| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Articles: Hormones and Reproductive Health |
Reproductive Endocrine Unit, Reproductive Endocrine Sciences Center, and National Center for Infertility Research, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Dr. Corrine K. Welt, Reproductive Endocrine Unit, BHX 511, Fruit Street, Massachusetts General Hospital, Boston, Massachusetts 02114.
Abstract
Inhibin B and inhibin A exhibit unique patterns of secretion across the follicular phase of the menstrual cycle. To test the hypothesis that the distinct patterns of inhibin B and inhibin A secretion result from differential regulation by LH and FSH, a series of controlled experiments was designed to dissect the specific effects of LH and FSH at distinct stages of follicle development. After GnRH agonist desensitization, women with small antral follicles were treated with recombinant human LH (rhLH), rhFSH, or rhFSH and estradiol (E2). rhLH or rhFSH was also administered when follicles reached the preovulatory stage in gonadotropin-stimulated or spontaneous cycles.
At the small antral stage of development, rhFSH, but not rhLH, administration increased inhibin B (17.4 ± 4.6 to 321.0 ± 97.0 pg/mL; P < 0.05), inhibin A (0.6 ± 0.1 to 2.6 ± 0.6 IU/mL; P < 0.05), and E2 [15.8 ± 3.6 to 95.3 ± 26.9 pg/mL (58.0 ± 13.2 to 349.8 ± 98.7 pmol/L); P < 0.05]. The inhibin B increase preceded inhibin A by 48 h. Addition of E2 to FSH resulted in a greater increase in inhibin B (23.2 ± 6.4 to 865.2 ± 294.5 pg/mL; P < 0.05) than FSH alone (P < 0.05). At the preovulatory stage, rhLH administration increased inhibin A (15.9 ± 10.3 to 21.5 ± 13.7 IU/mL; P < 0.05) and E2 [669.4 ± 285.5 to 943.6 ± 388.1 pg/mL (2457.4 ± 1048.1 to 3464.0 ± 1424.7 pmol/L); P < 0.05], but not inhibin B, as did rhFSH administration in spontaneous cycles [E2: 226.4 ± 102.7 to 264.7 ± 121.0 pg/mL (831.1 ± 377.0 to 971.7 ± 444.2 pmol/L); P < 0.05; inhibin A: 2.6 ± 1.3 to 3.7 ± 1.9 IU/mL; P < 0.05; and inhibin B: 76.3 ± 32.2 to 77.6 ± 32.8 pg/mL; P = NS].
These findings suggest that increases in both FSH and E2 in the early follicular phase result in increased inhibin B secretion at early stages of follicle development, whereas the selective LH rise in the late follicular phase favors inhibin A secretion from more mature follicles. Thus, both differential secretion of LH and FSH and the stage of follicle development determine the patterns of inhibin A and inhibin B secretion in the normal menstrual cycle.
OVARIAN FOLLICLE growth is a complex process involving changes in gonadotropin responsiveness (1, 2). Although initiation of primordial follicle growth is largely FSH independent, recruitment of the cohort of antral follicles, which begins in the late luteal phase of the menstrual cycle, requires FSH acting through its receptors on granulosa cells. The subsequent maturation of the dominant follicle requires acquisition of LH receptors on granulosa cells, which impart the ability to ovulate in response to the midcycle LH surge. Thus, follicular maturation is accompanied by a progressive ability of the follicle to respond to FSH at the small antral stage and to both LH and FSH at the preovulatory stage.
The two inhibins, heterodimers composed of an 
-subunit and one of
two ß-subunits forming inhibin A (ßA) and inhibin B (ßB),
also exhibit unique patterns of expression and secretion during the
menstrual cycle. In vitro studies demonstrate that inhibin
ßB-subunit messenger ribonucleic acid (mRNA) is expressed
predominantly in small antral follicles, and ßA mRNA is expressed in
the preovulatory follicle of the nonhuman primate (3) and
human (4), whereas -subunit mRNA expression is similar
at both follicle stages in the human (4). Therefore, small
antral follicles have the potential to secrete inhibin B, whereas
preovulatory follicles may secrete inhibin A. Consistent with this
hypothesis, inhibin B rises across the luteal-follicular transition and
reaches its initial peak in the midfollicular phase coincident with
growth of the cohort of recruited follicles (5, 6, 7). In
contrast, inhibin A increases in the late follicular phase concomitant
with the selection and maturation of the preovulatory follicle
(7, 8, 9).
The patterns of inhibin B and inhibin A secretion in the follicular
phase mirror those of FSH and LH, respectively, suggesting that FSH and
LH may differentially regulate inhibin B and inhibin A secretion from
small antral and preovulatory follicles. In vitro studies in
the rat (10, 11, 12, 13), nonhuman primate (14), and
human (15) indicate that FSH stimulates inhibin secretion
from granulosa cells of immature or small antral follicles, and both
FSH and LH stimulate inhibin secretion from granulosa cells of mature
or preovulatory follicles. However, these studies used RIAs directed at
the -subunit of inhibin (16) and therefore did not
distinguish between inhibin A and inhibin B. Recent human studies
attempting to dissect the specific regulation of inhibin A and inhibin
B demonstrated that both FSH and LH stimulate inhibin A and inhibin B
secretion from luteinized granulosa cells (17). However,
because these granulosa cells were stimulated with exogenous
gonadotropins and hCG for in vitro fertilization and were
well on the pathway to luteinization, their activities may not be
representative of granulosa cells from normal follicles at earlier
stages of development. The only human study examining dimeric inhibin
regulation by FSH in vivo in subjects not previously
stimulated by exogenous gonadotropins suggested that FSH stimulates
both inhibin A and inhibin B (18). However, the size and
developmental stage of the follicles were not determined in this study,
nor was regulation by LH examined.
We hypothesized that the distinct regulation of inhibin A and inhibin B across the follicular phase of the menstrual cycle would depend on both the stage of follicle development and the predominance of either FSH or LH at that cycle stage. Specifically, we hypothesized that FSH stimulates inhibin B, but not inhibin A, secretion from small antral follicles during the early follicular phase, whereas both FSH and LH stimulate inhibin A, but not inhibin B, secretion from preovulatory follicles in the late follicular phase. To test these hypotheses, a series of controlled experiments was designed to dissect the specific effects of LH and FSH at the small antral and preovulatory follicle stage. Endogenous gonadotropin secretion was suppressed using a GnRH agonist to examine the independent effects of recombinant human LH (rhLH) or rhFSH administration on inhibin A and inhibin B secretion in women across all stages of small antral follicle development. Additional subjects were treated with rhFSH and estradiol (E2) to ensure that E2 levels were in the physiological range in GnRH agonist-treated subjects given rhFSH, alone. In a second group of GnRH agonist-treated subjects, a defined regimen of rhLH and rhFSH was used to induce development of a preovulatory follicle, at which time either rhLH or rhFSH was administered, and the inhibin A and inhibin B responses were measured. rhFSH was also administered to subjects with a preovulatory follicle in a spontaneous cycle in the absence of gonadotropin-induced multiple follicle development. Our results provide evidence that both gonadotropin stimulation and stage of follicle development contribute to the differential regulation of inhibin B and inhibin A in the follicular phase of the menstrual cycle.
Subjects and Methods
Experimental subjects
The study population consisted of 29 healthy subjects, aged 19–40 yr (mean ± SD, 28.2 ± 6.5 yr), with normal thyroid hormone and PRL levels, normal body mass index (mean ± SD, 23.4 ± 3.3 kg/m2), regular menstrual cycles (25–35 days in length), and ovulation in the cycle preceding the start of the study confirmed by a progesterone level greater than 3.5 ng/mL (11 nmol/L). Two subjects underwent 2 studies each, 1 spontaneous cycle and 1 with GnRH agonist treatment, for a total of 31 studies. Subjects had been taking no hormonal medications for at least 3 months and used barrier contraception throughout the study. Subjects were recruited using e-mail, posted, and newspaper advertisements and were remunerated for their time and effort. The study was approved by the subcommittee on human studies at the Massachusetts General Hospital, and all subjects gave written informed consent.
GnRH agonist pretreatment
Subjects in the first three series of studies (n = 26) received a GnRH agonist (3.75 mg Decapeptyl Depot, Ferring Pharmaceuticals Ltd., Kiel, Germany) 1–3 days after the onset of menses. Two weeks later, subjects underwent a GnRH stimulation test, 100 µg GnRH, iv (Factrel, Wyeth-Ayerst Laboratories, Inc., Philadelphia, PA), to document gonadotropin suppression. The GnRH stimulation test was not performed in the final eight subjects studied due to unavailability of GnRH; however, baseline LH was 6 IU/L or less and FSH was 4 IU/L or less in these subjects. After completion of the GnRH stimulation test, subjects were randomly assigned to groups examining small antral follicles or preovulatory follicles.
Small antral follicles
LH or FSH administration. To analyze regulation of dimeric inhibin secretion across all stages of small antral follicle development, rhLH (Luveris, Serono Laboratories, Inc., Norwell, MA; 75 IU, sc, twice daily; n = 3) or rhFSH (Gonal-F, Serono Laboratories, Inc.; 150 IU, sc, daily; n = 6) was administered starting the day after documented gonadotropin suppression. The dose of rhLH has previously been demonstrated to increase E2 levels during ovulation induction in women with hypogonadotropic hypogonadism (19). Treatment was continued until a dominant follicle developed (maximum follicle diameter of at least 10 mm) as documented by ultrasound. In subjects treated with rhLH, there was no change in follicle size, and hormone administration was discontinued after 7 days. Transvaginal ultrasounds (Toshiba SAL 77B) were performed at baseline and on days 5–6 of treatment. The growth rate of the largest follicle was estimated to be 2 mm/day, and subjects returned for a final ultrasound when their largest follicle was expected to be at least 10 mm. All follicles 9 mm or larger were counted. Blood samples were drawn daily before gonadotropin administration for measurement of FSH, LH, E2, inhibin A, and inhibin B.
FSH and E2 administration. To ensure that E2 levels were in the physiological range in GnRH agonist-treated subjects given rhFSH alone, an additional group of subjects was treated with rhFSH and E2. E2 (2 mg, orally, daily; Estrace, Bristol-Myers Squibb Co., Princeton, NJ) was administered for 2 days, at which time rhFSH (150 IU, sc) was added, and both were administered daily until follicle size reached at least 10 mm (n = 5). Ultrasound and blood sampling were performed as described for LH and FSH administration above.
Preovulatory follicles
LH or FSH administration. To examine dimeric inhibin regulation at the preovulatory stage of follicle development, subjects were treated with a GnRH agonist, then received an identical decreasing dose regimen of exogenous rhLH and rhFSH as previously described (20). Specifically, rhLH (75 IU, sc) and rhFSH (150 IU, sc) were administered daily until at least one follicle reached a diameter of 11 mm as documented by ultrasound, at which time the dose of rhFSH was reduced to 112.5 IU daily, then to 75 IU daily 3 days later. An ultrasound was performed at baseline, on day 7, then every 1–2 days until the maximum follicle diameter reached 16 mm. The maximum diameter and number of all follicles 7–9 mm or larger were measured. Blood samples were drawn daily for measurement of LH, FSH, E2, inhibin A, and inhibin B. When at least one follicle reached 16 mm, rhLH (150 IU; n = 6) or rhFSH (150 IU; n = 6) was administered sc between 0800–0900 h. Blood was drawn at baseline, 2 h, then every 6 h from baseline for 36 h, with additional blood samples at 48, 72, and 96 h. Blood was analyzed for LH, FSH, E2, inhibin A, and inhibin B. A final ultrasound was performed at 96 h to document follicle size and to assess the occurrence of ovulation.
FSH administration in spontaneous cycles. Treatment with exogenous gonadotropins resulted in the growth of more than 1 dominant follicle in 6 of 12 subjects along with multiple small and medium-sized follicles. Therefore, FSH regulation of dimeric inhibin secretion by the preovulatory follicle was studied in a second group of 5 subjects during a spontaneous menstrual cycle to avoid multifollicular development. Two subjects who participated in the GnRH agonist portion of the protocol were also studied during a spontaneous cycle at least 3 months later.
Starting on the day of or the day after menses, blood was drawn daily for measurements of LH, FSH, E2, inhibin A, and inhibin B. Ultrasounds were performed at baseline, on day 7, then every 2–3 days until the maximum follicle diameter reached at least 16 mm. When one follicle reached 16 mm, rhFSH (150 IU) was administered sc between 0800–0900 h. Blood was drawn at baseline, at 2 h, then every 6 h from baseline for 36 h, with additional blood samples at 48, 72, and 96 h. Blood was analyzed for LH, FSH, E2, inhibin A, and inhibin B. A final ultrasound was performed at 96 h to document follicle size and to assess the occurrence of ovulation.
Assays
FSH, LH, and E2 concentrations were determined using a two-site monoclonal nonisotopic system (Abbott Laboratories, Chicago, IL) according to the manufacturer’s directions, as previously described (21). Inhibin A and inhibin B were determined using commercial enzyme-linked immunosorbent assay kits (Serotec, Oxford, UK) according to the manufacturer’s directions, as previously described (7, 9). Inhibin A results are reported as international units per mL WHO International Standard 91/624 (0.15 IU International Standard = 1 pg inhibin A against Serotec calibrators). The limit of detection was 0.6 IU/mL for inhibin A and 15.6 pg/mL for inhibin B, with an intraassay coefficient of variation of 10% and an interassay coefficient of variation of less than 20% for both assays.
Data analysis
To examine gonadotropin regulation of inhibin, the data were log-normalized so that errors were normally distributed, and nonlinear mixed effects models (22) were fit to the observed trajectories of LH, FSH, E2, inhibin A, and inhibin B. A quadratic mean function over time (in days) was assumed for LH, FSH, E2, inhibin A, and inhibin B levels from small follicles. Separate coefficients were assumed for the LH, FSH, and FSH plus E2 experiments, and a random subject intercept was included to model subject-specific deviations from the mean. A piecewise quadratic mean function over time (in hours) with a cut-point at 24 h was assumed for LH, FSH, E2, inhibin A, and inhibin B levels from preovulatory follicles in the GnRH agonist-treated cycles. A quadratic mean function over 24 h was assumed for LH, FSH, E2, inhibin A, and inhibin B in spontaneous cycles to exclude the onset of ovulation. For analyses of all preovulatory follicles (GnRH agonist-treated and spontaneous cycles), separate coefficients were assumed for the LH and FSH experiments, and a random intercept was included to model subject-specific deviations from the mean. Models were fit using a Bayesian analysis implemented via the BUGS software package (23). The Bayesian analysis yields a posterior probability distribution for each parameter in the model.
To assess whether a significant rise in hormone level occurred over the time course of each experiment, the posterior distribution of the difference between the maximum of the estimated mean function and the baseline value on the log scale was examined. A rise was considered significant if a 95% probability interval for the difference between maximum and baseline (an interval that contains the true difference with a probability of 0.95) fell to the right of the null value of zero. This corresponds to a posterior P < 0.05. To compare the effects of rhFSH and rhFSH plus E2 stimulation on small follicles, a 95% probability interval for the difference in the rise between the two experiments was examined and was determined to be significant (P < 0.05) if it did not contain the null difference of zero. Estimated baseline and maximum values are reported on the original untransformed scale as the mean ± SEM.
The follicle number and maximum follicle diameter for all follicles 4 mm or larger after rhLH administration and 9 mm or larger after rhFSH or rhFSH plus E2 administration was compared at baseline and at the completion of the small antral follicle study using paired t test. At the time of final rhLH or rhFSH administration in the preovulatory follicle study, follicle diameter for all follicles more than 9 mm and follicle number were compared 96 h later using paired t test.
Results
Small antral follicles
rhFSH or rhLH administration. rhLH administration resulted in
a significant increase in serum LH levels (baseline to maximum,
6.3 ± 1.9 to 7.5 ± 2.4 IU/L; P < 0.05),
with maximum levels achieved after 4.7 ± 0.7 days. There was no
increase in FSH (4.0 ± 0.7 to 4.8 ± 0.8 IU/L;
P = NS), E2 [14.4 ± 4.3 to
22.3 ± 6.4 pg/mL (52.9 ± 15.8 to 81.9 ± 23.5 pmol/L);
P = NS], inhibin A (0.6 ± 0.2 to 0.7 ± 0.2
IU/mL; P = NS), or inhibin B levels (22.3 ± 8.6
to 27.0 ± 8.9 pg/mL; P = NS) during the 7 days of
rhLH administration (Fig. 1
). Maximum
follicle diameter did not change after rhLH administration (5.0 ±
0 to 4.3 ± 0.3 mm; P = NS).
|
), and no increase in LH (4.0 ±
0.7 to 4.8 ± 0.8 IU/L; P = NS). During rhFSH
administration, there was an increase in E2
[15.8 ± 3.6 to 95.3 ± 26.9 pg/mL (58.0 ± 13.2 to
349.8 ± 98.7 pmol/L); P < 0.05], inhibin A
(0.6 ± 0.1 to 2.6 ± 0.6 IU/mL; P < 0.05),
and inhibin B (17.4 ± 4.6 to 321.0 ± 97.0 pg/mL;
P < 0.05), with maximum levels achieved on the final
day (Figs. 1
and 2A). In contrast to inhibin B, which increased within
24 h of rhFSH administration, inhibin A levels remained
undetectable for 3 days before beginning a steady increase (Fig. 2A, inset), and were first detectable when follicles reached a
diameter of 6–9 mm. Maximum follicle diameter (5.2 ± 0.9 to
11.2 ± 0.2 mm; P < 0.002) and number of
follicles 9 mm or larger (0.2 ± 0.2 to 4.0 ± 1.2 follicles;
P < 0.03) also increased significantly during the
period of rhFSH administration.
|
), whereas there was no increase in LH (4.4 ± 1.0 to 4.4
± 1.0 IU/L; P = NS). There was an increase in
E2 [64.5 ± 16.7 to 205.7 ± 67.0
pg/mL (236.8 ± 61.3 to 755.1 ± 246.0 pmol/L);
P < 0.05], inhibin A (0.7 ± 0.1 to 5.2 ±
1.5 IU/mL; P < 0.05), and inhibin B (23.2 ± 6.4
to 865.2 ± 294.5 pg/mL; P < 0.05) during rhFSH
plus E2 administration, with maximum levels
achieved on the final day of treatment (Figs. 1 and 2B). In the rhFSH
plus E2 group, E2 was
higher at the beginning of rhFSH treatment (after 2 days of
E2 administration) than with rhFSH alone, whereas
inhibin B was not different at baseline in the two groups. However, the
increase in inhibin B was greater after stimulation with rhFSH and
E2 than with rhFSH alone (P <
0.05; Figs. 1 and 2). There was no difference in the LH, FSH,
E2, or inhibin A changes (baseline to maximum)
between rhFSH and rhFSH plus E2 stimulation.
Maximum follicle diameter (4.8 ± 0.4 to 11.0 ± 0.3 mm;
P < 0.001) and number of follicles 9 mm or larger
(0 ± 0 to 4.6 ± 0.9 follicles; P < 0.01)
increased after rhFSH plus E2 administration, but
were not greater than those after rhFSH alone. Preovulatory follicles
rhLH or rhFSH administration. Administration of a single dose
of rhLH in the presence of a preovulatory follicle resulted in a
significant increase in serum LH levels (3.7 ± 0.9 to 4.4 ±
0.9 IU/L; P < 0.05), with maximum levels achieved
after 10.9 ± 1.6 h. FSH levels declined steadily over
96 h (baseline to minimum, 14.0 ± 3.8 to 1.7 ± 1.4
IU/L; P < 0.05), consistent with discontinuation of
rhFSH treatment. The increase in LH was accompanied by an increase in
E2 [669.4 ± 285.5 to 943.6 ± 388.1
pg/mL (2457.4 ± 1048.1 to 3464.0 ± 1424.7 pmol/L);
P < 0.05] at 23.7 ± 2.8 h and inhibin A
(15.9 ± 10.3 to 21.5 ± 13.7 IU/mL; P <
0.05) at 16.3 ± 4.0 h (Fig. 3). Inhibin B did not increase in
response to rhLH administration (1108.0 ± 550.8 to 927.4 ±
467.1 pg/mL; P = NS; Fig. 3). The number of follicles
16 mm or larger (1.7 ± 0.2 to 4.0 ± 0.7 follicles;
P < 0.05) and maximum follicle size (17.7 ± 0.8
to 21.2 ± 1.2 mm; P = 0.05) increased in the
96 h after rhLH administration; however, the number of follicles 9
mm or larger (7.8 ± 1.9 to 9.0 ± 1.9 follicles;
P = NS) did not. There was no evidence of ovulation in
these subjects.
|
). In the 96 h after
rhFSH administration, there was no change in the number of follicles 9
mm or larger (8.5 ± 7.6 to 9.6 ± 3.7 follicles;
P = NS) or 16 mm or larger (2.5 ± 1.3 to 5.0
± 2.8 follicles; P = NS) or in maximum follicle size
(17.5 ± 0.6 to 20.8 ± 2.3 mm; P = NS).
There was also no difference in follicle number or maximum follicle
size after administration of rhFSH compared with rhLH. There was no
evidence of ovulation in these subjects. rhFSH administration in spontaneous cycles. In spontaneous cycles, FSH increased 12 h after rhFSH administration (8.8 ± 3.6 to 11.8 ± 5.0 IU/L; P < 0.05). The increase in FSH was accompanied by increases in E2 [226.4 ± 102.7 to 264.7 ± 121.0 pg/mL (831.1 ± 337.0 to 971.7 ± 444.2 pmol/L); P < 0.05] and inhibin A (2.6 ± 1.3 to 3.7 ± 1.9 IU/mL; P < 0.05), but not inhibin B (76.3 ± 32.2 to 77.6 ± 32.8 pg/mL; P = NS). At the time of rhFSH administration, subjects had only one follicle 16 mm or larger (mean size, 17.6 ± 1.1 mm). Zero to three follicles 9 mm or larger were observed in addition to the largest follicle, with a maximum size of 11 mm in two of five subjects. The LH surge occurred 36–96 h after rhFSH administration in all subjects.
Discussion
The distinct patterns of inhibin A and inhibin B secretion observed across the follicular phase of the menstrual cycle (5, 6, 7, 8, 9) and their relationship to LH and FSH secretion suggest that the regulation of inhibin A and inhibin B secretion is dependent on stage of follicle development and/or differential gonadotropin stimulation. The current study attempted to resolve these effects using a controlled series of experiments that isolated LH and FSH and the distinct follicle stage. The results indicate that both gonadotropin stimulation and the stage of follicle development determine the species of dimeric inhibin secreted.
Suppression of endogenous gonadotropin secretion using a GnRH agonist allowed us to examine dimeric inhibin regulation from small antral follicles in response to physiological doses of either rhLH or rhFSH. Both inhibin A and inhibin B were stimulated by rhFSH administration, whereas rhLH did not stimulate either inhibin, consistent with the absence of LH receptors on follicles at this stage of development (1). Previous studies using varying FSH doses in GnRH agonist-treated infertility patients (24) and two defined FSH doses in normal women in the early follicular phase (18) demonstrated that FSH stimulates both inhibin A and inhibin B. It is clear from the current data, however, that inhibin B secretion precedes that of inhibin A during small antral follicle development. Similarly, FSH stimulation results in preferential inhibin B secretion across the luteal-follicular transition in GnRH-deficient women (6) and a more pronounced inhibin B than inhibin A increase in the early follicular phase of normal women (18). In vitro studies suggest that inhibin ßB-subunit mRNA is more abundant than ßA-subunit mRNA in small antral follicles (4), perhaps accounting for the more robust inhibin B protein secretion. Taken together, these studies indicate that control of inhibin B secretion is dependent on FSH, but not LH, during the early stages of follicular growth.
However, recent in vitro studies suggest that FSH regulation of inhibin B secretion may be more complex than it appears from the in vivo data (25). In vitro, the granulosa cells of small antral follicles secrete inhibin B in the absence of gonadotropin stimulation (25). FSH alone did not stimulate inhibin B secretion from the granulosa cells of these small antral follicles. Indeed, it was only the combination of IGF-I and FSH that stimulated inhibin B secretion, suggesting that IGF-I is critical for inhibin B stimulation by FSH. Even in the presence of IGF-I, however, the increase in inhibin B (<2-fold) was less pronounced than the increase in inhibin A (>4-fold) under these in vitro conditions (25). In contrast to in vitro findings, the current in vivo data demonstrate that FSH stimulation increases serum inhibin B levels almost 20-fold, whereas inhibin A increases only 5-fold. FSH stimulation is known to have a dramatic effect on granulosa cell number and follicle size (1). Thus, it is likely that the increases in the number of granulosa cells and the number of developing follicles induced by FSH stimulation play a more prominent role in increasing inhibin B serum levels than the direct stimulation of inhibin B from granulosa cells, even in the presence of IGF-I.
In contrast to the early increase in inhibin B in response to rhFSH, inhibin A did not reach the limit of detection in all six subjects until the fourth day of rhFSH stimulation in the current study, when the largest follicle was 6–9 mm in diameter. This finding is consistent with previous studies demonstrating that inhibin A is not detectable in serum until relatively late in the process of follicular maturation (7, 21). In contrast to the current study, FSH administration on days 3–5 in the early follicular phase of normal cycling women resulted in a rapid and immediate increase in inhibin A (18), probably related to the more advanced development of the follicles in the absence of gonadotropin suppression in that study. Nonetheless, inhibin A protein is detectable in the follicular fluid of follicles as small as 6 mm (26, 27), and inhibin A is secreted from granulosa cells of follicles as small as 2–4 mm (25). Taken together, these findings point to the requirement for a critical mass of granulosa cells or follicles before inhibin A is detectable in serum.
The combination of E2 and rhFSH resulted in a
greater increase in inhibin B from small antral follicles than did
rhFSH administration alone. This effect was not related to an increase
in the number or size of the follicles detectable on ultrasound,
although granulosa cell proliferation or an increase in the number of
undetectable follicles cannot be ruled out. Further, the effect
appeared to be specific to small antral follicles, as neither rhLH nor
rhFSH stimulated inhibin B secretion from preovulatory follicles
despite high E2 levels. Previous studies in rat
granulosa cells demonstrate that E2 alone
increased inhibin A, with only a minor effect on inhibin B secretion
(28). E2 augmented FSH-stimulated
inhibin secretion in nonhuman primate granulosa cells from immature
follicles (14); however, only testosterone and
nonaromatizable 5-dihydrotestosterone enhanced FSH-stimulated
inhibin secretion in human granulosa cells from immature follicles
(15). The effect of E2 alone on
dimeric inhibin secretion in the human has not been evaluated in
vitro. Although current in vivo studies suggest that
E2 enhances inhibin B secretion from small antral
follicles, further studies will be needed to confirm this effect and to
determine the mechanism by which E2 augments
inhibin B secretion at early stages of follicle development.
In preovulatory follicles, rhLH stimulated inhibin A secretion exclusively. The inhibin A increase was seen despite an increase in LH of only 0.7 IU/L. We previously found a strong correlation between inhibin A and LH during recovery after administration of a GnRH antagonist in the late follicular phase (21) and hypothesized that LH is the primary factor stimulating inhibin A in the preovulatory follicle. Similarly, LH and hCG consistently stimulated inhibin A, but not inhibin B, secretion in vitro from luteinized granulosa cells from women undergoing in vitro fertilization (17). The absence of an associated inhibin B increase after rhLH stimulation in the preovulatory follicle may be due to absent (4) or low levels (3) of inhibin ßB-subunit mRNA in dominant follicles. Alternatively, the different mechanisms by which inhibin A and inhibin B are regulated may play a role. Recent studies in human granulosa cells demonstrate that cAMP stimulates inhibin A, but not inhibin B, secretion from granulosa cells in vitro (25), and LH signals intracellularly via cAMP. Taken together with the parallel rise in LH and inhibin A levels during the late follicular phase, these findings support the hypothesis that LH directly controls the pattern of inhibin A secretion in mature follicles.
FSH stimulation is more complex. Although rhFSH predominantly
stimulated inhibin A secretion in GnRH agonist-treated subjects after
the development of a preovulatory follicle, inhibin B levels also
increased. The standardized exogenous rhFSH and rhLH regimen used to
ensure uniform gonadotropin stimulation in all subjects resulted in the
growth of multiple small (<10 mm) and medium-sized (10–15 mm)
follicles in addition to preovulatory follicles (
16 mm) in these
subjects. As rhFSH increased both inhibin A and inhibin B secretion
from small follicles, but rhLH did not, it was unclear whether small
and medium-sized follicles and/or the preovulatory follicle contribute
to rhFSH-stimulated serum inhibin A and inhibin B levels in the late
follicular phase after rhFSH stimulation.
We therefore examined rhFSH-stimulated dimeric inhibin secretion from preovulatory follicles in spontaneous menstrual cycles to avoid the confounding effect of multifollicular development. In these spontaneous cycles, only one preovulatory follicle developed, accompanied by one to four additional follicles of 11 mm or less. When rhFSH was administered, inhibin A and E2, but not inhibin B, secretion was elicited. These findings suggest that gonadotropin stimulation of the preovulatory follicle does not contribute to serum inhibin B levels in the late follicular phase of the normal menstrual cycle. Further, they are consistent with our previous findings in which the recovery of inhibin A secretion from the preovulatory follicle was not correlated with inhibin B after treatment with a GnRH antagonist in the late follicular phase in normal women (21). If the majority of serum inhibin B results from increased granulosa cell and follicle number (25) as discussed previously, the continued rise in inhibin B with FSH stimulation in gonadotropin-stimulated cycles (24, 29), as opposed to the decrease seen in the late follicular phase of spontaneous cycles (5, 7), probably results from inhibin B secretion from continued growth of small and medium-sized follicles. In contrast, after the development of a dominant follicle in spontaneous cycles, administration of FSH does not increase the number of nondominant follicles (30) or granulosa cell proliferation (31), thus accounting for the absence of a serum inhibin B increase after rhFSH stimulation in the late follicular phase of spontaneous cycles in the current study.
In summary, in small antral follicles, inhibin B is an immediate product of FSH stimulation, and its increase may result indirectly from FSH-stimulated follicle growth rather than direct stimulation. Inhibin B, but not inhibin A, secretion from these small antral follicles is enhanced by E2. Inhibin A is stimulated at a later stage of follicle growth and becomes the predominant form of inhibin secreted in response to both FSH and LH from the preovulatory follicle. In conclusion, the increase in both FSH and E2 in the early follicular phase results in an increase in inhibin B secretion, whereas the selective LH rise in the late follicular phase favors inhibin A secretion. Thus, both the differential secretion of LH and FSH and the stage of follicle development determine the patterns of inhibin A and inhibin B secretion in the normal menstrual cycle.
Acknowledgments
We thank the nurses of the Mallinkrodt General Clinical Research Center for their exceptional patient care and organization. We also acknowledge Patrick Sluss, Ph.D., and the members of the Reproductive Sciences Core Laboratory for their assay expertise. We thank Ferring Pharmaceuticals Ltd. (Kiel, Germany) for providing Decapeptyl Depot, and Serono Laboratories, Inc. (Norwell, MA), for supplying the recombinant human LH (Luveris) and recombinant human FSH (Gonal-F).
Footnotes
1 This work was supported by NIH Grants U54-HD-29164, M01-RR-01066,
K24-HD-01290, and P30-HD-28138. 
Received September 11, 2000.
Revised December 1, 2000.
Accepted December 8, 2000.
References
-subunit in human serum.
Implications for radioimmunoassay. J Clin Endocrinol Metab. 70:1208–1212.This article has been cited by other articles:
 |
|
 |
|
  S. G. Hillier Paracrine support of ovarian stimulation Mol. Hum. Reprod., December 1, 2009; 15(12): 843 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Broekmans, M. R. Soules, and B. C. Fauser Ovarian Aging: Mechanisms and Clinical Consequences Endocr. Rev., August 1, 2009; 30(5): 465 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hirshfeld-Cytron, R. B. Barnes, D. A. Ehrmann, A. Caruso, M. M. Mortensen, and R. L. Rosenfield Characterization of Functionally Typical and Atypical Types of Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1587 - 1594. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Kevenaar, A. P.N. Themmen, J. S.E. Laven, B. Sonntag, S. L. Fong, A. G. Uitterlinden, F. H. de Jong, H. A.P. Pols, M. Simoni, and J. A. Visser Anti-Mullerian hormone and anti-Mullerian hormone type II receptor polymorphisms are associated with follicular phase estradiol levels in normo-ovulatory women Hum. Reprod., June 1, 2007; 22(6): 1547 - 1554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Y. Andersen, P. Humaidan, H. B. Ejdrup, L. Bungum, M.L. Grondahl, and L.G. Westergaard Hormonal characteristics of follicular fluid from women receiving either GnRH agonist or hCG for ovulation induction Hum. Reprod., August 1, 2006; 21(8): 2126 - 2130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Lambert-Messerlian and B. L. Harlow The Influence of Depression, Body Mass Index, and Smoking on Serum Inhibin B Levels in Late Reproductive-Aged Women J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1496 - 1500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, H. Newton, J. A. Spaliviero, C. M. Allan, B. Marshan, D. J. Handelsman, and P. J. Illingworth Gonadotropin Control of Inhibin Secretion and the Relationship to Follicle Type and Number in the hpg Mouse Biol Reprod, October 1, 2005; 73(4): 610 - 618. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. K. Welt, P. C. Smith, and A. E. Taylor Evidence of Early Ovarian Aging in Fragile X Premutation Carriers J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4569 - 4574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Popovic, M. Djurovic, A. Cetkovic, D. Vojvodic, S. Pekic, S. Spremovic, M. Petakov, S. Damjanovic, N. Milic, C. Dieguez, et al. Inhibin B: A Potential Marker of Gonadal Activity in Patients with Anorexia Nervosa during Weight Recovery J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1838 - 1843. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.I. Parker, D.M. Robertson, N.P. Groome, and K.L. Macmillan Plasma Concentrations of Inhibin A and Follicle-Stimulating Hormone Differ Between Cows with Two or Three Waves of Ovarian Follicular Development in a Single Estrous Cycle Biol Reprod, March 1, 2003; 68(3): 822 - 828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. K. Welt, A. E. Taylor, K. A. Martin, and J. E. Hall Serum Inhibin B in Polycystic Ovary Syndrome: Regulation by Insulin and Luteinizing Hormone J. Clin. Endocrinol. Metab., December 1, 2002; 87(12): 5559 - 5565. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. A. Klein, A. J. Harper, B. S. Houmard, P. M. Sluss, and M. R. Soules Is the Short Follicular Phase in Older Women Secondary to Advanced or Accelerated Dominant Follicle Development? J. Clin. Endocrinol. Metab., December 1, 2002; 87(12): 5746 - 5750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Welt, Y. Sidis, H. Keutmann, and A. Schneyer Activins, Inhibins, and Follistatins: From Endocrinology to Signaling. A Paradigm for the New Millennium Experimental Biology and Medicine, October 1, 2002; 227(9): 724 - 752. [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 |
)
and maximum (
) inhibin A (Inh A), E2, and inhibin B (Inh
B) levels in small antral follicles after rhLH (A), rhFSH (B), or rhFSH
plus E2 (C) stimulation in GnRH agonist-treated subjects.
The scales for Inh A are located on the left axes, and
those for E2 and Inh B are on the right
axes. To convert estradiol to picomoles per L, multiply by
3.671. Note the difference in scale of the hormone data in rhLH
stimulation cycles (A). **, P < 0.05.
