A and inhibin B levels in size-matched PCOS and normal follicles despite follicular fluid inhibin A and B concentrations similar to those in the current study (21). In this study, follicular fluid was collected from a single surgical visit in women given anesthesia, which can alter intrafollicular hormone concentrations (22). In contrast, in the current study, PCOS and normal follicular fluids were obtained from the largest follicle in well-characterized women followed serially during normal follicle development and may therefore better represent physiological hormone levels during follicle development. Perhaps as a result of the different collection methods, the A:E ratio was greater than 4 in all PCOS follicles and in the majority of size-matched normal follicles in the previous study (21), whereas in the current study, two of 14 PCOS follicles and six of 13 normal, size-matched, lead follicles had an A:E ratio less than 4, a difference that could account for the discrepant inhibin results. These data also suggest that aromatase turns on even sooner than at an 8- to 10-mm follicle size, as suggested by previous studies of normal follicle development (11, 23).
Inhibin A and inhibin B are regulated differently in vitro, with FSH stimulating inhibin A, but not inhibin B, secretion from granulosa cells in culture (24). Therefore, deficient follicular fluid inhibin A and inhibin B protein concentrations argue against a defect in FSH stimulation as the etiology. It is possible that the increased serum LH concentrations (9) play a role in inhibin deficiency in women with PCOS. LH stimulates granulosa cell E2 and P4 secretion earlier in follicle development in PCOS compared with normal follicles (25). Therefore, the high LH concentrations could cause premature luteinization and differentiation, which would inhibit proliferation, resulting in a smaller number of granulosa cells in PCOS follicles and lower inhibin A, inhibin B, and P4 concentrations in PCOS follicular fluid. A factor that regulates -subunit mRNA expression could also be responsible for the decreased inhibin A and inhibin B concentrations, because -subunit expression is reduced in PCOS granulosa cells (12, 20).
Deficient inhibin production could have functional consequences for antral follicle maturation (reviewed in Ref.26). Inhibin signals via a type II activin receptor and type III TGFß receptor (betaglycan) complex (27), and both have been identified on granulosa and thecal cells (28) of growing antral follicles. In rodent models, administration of inhibin to adult ovaries accelerated antral follicle development, whereas administration of activin increased follicular atresia (29). Furthermore, when the inhibin ßB-subunit gene was inserted into the ßA locus to rescue the lethal outcome of the ßA-subunit knockout mouse, females had smaller litter sizes and a reduced number of large, preovulatory follicles, suggesting that inhibin A and/or activin A play a role in the final stages of follicle maturation (30). Taken together, these studies suggest that defective inhibin biosynthesis may be functionally related to follicular arrest in PCOS.
Serum inhibin B is a marker of the number of developing follicles and is secreted starting in the primordial follicle stage (10). Therefore, serum inhibin B was universally expected to be higher in PCOS subjects compared with normal subjects, because the total number of follicles, from the primary to the Graafian stage, is increased in PCOS compared with normal ovaries (31). However, the current study is consistent with the majority of previous studies, demonstrating that serum inhibin B and A concentrations are similar despite the increased number of follicles in women with PCOS compared with normal women (32, 33, 34, 35, 36, 37, 38). The current study provides evidence that follicles in women with PCOS are deficient in inhibin A and inhibin B production, accounting for the similar serum inhibin A and B concentrations in PCOS and normal women.
The serum and follicular fluid E2 and inhibin A concentrations correlated with each other, but follicular fluid E2 levels were approximately 15,000 times greater, and inhibin A levels were approximately 350 times greater than serum levels, as described previously (11). These findings suggest that follicular fluid inhibin and E2 contribute to serum concentrations, but that the contribution is dependent on individual hormone transport or hormonal structure, i.e. steroid vs. protein. Interestingly, serum FSH levels correlated inversely with follicular fluid E2, P4, and inhibin A concentrations. This inverse correlation presumably reflects the close correlation between serum and follicular fluid E2 and inhibin A concentrations. Finally, it is interesting that serum FSH levels increased over 10 d in subjects with PCOS, a change that has not been observed in previous studies (39) and is not easily explained. This change was not related to recent ovulation, because an increase in FSH was seen in subjects who had not had menses for more than 2 months. It was also not biologically relevant, because follicle size did not increase across the 10 d in PCOS subjects.
Through careful examination of follicular fluid aspirated from women with PCOS and women with regular menstrual cycles, we have demonstrated a selective decrease in inhibin A and inhibin B protein concentrations in the follicular fluid of women with PCOS. These findings add to the accumulating evidence for a selective deficiency of inhibin A and inhibin B biosynthesis in the arrested follicles of women with PCOS. This deficient inhibin biosynthesis in PCOS follicles points to a potential mechanism for follicular arrest in women with PCOS and emphasizes the important roles of inhibin A and inhibin B in normal follicle development.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Current address for A.E.T.: Pfizer Global Research and Development, Groton Laboratories, Groton, Connecticut 06340.
First Published Online July 19, 2005
Abbreviations: A:E ratio, Androstenedione/estradiol ratio; BMI, body mass index; CV, coefficient of variation; E2, estradiol; FS, follistatin; P4, progesterone; PCOS, polycystic ovary syndrome.
Received March 30, 2005.
Accepted July 11, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Antenos, M. Stemler, I. Boime, and T. K. Woodruff N-Linked Oligosaccharides Direct the Differential Assembly and Secretion of Inhibin {alpha}- and {beta}A-Subunit Dimers Mol. Endocrinol., July 1, 2007; 21(7): 1670 - 1684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Dewailly, S. Catteau-Jonard, A.-C. Reyss, C. Maunoury-Lefebvre, E. Poncelet, and P. Pigny The excess in 2-5 mm follicles seen at ovarian ultrasonography is tightly associated to the follicular arrest of the polycystic ovary syndrome Hum. Reprod., June 1, 2007; 22(6): 1562 - 1566. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. E. Messinis Ovarian feedback, mechanism of action and possible clinical implications Hum. Reprod. Update, September 1, 2006; 12(5): 557 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Wachs, M. S. Coffler, P. J. Malcom, and R. J. Chang Comparison of Follicle-Stimulating-Hormone-Stimulated Dimeric Inhibin and Estradiol Responses as Indicators of Granulosa Cell Function in Polycystic Ovary Syndrome and Normal Women J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 2920 - 2925. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Farnworth, P. G. Stanton, Y. Wang, R. Escalona, J. K. Findlay, and G. T. Ooi Inhibins Differentially Antagonize Activin and Bone Morphogenetic Protein Action in a Mouse Adrenocortical Cell Line Endocrinology, July 1, 2006; 147(7): 3462 - 3471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. Foong, D. H. Abbott, M. A. Zschunke, T. G. Lesnick, J. L. Phy, and D. A. Dumesic Follicle Luteinization in Hyperandrogenic Follicles of Polycystic Ovary Syndrome Patients Undergoing Gonadotropin Therapy for in Vitro Fertilization J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2327 - 2333. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
