0021-972X/87/6505-0954$02.00/0
Journal of Clinical Endocrinology and Metabolism
Copyright © 1987 by The Endocrine Society
Vol. 65, No. 5
Printed in U.S.A.
R. I. McLACHLAN, D. M. ROBERTSON, D. L. HEALYt, H. G. BURGER AND
D. M. DE KRETSER
Medical Research Centre, Prince Henry's Hospital (R.I.M., D.L.H., H.G.B.), and the Departments of
Anatomy (D.M.R., D.M.d.K.) and Obstetrics and Gynecobgy (D.L.H.), Monash University,
Melbourne, Australia
ABSTRACT. Serum inhibin concentrations were measured daily by RIA in six normal women throughout one menstrual cycle. The RIA was specific for inhibin, and inhibin subunits and related proteins cross-reacted minimally in it. In the early to midfollicular phase, inhibin levels changed little, while in the late follicular phase, inhibin levels rose, in parallel with estradiol
(r = 0.43; P < 0.05; n = 22), to a peak level of 714 (407-1267)
U/L (geometric mean ± 67% confidence limits) coincident with the midcycle LH and FSH surges. An inverse relationship was found between serum inhibin and FSH during the mid- to late follicular phase (r = 0.42; P < 0.01; n = 45). Inhibin levels rose further during the luteal phase to a peak level of 1490 (1086-
2028) U/L 7-8 days after the LH surge, and they correlated positively with serum progesterone (r = 0.76; P < 0.001; n = 49) and inversely with serum FSH (r = 0.43; P < 0.01; n = 49) throughout the luteal phase. We conclude that 1) circulating inhibin is detectable throughout the normal menstrual cycle; 2) in the late follicular phase, inhibin levels rise in parallel with estradiol, consistent with the concept that both are products of the maturing follicle; 3) in the luteal phase, the profile of inhibin suggests that it is a secretory product of the corpus luteum; and
4) the inverse relationship between inhibin and FSH in the follicular phase is consistent with the inhibin hypothesis, while at midcycle there is loss of the inhibitory effect of inhibin on
FSH secretion. The inverse relationship between FSH and inhibin during the luteal phase suggests a hitherto unsuspected role for inhibin in the feedback regulation of FSH secretion. (J
Clin Endocrinol Metab 6 5 : 954, 1987)
T
HE NORMAL menstrual cycle has been characterized by the measurement of a number of pituitary and ovarian hormones in serum, including LH, FSH,
PRL, estradiol, and progesterone (1-4). However, understanding of the hormonal control of human folliculogenesis is incomplete due to limited information concerning the changes in the gonadal hormone inhibin, which is considered important in the feedback regulation of FSH. After its purification from bovine follicular fluid
(bFF) (5, 6), a RIA for serum inhibin was developed (7) and used to demonstrate that exogenous FSH stimulates inhibin secretion in women undergoing ovarian hyperstimulation before in vitro fertilization (IVF) (8). In that study, a close relationship between serum inhibin and estradiol was found, suggesting that inhibin was produced by the granulosa cells of the developing follicles.
However, the sensitivity of that assay was insufficient to measure circulating inhibin levels during the normal menstrual cycle.
Received March 17,1987.
Address requests for reprints to: Dr. R. I. McLachlan, Department of Anatomy, Monash University, Clayton, Victoria 3168, Australia.
* This work was supported by the National Health and Medical
Research Council of Australia.
t Wellcome Trust Senior Clinical Fellow.
954
In this study, we describe the development of a more sensitive RIA and its application to the measurement of serum inhibin during the normal human menstrual cycle.
We then assessed the relationships between serum inhibin, FSH, LH, estradiol, and progesterone during the cycle.
Subjects
The six normal women studied were ovulating regularly, as judged by their menstrual history (mean cycle length, 28.0 days; range, 25-35 days) and luteal phase serum progesterone concentrations (>8 ng/mL). Three had borne children. The women were bled daily from the first day of menses until the day before the next menses. Serum was stored at - 2 0 C before assay.
Preparations
Pools of serum were obtained from 1) eight postmenopausal women over 55 yr of age (PMS), 2) seven women undergoing ovarian hyperstimulation for the purpose of IVF (IVF-QC), 3) a woman during the midluteal phase of a normal cycle (NCS)', and 4) an ovariectomized woman (OS).
bFF was obtained from the local abbatoir. Human follicular fluid (hFF) was obtained at the time of oocyte collection from
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

SERUM INHIBIN DURING THE MENSTRUAL CYCLE women participating in the IVF program at the Queen Victoria
Medical Centre/Epworth Hospital, Melbourne.
Hormones. Rat(r) FSH (1-5), ovine(o) FSH (1-1), human(h)
FSH (AFP 4161B), rLH (1-6), rLHa (AFPCO43), rLH/3 (AFP
A805), rTSH (1-6), TSH (1-5), and 4PRL (1-5) were generously donated by the NIDDK, NIH (Bethesda, MD). Mouse epidermal growth factor, ovine insulin-like growth factor II (oIGF-II;
1-55), and mouse nerve growth factor were kindly provided by
Dr. C. Brown, Department of Physiology, Monash University.
hCG was kindly provided by the Centre for Population Res e a r c h , NICHHD (Bethesda, MD). Bovine Mullerian inhibitory substance (MIS) was generously provided by Dr. J. Hutson,
Royal Children's Hospital, Melbourne. Transforming growth factor-/3 was obtained from R. & D. Systems (Minneapolis,
MN); bTSH and BSA were purchased from Sigma Chemical
Co. (St. Louis, MO). Recombinant-derived human Thr
59
-IGF-
I was obtained from (Amgen Biologicals, Thousand Oaks, CA), and LHRH from Hoechst (Melbourne, Australia).
955
Inhibin antiserum
Immunization procedure. The inhibin antiserum (As749) was raised by immunization with pure 3IK bFF inhibin in a castrate male New Zealand White rabbit using a protocol described previously (7). A primary immunization of 20 tig was followed by booster immunizations of approximately 5 fig at 1, 4, and 6 months. The animal was then bled once or twice weekly after the final booster injection, and the antibody titer was assessed by the ability of the antiserum to bind
125
I-labeled 31K inhibin.
A pool of serum from 1-7 weeks postbooster immunization was then made and used in the RIA.
In vitro neutralization of inhibin bioactivity. The capacity of inhibin antiserum to neutralize the inhibin bioactivity of hFF was examined in the inhibin in vitro bioassay. A maximal FSHsuppressing dose of 31K bFF inhibin was more than 90% neutralized by coincubation of bFF and 0.5 /xh antiserum for 1 h before addition to culture wells. The corresponding degrees of neutralization with the partially purified hFF inhibin were
56% (0.5 nL) and more than 90% (2 ML).
In vitro bioassays
Inhibin bioactivity was measured using an in vitro bioassay based on the dose-dependent reduction of FSH cell content of rat pituitary cells in culture (9). This bioassay used a bFF standard preparation previously calibrated against an ovine testicular lymph preparation with an arbitrarily defined unitage
>- of 1 U/mg. In vitro neutralization of inhibin bioactivity by the inhibin antiserum was assessed in this bioassay system, as previously described (7).
Isolation of hormones
Iodination of inhibin
The 58K and 31K inhibin forms have been isolated from bFF and found to have similar in vitro bioactivities and immunoactivities (5, 6). These hormones were iodinated using the chloramine-T procedure and purified on Matrex red A (Amicon,
Danvers, MA), as previously described (7). Both iodinated forms were stable in bFF, while only the 31K form was stable during incubation with serum under RIA conditions; the 59K form was degraded to the 3IK form, presumably as the result of serum proteolytic cleavage of the extended A inhibin subunit
(7, 9A). Thus, iodinated 31K bFF inhibin, with a specific activity of 40 fiCi/ng, was used in this assay as well as in all other serum inhibin RIA systems described (7, 8).
bFF was collected in the presence of proteolytic enzyme inhibitors, and 31K bFF inhibin was isolated as previously described (5, 6). hFF inhibin, used as RIA standard (hFF-std), was partially purified using the first three steps employed in the purification of 31K bFF inhibin. Briefly, 200 mL hFF was fractionated on Sephacryl S-200 in 0.05 mol/L ammonium acetate, pH 7.0, and the inhibin bioactivity was recovered in the void volume fractions. These fractions were pooled, pH
> precipitated, and fractionated on Sephadex G-100 in 4 mol/L acetic acid. The inhibin fractions were lyophilized, resuspended in 4 mol/L acetic acid, and fractionated by reverse phase high pressure liquid chromatography. Inhibin biological and immunological activities coeluted from this high pressure liquid chromatographic column in a fashion similar to those of 3IK
>
bFF inhibin (6). The material was defined in terms of its inhibin bioactivity, as the specific activity of pure hFF inhibin is unknown. Assuming a specific activity similar to that described for 31K bFF inhibin (750,000 U/mg) (6), the RIA standard curve covered a range from 0.15-4.80 ng/mL (100-3,600 U/L).
Bovine inhibin /3-subunit dimer was purified from bFF using a purification procedure similar to that described for 31K bFF
> inhibin (6), and its purity was confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and NH
2
-terminal amino acid sequencing (unpublished data).
Inhibin RIA procedure
A double antibody RIA was established for the measurement of human serum inhibin. Serum and standards (200 nL) and antiserum (1:2666; 50 ML in 100 mmol/L phosphate and 0.5%
BSA, pH 7.2, containing 1:400 normal rabbit serum), in duplicate or triplicate, respectively, were incubated for 2 days at 4 C before the addition of
125
I-labeled 31K bFF inhibin (50 nL;
10,000 cpm in 100 mmol/L phosphate buffer, pH 7.4, containing
0.5% BSA and 0.1% Triton X-100), and the incubation was continued for a further 2 days at 4 C. Second antibody (goat antirabbit serum; 1:50; 100 nh) was added, and the incubation was continued overnight at 4 C, after which 2 mL 150 mmol/L
NaCl were added, the tubes were centrifuged at 2,000 x g for
45 min, decanted, and counted. Standards and samples were diluted in PMS to a total volume of 200 /xL to compensate for the nonspecific effects of serum in the RIA.
Effects of serum in the RIA
Addition of either OS or PMS at volumes greater than 100
^L/tube resulted in displacement of the inhibin tracer compared to phosphate buffer (Fig. la). Consequently, studies were
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

956
+ 2
( b )
P M S O S
Ml M>
45 180 720
31kDa bFF mil
45 180 720 hFF-std mU i
6-25 25
IVF QC
100
McLACHLAN ET AL. JCE&M«1987
Vol 65 • No 5
0.93; P < 0.001; n = 30; slope = 1.07). Since the error of the intercept of the regression line on the x,y-axis included the origin, and the slope of the line was unity, these data indicated that the addition of PMS to the RIA tubes compensated for the nonspecific effect of serum in the new RIA. The nature of this nonspecific effect is unknown.
Despite the demonstrated high specificity of the RIA (see below), the measured inhibin immunoreactivity may reflect cross-reactivity of an unknown protein(s). Other possibilities include the presence of binding proteins that may interfere with the assessment of bound and free hormone. Finally, the measured immunoreactivity may represent nonspecific interference by the high concentration of serum in the RIA tubes, as described for human gonadotropin assays (9B).
To assess whether serum had any confounding effects in the
RIA, the hFF standard was assayed in the presence or absence of serum (Fig. 1). The dose-response lines were parallel. Graded doses of the hFF standard were added to a set volume of the
IVF-QC serum, and inhibin levels were determined against the same doses of the standard. Regression analysis of the anticipated and measured inhibin values revealed a slope of 1.06, with an intercept not significantly different from the activity in IVF-QC serum alone. These results indicate that serum did not interfere with the measurement of inhibin in the RIA.
45 180 720 hFF-std mU
FiG. 1. a, Logit-log dose response lines in the RIA of 31K bFF inhibin, hFF-std, PMS, OS, and IVF-QC. In this RIA, samples were diluted in
10 mM phosphate-0.5% BSA buffer, b, Logit-log dose-response lines in the RIA of the hFF-std, IVF-QC, and NCS. In this RIA, the standard and samples were diluted to a volume of 200 nL in PMS (for further details see text).
Assay characteristics
Neither the collection of blood as serum or plasma nor repeated freeze-thawing (up to three times) affected the measured inhibin immunoactivity. Parallelism was assessed by comparison of the slopes of the logit-log dose-response lines for the hFF std [mean, 1.18 ± 0.13 (±SD); n = 12] and IVF-QC serum pool (1.13 ± 0.12; n = 12), which were not significantly different. The sensitivity of the assay (ED
5
) was 90 U/L, with an
ED
50
of 570 U/L. Within- and between-assay coefficients of variation (CV) were derived from the repeated measurement of multiple dilutions of the IVF-QC serum covering the logit range
+2 to - 2 (bound to free ratio, 88%:12%) and calculated using parallel line bioassay statistics. The within-assay CV, obtained from the mean index of precision of 12 assays, was 0.064. In addition, the CVs in the upper, mid-, and lower regions of the standard curves were 8%, 5%, and 6%, respectively. The between-assay variation, as determined from the CV of repeated measurement of the IVF-QC serum pool, was 8.0% from 12 assays. Examples of the logit-log dose-response lines for the hFF-std and IVF-QC and NCS serum pools are shown in Fig.
lb.
undertaken to establish if this displacement was attributable to inhibin or to nonspecific effects. Evidence that the effects of these serum samples were nonspecific was obtained by a comparison of serum inhibin levels obtained using this RIA and those obtained using a previously described method based on an antiserum (As474) to 58K bFF inhibin (7). The latter RIA, although 4-fold less sensitive than the new RIA, had no crossreactivity with OS or PMS. A close correlation was found between values obtained using both assay systems for serum samples obtained from women during ovulation induction (r =
Cross-reaction of hormones
After logit-log dose transformation of the response curves,
125 linear and parallel displacement of I-labeled 31K bFF inhibin was produced by 31K bFF inhibin and hFF std (Fig. la).
The cross-reactivities of the following purified glycoproteins, growth factors, and inhibin-related peptides were less than
0.1% relative to 31K bFF inhibin: rFSH, oFSH, hFSH, rLH, oLH, hLH, rLHa, rLHft rTSH, bTSH, hTSH, hCG, mouse epidermal growth factor, oIGF-II, mouse nerve growth factor, hIGF-I, bMIS, rPRL, LHRH, and BSA. Porcine transforming
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

SERUM INHIBIN DURING THE MENSTRUAL CYCLE growth factor-/3 had less than 0.7% cross-reactivity. Bovine inhibin 0-subunit dimer had less than 1% cross-reactivity.
Bovine 31K inhibin subunits obtained after reduction and alkylation had less than 0.2% cross-reactivity. After reduction and alkylation of the hFF std preparation (10), the inhibin subunits had 0.4% cross-reactivity in the RIA.
Other RIAs
Serum LH and FSH concentrations were determined in duplicate by RIA using reagents supplied by the WHO Matched
Reagents Program (11) and are expressed in terms of the First
International Reference Preparation of LH and the Second
International Reference Preparation of FSH immunoassay standards, with interassay CVs of 7.3% and 7.8%, respectively.
Both serum estradiol and progesterone were measured in duplicate by RIA (Coat-a-Count, Diagnostic Products Corp., Los
Angeles, CA), with interassay CVs of 8.7% and 8.1%, respectively, from 150 assays.
957
Luteal phase
After the midcycle rise, serum inhibin levels fell approximately 50% to a nadir on day 2, although this fall was not significant. Subsequently, inhibin levels rose during the luteal phase to a peak of 1490 (1086-2028) U/
L on days 7 and 8. This value was significantly higher
(P < 0.05) than that at midcycle. Both FSH and LH levels declined rapidly in the early luteal phase and remained low until the perimenstrual period. In the early to midluteal phase (days 2-10), inhibin was inversely correlated with FSH (r = 0.43; P < 0.01; n = 49) and directly correlated with progesterone (r = 0.76; P < 0.001; n = 49), but not with estradiol.
Statistical analysis
Hormone levels were determined at one dilution against the standard preparation using an iterative curve-fitting procedure
(12). In the calculation of results, a log-normal distribution of individual observations (13) was assumed, i.e. all calculations were performed using logarithmically transformed values to give geometric means and 67% confidence intervals. Statistical comparisons between hormone levels on different days were made using Duncan's multiple range test (14) and, in some cases, Kramer's (15) modification to group means with unequal numbers of replications.
Perimenstrual period
In view of the variable cycle length, the late luteal phase and perimenstrual period were examined by combining, for each woman, the first and last 4 days of the study cycle so as to normalize to the day of menses
(defined as day 0; Fig. 3). A significant (P < 0.01) fall in inhibin occurred in the 2 days before the onset of menses, while FSH rose in the same period (P < 0.01). Inhibin was inversely correlated with FSH (r = 0.40; P < 0.05; n
= 21) and directly correlated with progesterone (r = 0.82;
P < 0.001; n = 20), but not with estradiol, on days - 4 to
0 inclusively.
Mean serum inhibin, FSH, LH, estradiol, and progesterone levels during the menstrual cycle in normal women are shown in Fig. 2. All cycles were normalized according to the day of the LH surge (defined as day 0).
. Follicular phase
Serum inhibin levels did not vary (range of means,
286-428 U/L) between days -14 to - 3 and then increased significantly (P < 0.05) to a midcycle peak of
714 (407-1267) U/L (geometric mean ± 67% confidence limits). Serum FSH rose significantly (P < 0.05) between days -14 and - 1 1 , fell (P < 0.01) to a nadir on day - 3 ,
> then rose to a peak at midcycle (day 0). Serum inhibin was inversely correlated with serum FSH in the mid- to late follicular phase (days -10 to - 2 ; r = 0.42; P < 0.01; n = 45). Serum estradiol rose in parallel with inhibin in the late follicular phase, although estradiol peaked 1 day earlier (day —1). A significant correlation was found
>• between serum estradiol and inhibin in the late follicular phase (between days - 4 and - 1 (r = 0.43; P < 0.05; n =
22).
This study demonstrates that immunoreactive inhibin is detectable in the circulation throughout the menstrual cycle in normal women and presents evidence in support of a physiological role for inhibin in the modulation of pituitary FSH release during both the follicular and luteal phases.
The physiological validity of the observations made in this study is dependent upon the specificity of the RIA.
This was established by several criteria. First, the polyclonal antiserum, generated using pure 31K bFF inhibin, recognized the site(s) required for expression of bFF and hFF inhibin bioactivity, since the antiserum neutralized their biological activities in vitro. Secondly, the crossreactivity of a wide range of purified glycoprotein hormones and, in particular, inhibin-related proteins
(TGFft MIS, bFF, and inhibin 0-subunit dimer) in the
RIA was minimal, underlining that the conformation of the A and B subunits of intact inhibin was fundamental to this antibody-ligand interaction. Finally, in view of the limited displacement of tracer that occurred compared to RIA buffer when OS or PMS was added, we compared the inhibin values obtained using the present
RIA with those measured by our previous assay method
(7), which was unaffected by these serum samples. A
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

958
2000 -
1500 z
1000h • m i
- 500
McLACHLAN ET AL.
/
JCE & M • 1987
Vol 65 • No 5
T 6 0
FIG. 2. Serum inhibin, FSH, LH, estradiol, and progesterone concentrations during the normal menstrual cycle. Data were normalized around the day of the
LH surge ( ). The data are expressed as the geometric mean and 67% confidence limits with between four and six observations per point. The factors for conversion of values to Systeme International (SI) units are: estradiol, pg/ml x 3.671 = pmol/L, progesterone, ng/mL
X 3.180 = nmol/L.
400 r
200 -
0
L
-14 - 1 2 - 1 0 close correlation was found between the two assays, indicating that the new RIA was not detecting additional immunoactivity, justifying the assumption that the addition of PMS to the assay tubes compensated for the nonspecific inhibition of tracer binding by PMS.
Follicular phase
During the early follicular phase both serum inhibin and estradiol levels were low. It is during this period that the perimenstrual FSH rise leads to the recruitment of a cohort of follicles from which the dominant follicle will subsequently arise (16). The relatively constant inhibin and estradiol levels are consistent with a small biosynthetic capacity for inhibin in these small follicles.
The absence of a significant rise in serum inhibin in the midfollicular phase, during which selection and emergence of the dominant follicle occur, was surprising, particularly as the number of granulosa cells and their capacity to secrete estradiol and respond to FSH increase with follicular development (17). It might, therefore, be expected that their capacity to secrete inhibin would similarly increase. A significant rise in inhibin levels also might be likely during this time in view of the significant inverse relationship with FSH which itself declines during this period. Our failure to find a rise in inhibin levels could be attributed to the wide subject variation and small sample number. On the other hand, inhibin levels
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

1500 -
SERUM INHIBIN DURING THE MENSTRUAL CYCLE 959 studies indicating that granulosa cells produce inhibin
(18, 19) and with the finding of inhibin bioactivity in human ovarian venous blood at this time of the cycle
(20). The data are also consistent with observations suggesting that granulosa cells from large follicles produce more inhibin (21, 22). The findings are in agreement with our previous report using a less sensitive RIA, in which rapid and parallel rises in inhibin and estradiol were found during exogenous FSH administration in women undergoing ovulation induction for IVF (8).
These results support the view that inhibin is a marker of granulosa cell function. It is interesting that estradiol reached a peak on day —1 (as described previously) (2), yet inhibin continued to increase to a midcycle peak on day 0, suggesting different regulation of the secretion of these two parameters of follicular growth.
J
0
-120
15
1 0
5
u
O
O oc
0> c
UJ z o a
UJ
- 4 - 3 - 2 - 1 0 1 2 3
DAYS
FIG. 3. Serum inhibin, FSH, LH, estradiol, and progesterone concentrations during the perimenstrual period. For this figure, the first and last 4 days of the study cycle for each individual were normalized around the first day of menses (the noncontinuity between days - 1 and 0 is indicated by the vertical lines). This figure includes data from
Fig. 2 (days -14 to +10) plus a substantial number of data points that fell outside that time frame due to the variable cycle length. The time of menses is indicated by the shaded area. These data are expressed as the geometric mean and 67% confidence limits, with between four and six observations per point. SI conversion values are given in Fig. 2.
may, in fact, not rise during the midfollicular phase, in which case another mechanism(s) (e.g. increased sensitivity of the pituitary to inhibin feedback, the effect of the small rise in estradiol) must operate to progressively reduce FSH levels during this period.
The parallel rises in inhibin and estradiol levels during the late follicular phase are consistent with previous
Periovulatory phase
Serum FSH and inhibin peaked simultaneously with the LH surge, a pattern that seems contrary to a feedback role for inhibin. We speculate that FSH rises in spite of the increasing inhibin concentration as a result of the overriding effect of increased LHRH pulse frequency and amplitude, as implied by the measurement of LH pulses at this stage of the cycle (23, 24). In addition, the rising estradiol and progesterone levels, which are also important in the induction of the midcycle gonadotropin surge
(25), may result in the loss of inhibin feedback upon
FSH.
Luteal phase
The decline in inhibin levels during the first 2 days of the luteal phase paralleled that in estradiol. The reason for this decline in both hormones is uncertain, but probably relates to the disruption of the architecture of the follicle at ovulation and the formation of a corpus hemorrhagicum.
The peak serum inhibin concentrations occurred during the midluteal phase. This was an unexpected finding in view of the failure of bovine (26) and rat (27) luteal cells to secrete significant amounts of inhibin in vitro.
The remarkably similar patterns of secretion of inhibin and progesterone during the mid- and late luteal phases strongly suggest that the corpus luteum is indeed the source of inhibin. The difference between species suggests that luteal phase inhibin secretion is limited to primates. On the other hand, the mRNA for the inhibin
A subunit has been detected in rat corpus luteum (28).
Further studies of the secretory pattern of corpora lutea are required to resolve these differences. Moreover, the regulation of inhibin secretion by the human corpus luteum requires further study, since the progressively decreasing levels of FSH would be an unlikely stimulus.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

960
McLACHLAN ET AL.
Whether inhibin secretion by the corpus luteum is subject to control by LH, as has been suggested for progesterone (29), is presently unknown.
The production of significant amounts of inhibin by small antral follicles during the mid- to late luteal phase seems unlikely, in that such follicles are few in number
(zero to four per ovary), are small in size (<5 mm in diameter), and contain low aromatase activity and low estradiol concentrations in their follicular fluids (30).
Although granulosa cells from these follicles respond to
FSH in vitro, serum FSH levels are low during the luteal phase.
The significant inverse relationship between FSH and inhibin during the luteal phase suggests a role for inhibin in the regulation of FSH during this phase of the cycle.
The suppression of FSH during the luteal phase has hitherto been attributed to progesterone and estrogen secretion by the corpus luteum, while the perimenstrual rise in FSH has been attributed to the withdrawal of these steroids as the corpus luteum regresses (1). Nonetheless, the possibility that a nonsteroidal factor(s) may be important at this time is supported by the observation that the perimenstrual rise of FSH in monkeys cannot be blocked by the administration of progesterone in a manner designed to maintain progesterone concentrations in the peak luteal phase range (31). The luteal phase peak of inhibin described in this study cells for a revaluation of the relative contribution of inhibin and steroids to the suppression of FSH. It is tempting to postulate that the decline in inhibin that occurs in conjunction with regression of the corpus luteum is associated with the rise in FSH required to stimulate a new cohort of follicles for the next menstrual cycle.
We thank F. de Vos, M. Giacometti, S. Hayward, M. Bangah, and
H. Quigg for their excellent technical assistance and J. Stirling and S.
Simpson for the preparation of this manuscript.
1. Ross GT, Cargille CM, Lipsett MB, Rayford PL, Marshall JR,
Strott CA, Rodbard D 1970 Pituitary and gonadal hormones in women during spontaneous and induced ovulatory cycles. Recent
Prog Horm Res 26:1
2. Korenman SG, Sherman BM 1973 Further studies of gonadotropin and estradiol secretion during the preovulatory phase of the human menstrual cycle. J Clin Endocrinol Metab 36:1205
3. Judd HL, Yen SSC 1973 Serum androstenedione and testosterone levels during the menstrual cycle. J Clin Endocrinol Metab 36:475
4. Vekemans M, Delvoye P, L'Hermite M, Robyn C 1977 Serum prolactin levels during the menstrual cycle. J Clin Endocrinol
Metab 44:989
5. Robertson DM, Foulds DM, Leversha L, Morgan FJ, Hearn MTW,
Burger HG, Wettenhall REH, de Kretser DM 1985 Isolation of inhibin from bovine follicular fluid. Biochem Biophys Res Commun
126:220
6. Robertson DM, de Vos FL, Foulds LM, McLachlan RI, Burger
HG, Morgan FJ, Hearn MTW, de Kretser DM 1986 Isolation of
JCE & M • 1987
Vol 65 • No 5
31 kD form of inhibin from bovine follicular fluid. Mol Cell
Endocrinol 44:271
7. McLachlan RI, Robertson DM, Burger HG, de Kretser DM 1986 <
The radioimmunoassay of bovine and human follicular fluid and serum inhibin. Mol Cell Endocrinol 46:175
8. McLachlan RI, Robertson DM, Healy DL, de Kretser DM, Burger
HG 1986 Plasma inhibin levels during gonadotropin-induced ovarian hyperstimulation for IVF: a new index of follicular function?
Lancet 1:1233
9. Scott RS, Burger HG, Quigg H 1980 A simple and rapid in vitro bioassay for inhibin. Endocrinology 107:1536 4
9A. Forage RG, Ring JM, Brown RW, Mclnerney BV, Cobon GS,
Gregson RP, Robertson DM, Morgan FJ, Hearn MTW, Findlay
JK, Wettenhall REH, Burger HG, de Kretser DM 1986 Cloning and sequence analysis of cDNA species coding for the two subunits of inhibin from bovine follicular fluid. Proc Natl Acad Sci USA
83:3091
9B. Hunter WM, Bennie JG 1979 Reduction of non-specific serum responses in human pituitary gonadotrophin radioimmunoassays. *
J Endocrinol 80:59
10. Leversha LJ, Robertson DM, de Vos FL, Morgan FJ, Hearn MTW,
Wettenhall REH, Findlay JK, Burger HG, de Kretser DM 1987
Isolation of inhibin from ovine follicular fluid. J Endocrinol 113:213
11. 1986 Programme for the Provision of Matched Assay Reagents in the Radioimmunoassay of Hormones in Reproductive Physiology:
Method Manual, ed 10. WHO, Geneva, p 17
12. Burger HG, Lee VWK, Rennie GC 1972 A generalized computer 4 program for the treatment of data for competitive protein-binding assays including radioimmunoassays. J Lab Clin Med 80;302
13. Gaddum JH 1945 Lognormal distributions. Nature (Lond) 156:463
14. Duncan DB 1985 Extension of multiple F tests. Biometrics 11:1
15. Kramer CY 1956 Extension of multiple range tests to groups means with unequal numbers of replications. Biometrics 12:307
16. Baird DT 1983 Factors regulating the growth of the preovulatory follicle in the sheep and human. J Reprod Fertil 69:343
17. McNatty KP, Smith DM, Makris A, Osthanodh R, Ryan KJ 1979
The microenvironment of the human antral follicle: interrelationships among the steroid levels in antral fluid, the population of granulosa cells, and the status of the oocyte in vivo and in vitro. J
Clin Endocrinol Metab 49:851
18. Erickson GF, Hseuh AJW 1978 Secretion of "inhibin" by rat granulosa cells in vitro. Endocrinology 103:1960
19. Channing CP, Anderson LD, Hoover DG, Kolena J, Osteen KG,
Pomerantz SH, Tanabe K 1982 The role of nonsteroidal regulators in control of oocyte and follicular maturation. Recent Prog Horm <
Res 38:331
20. Channing CP, Gagliano P, Tanabe K, Fortuny A, Cortes-Prieto J
1985 Demonstration of a gradient in inhibin activity, estrogen, progesterone and A
4
-androstenedione in follicular fluid, ovarian vein blood, and peripheral blood of normal women. Fertil Steril
43:142
21. Channing CP, Hoover DJ, Anderson LD, Tanabe K 1982 Control of follicular secretion of inhibin in vitro and in vivo. In Fujii T, 4
Channing CP (eds) Non-Steroidal Regulators in Reproductive
Biology and Medicine. Pergamon Press, New York, p 41
22. Henderson KM, Franchimont P, Charlet-Renard Ch, McNatty KP
1984 Effect of follicular atresia on inhibin production by bovine granulosa cells in vitro and inhibin concentrations in the follicular fluid. J Reprod Fertil 72:1
23. Veldhuis JD, Beitins IZ, Johnson ML, Serabian MA, Dufau ML
1984 Biologically active luteinizing hormone is secreted in episodic pulsations that vary in relation to stage of the menstrual cycle. J
Clin Endocrinol Metab 58:1050
24. Reame N, Sauder SE, Kelch RP, Marshall JC 1984 Pulsatile gonadotropin secretion during the human menstrual cycle: evidence for altered frequency of gonadotropin-releasing hormone secretion. J Clin Endocrinol Metab 59:328
25. Liu JH, Yen SSC 1983 Induction of midcycle gonadotropin surge by ovarian steroids in women: a critical evaluation J Clin Endocrinol Metab 57:797
26. Henderson KM, Franchimont P 1981 Regulation of inhibin production by bovine ovarian cells in vitro. J Reprod Fertil 63:431
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.

SERUM INHIBIN DURING THE MENSTRUAL CYCLE
27. Carson RS, Lee VWK, Inhibin content and production by antral, preovulatory and luteinized rat ovarian follicles. 26th Annual Meeting of the Endocrine Society of Australia Canberra, Australia, 1982
(Abstract 33)
28. Davis SR, Dench F, Nikolaidis I, Clements JA, Forage RG, Burger
HG 1986 Inhibin A-subunit gene expression in the ovaries of immature female rats is stimulated by pregnant mare serum gonadotropin. Biochem Biophys Res Commun 138:1196
29. Filicori M, Butler J, Crowley WF 1984 Neuroendocrine regulation
961 of the corpus luteum in the human. J Clin Invest 73:1638
30. McNatty KP, Hillier SG, van den Boogaard AMJ, Trimbos-Kemper TCM, Reichart Jr LE, van Hall EV 1983 Follicular development during the luteal phase of the human menstrual cycle. J Clin
Endocrinol Metab 56:1022
31. Resko JA, Normal RL, Niswender GD, Spies HG 1974 The relationship between progestins and gonadotropins during the late luteal phase of the menstrual cycle in rehesus monkeys. Endocrinology 94:128
Travel Grants For The Annual Meeting, 1988
The Endocrine Society is providing grants for travel and registration fees for the 70th Annual Meeting,
June 8—10, 1988 in New Orleans, Louisiana. Recipients will be selected on the basis of merit.
Application forms may be obtained from Francine Staiman, The Endocrine Society, 9650 Rockville
Pike, Bethesda, MD 20814. Completed applications must be received at that address no later than January
15, 1988.
The following restrictions apply:
1. The applicant must be an author on an abstract submitted for the 70th Annual Meeting.
2. Funds are available for travel within the United States or Canada only.
3. Recipients of 1987 travel grants, scientists holding postdoctoral research positions for more than four years, and undergraduate students are ineligible.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 04:33 For personal use only. No other uses without permission. . All rights reserved.