| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Reproductive Endocrinology |
Department of Obstetrics and Gynecology, University of Edinburgh, Center for Reproductive Biology; Simpson Memorial Maternity Pavilion (A.H.); the and Department of Pediatric Pathology and Cytogenetics, Royal Hospital for Sick Children (M.S., P.M.E.), Edinburgh; Duncan Guthrie Institute (J.A.C., D.A.A.), Glasgow; and School of Biological and Molecular Sciences, Oxford Brookes University (N.P.G.), Oxford, United Kingdom
Address all correspondence and requests for reprints to: Dr. E. M. Wallace, Department of Obstetrics and Gynecology, Monash University, Monash Medical Center, 246 Clayton Road, Clayton, Victoria 3168, Australia. E-mail: Euan.Wallace{at}med.monash.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inhibin A levels in AF rose from a median (10–90th percentile) level of 615 (158.2–1124.6) pg/mL at 14 weeks to 1336.0 (489.4–2084.1) pg/mL at 20 weeks, and inhibin B rose from 216.6 (67.4–554.6) to 1078.2 (439.3–2482.2) pg/mL over the same period. In MS, inhibin A levels fell from a median (10–90th percentile) level of 177.5 (101.4–290.7) pg/mL at 10 weeks to a nadir of 111.9 (59.5–200.3) pg/mL at 17 weeks, rising again to 180.3 (74.1–327.2) pg/mL at 20 weeks. No inhibin B was detectable in MS. In 47 pairs of matched samples (14–16 weeks gestation) there was no correlation of inhibin A levels in AF with those in MS (r = 0.19; P > 0.05). In 45 term umbilical cord serum samples, no dimeric inhibin was detectable in serum from female babies, but inhibin B was detectable in male sera; the median (10–90th percentile) concentration was 167.4 (111.2–224.8) pg/mL.
These data suggest that for the gestation periods studied, although the placenta secretes inhibin A, another source, probably the fetal membranes, secretes both inhibin A and inhibin B. Further, the presence of inhibin B in male fetuses is consistent with a testicular origin, suggesting that inhibin B may be important in the development of the fetal hypothalamo-pituitary-testicular axis.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit and one of two ß-subunits,
ßA or ßB, giving rise to two mature 32-kDa
inhibins: inhibin A (-ßA) and inhibin B
(-ßB) (2). Although inhibins were originally
identified from gonadal tissue, messenger ribonucleic acids (mRNAs) for
the inhibin subunits are expressed in many nongonadal sites (3),
including the placenta (4). Indeed, very high circulating levels of
inhibin have been reported during pregnancy, significantly higher than
those in nonpregnant subjects (5, 6, 7, 8), which then rapidly decline after
delivery (6, 8, 9). Current evidence derived from clinical studies
(4, 5, 6, 7, 8, 9, 10, 11), from the localization of inhibin subunit mRNA (4, 12) and
proteins (4, 13), and from trophoblast cell culture experiments (12)
suggests that the placenta is probably the principal source of inhibin
during pregnancy (14).
Until recently, the available assays for inhibin were unable to
differentiate between dimeric forms and partially processed free
-subunits or between dimers (15, 16, 17). Thus, our current
understanding of inhibins in pregnancy is largely based upon these
nondiscriminatory immunoreative inhibin assays (14). However, the
development of sensitive and specific enzyme-linked immunosorbent
assays (ELISAs) for inhibin A (18) and inhibin B (19) and their
subsequent application have offered novel and important insights into
inhibin biology in chromosomally abnormal pregnancies (20, 21, 22). To gain
further insight into the biology of inhibin in normal pregnancy,
therefore, we measured inhibin A and inhibin B in different pregnancy
compartments.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Amniotic fluid (AF). Aliquots of AF were obtained from the regional cytogenetics laboratory in Edinburgh, Scotland. These had been collected prospectively as part of a clinical amniocentesis service. AF was separated from fetal epithelial squamous cells by centrifugation at 250 x g within 24 h of collection, and an aliquot of each sample was stored specifically for this study at -20 C until assay. The karyotype from each sample was reported subsequently.
Maternal serum (MS). Blood samples were collected prospectively as part of the West of Scotland Down’s Syndrome and Neural Tube Defects Prenatal Screening Program and from an ongoing first trimester study of serum markers. Each sample was centrifuged, and the serum was separated within 3 days of collection and stored at -20 C. Information from early pregnancy karyotyping, where performed, and birth records, where karyotyping is not performed, is routinely used to identify chromosomally abnormal babies. Through the exclusion of these, sera from chromosomally normal pregnancies were identified and retrieved from storage. To gain sex details from this normal group, a cohort of case records was retrieved and analyzed.
Matched MS and AF. In 47 women venepuncture was performed, with informed consent, immediately before amniocentesis (14–16 weeks gestation). The blood was centrifuged, and the serum was separated on the day of collection and stored at -20 C until assay. The AF was processed as detailed above. In each case, a normal karyotype was subsequently reported.
Umbilical cord serum. In 45 normal pregnancies, at term (37–41 weeks) and in normal spontaneous labor, umbilical blood (arterial and venous) was collected after consent was obtained. After the delivery of each baby, the cord was double clamped, and blood was taken after delivery of the placenta. Blood was centrifuged on the day of collection, and serum was separated and stored at -20 C until assay.
For each sample, AF, MS, or umbilical cord serum, the time of gestation at sampling, in completed weeks, was calculated from certain menstrual dates or from an early pregnancy ultrasound scan. Ethical approval was granted by the Lothian research ethics committee.
Assays
Inhibin A. Inhibin A was measured using a two-site ELISA
(18) that has been previously validated for human serum (23) but with
some modifications. Before assay, samples (and standards) underwent two
preparatory steps. Each was mixed with 2% (final wt/vol) SDS and
heated in a water bath at 100 C for 3 min. After cooling, each sample
or standard was mixed with 1% (final wt/vol) hydrogen peroxide,
incubated at room temperature for 30 min, and then diluted 1:1 in assay
diluent (0.1 mol/L Tris-HCl, 0.15 mol/L NaCl with 5% Triton X-100,
10% BSA, and 5% normal mouse serum, pH 7.5). The assay uses an
immobilized anti-ßA-inhibin subunit monoclonal antibody (E4) as a
capture antibody, covalently coupled to hydrazide microplates
(Avidplate-HZ, UniSyn Technologies, Tustin, CA). The Fab fraction of a
mouse anti--inhibin subunit monoclonal antibody (R1) is used as a
second antibody diluted in assay diluent. This is conjugated to
alkaline phosphatase, allowing detection by the addition of an alkaline
phosphatase substrate, p-nitrophenylphosphate (Kirkegaard
and Perry Laboratories, Gaithersburg, MD). Inhibin, immunopurified from
follicular fluid and calibrated against recombinant 32-kDa human
inhibin A (Genentech, South San Francisco, CA), was used as a standard
preparation. Plates were read at 405 nm in a microplate reader
(Thermomax, Molecular Devices Corp., Menlo Park, CA) using dedicated
software (Softmax, Molecular Devices Corp.). Results are expressed as
picograms per mL, with an assay sensitivity of 23 pg/mL. The intra- and
interplate coefficients of variation were 4.3% and 5.6%,
respectively. The cross-reactivity of this assay with activin A,
activin B, follistatin, purified human pro-C, and inhibin B is less
than 0.1%. The recovery of recombinant human inhibin A spiked into AF
was quantitative (mean ± SEM, 108 ± 13%;
n = 10). Serial dilution of AF samples gave dose responses
parallel to that of the immunopurified standard (Fig. 1
).
|
C had less than 0.1%
cross-reaction, whereas recombinant inhibin A had 0.5% cross-reaction.
The intra- and interplate coefficients of variation were 7.6% and
8.1%, respectively. Recovery of recombinant human inhibin A spiked
into AF was quantitative (mean ± SEM, 88 ±
11%; n = 8). Serial dilution of AF gave a dose response parallel
to that of the immunopurified standard (Fig. 1
). Statistical analyses were performed using Statview 4.1 (Abacus, Berkeley, CA) and SPSS for Windows (SPSS, Chicago, IL). Regressed medians were calculated to compare the ontogeny of the inhibins in each compartment, and multiples of the median (MoM) were used to correct for gestational changes, allowing group comparisons across gestations.
The regression equations calculated were: AF inhibin A: median = e(11.3209–84.616/gestation); AF inhibin B: median = e(9.3254–41.656/gestation); MS inhibin A: median = -0.0129006x5 + 0.988003x4 - 29.6031x3 + 434.528x2 - 3136.00x + 9078.01 (where x is weeks gestation).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
AF samples from 618 consecutive amniocenteses were available for measurement of inhibin A content. In 15 (2.4%), karyotyping was abnormal, and these were excluded from this study. Of the 603 AF samples analyzed, 388 (64.3%) were unselected with regard to placental function [tests were performed for reasons of maternal age (n = 350), maternal anxiety (n = 19), or past history (n = 19)] and were grouped together (group A). The remaining 215 (35.7%) amniocenteses (group B) had been performed after an abnormal ultrasound scan (n = 15) or after positive MS screening, based on maternal age and MS levels of intact hCG and alphafetoprotein, had indicated an increased risk of fetal trisomy 21 (n = 200).
Inhibin A levels were analyzed for groups A and B separately by
completed week of gestation and by fetal sex. There were no significant
differences in inhibin A levels between the two groups
(P > 0.05, by Mann-Whitney U test; data not shown) or
between sexes in either group A or B (P > 0.05, by
Mann-Whitney U test; data not shown). Data from all AF samples were,
therefore, combined regardless of indication for sampling or fetal sex
and analyzed by gestational age. Table 1 details the
median and 10th and 90th percentiles for inhibin A in AF in these 603
pregnancies. Levels increased steadily across the gestational window of
14–20 weeks, as evidenced by the regressed medians (Fig. 2).
|
|
details
the median and 10th and 90th percentiles for inhibin B in these 189
pregnancies. Levels increased steadily across the gestational window,
as shown in Fig. 2.
|
). There was a significant, but weak,
association between the levels of inhibin A and B in the 189 amniotic
fluid samples (r = 0.42; P = 0.0001). MS
Sera from 807 chromosomally normal singleton pregnancies from
10–20 completed weeks of pregnancy were analyzed for inhibin A and
inhibin B. Table 3 details the distribution of these
sera by gestation and the median (10th and 90th percentiles) inhibin A
levels. Levels fell significantly from 10 weeks to a nadir at 17 weeks
(P < 0.0001, by Mann-Whitney U test), rising again to
20 weeks (P < 0.0001, by Mann-Whitney U test; Fig. 2).
|
Inhibin B was undetectable in MS.
Matched MS and AF
Inhibin A was measured in matched maternal sera, and AF samples were collected from 47 women at 14–16 weeks gestation. The median (10–90th percentiles) inhibin A levels in MS and AF were 123.7 (79.0–196.1) and 802.5 (250.6–1731.4) pg/mL, respectively; the levels in AF were significantly higher than those in MS (by paired t test, P < 0.001). Inhibin A levels in the AF and MS samples were not significantly correlated (r = 0.19; P > 0.05).
Umbilical cord serum
Cord blood was obtained from 45 pregnancies at term (37–41 weeks gestation), 24 with a female baby and 21 with a male, as evident on examination. Inhibin A was undetectable in all 45 samples, both arterial and venous. Inhibin B was undetectable in the cord serum from all female babies, but was present in serum from all 21 male babies, with no differences between arterial or venous blood levels. The median (10–90th percentile) inhibin B level in the umbilical vein was 167.4 (111.2–224.8) pg/mL. There was no association between cord inhibin B level and either gestation (r = 0.005; P = 0.98) or birth weight (r = 0.11; P = 0.6).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit-based assays (5, 6, 7, 8, 9).
Serum levels of immunoreactive inhibin rise from ovulation to a peak at
9–10 weeks gestation, falling to a plateau at approximately 15 weeks
and thereafter rising in the third trimester to another peak at term
(5, 6, 7, 8). A similar biphasic ontogeny for inhibin A has been recently
described (24, 25). On this much larger series of MS samples than
previously reported, our findings confirm the biphasic profile, but
show that inhibin A levels rise at 18 weeks gestation, earlier than
previously reported (24). Contrary to our findings, Qu et
al. (26) showed a steady rise in bioactive inhibin levels without
the biphasic profile. It would now appear that these bioactive inhibin
data have been confounded by either other FSH-regulating peptides in
MS, such as activin (27), or by inadequate stripping of the very high
levels of circulating sex steroids that exist in pregnancy (26). Our
studies also demonstrate that inhibin B is not detectable in MS at
10–20 weeks gestation, extending the finding of a previous report that
inhibin B was absent in MS up to 11 weeks gestation (25). This is the first report of specific inhibin dimers in AF. The higher level of inhibin A in AF than in MS is consistent with existing reports of immunoreactive inhibin levels in these compartments (28, 29). It is perhaps surprising that AF levels of inhibin A are higher than those in MS because hCG, which is also secreted by the placenta, is significantly lower in AF than in serum (30, 31), consistent with secretion from the placenta into the maternal circulation. Our data suggest that either inhibin A is preferentially secreted by the placenta into the AF, rather than the maternal circulation, which is unlikely considering anatomical relationships, or that there is another significant source of inhibin during pregnancy. The lack of correlation of AF inhibin A with MS inhibin A and the differing ontogenies of inhibin A in MS and AF support the latter explanation, suggesting that different sources contribute to different compartments. Furthermore, the presence of inhibin B in AF, but not in MS, and the finding that AF levels of inhibin A and inhibin B are only weakly associated are also consistent with the presence of at least two independent sources of inhibin.
Potential sources of inhibin in pregnancy include the decidua, the
fetus, and the fetal membranes, which all express inhibin subunit mRNAs
(32, 33, 34). The decidua is an unlikely candidate, as it preferentially
expresses ßB-subunit mRNA (32), suggesting that it would
secrete more inhibin B than inhibin A, the converse of our
observations. Further, if decidually derived, it might be expected that
inhibins would be detectable in both MS and AF, similar to other
decidual products, such as PRL and placental protein 14 (30, 35, 36),
but inhibin B is not detectable in MS. It is possible that the inhibins
in AF may be fetally derived, as alphafetoprotein, which is of fetal
origin, is found in higher concentrations in AF than MS (37), similar
to inhibin A. Previous studies of immunoreactive inhibin in cord serum
concluded that the fetus was not a significant source of inhibin (5, 6, 8), and although we were unable to collect fetal blood at 10–20 weeks,
we demonstrated that there is no inhibin A in the fetal circulation at
term, which would argue against a fetal source for the inhibin A in AF.
The fetal membranes are, therefore, the likely origin of the inhibin A
in AF at the gestation periods studied. Of these, the amnion
selectively expresses ßB-subunit mRNA (34), which is more
indicative of activin production, but the chorion expresses mRNA for
both the ßA- and -subunits (34), consistent with
inhibin A secretion.
The fetal membranes are also the probable source of the inhibin B detected in AF. We found no sex differences in AF inhibin B levels, which is not consistent with our umbilical cord data or previous midgestation data for immunoreactive inhibin (38). This would argue against a significant fetal contribution to the inhibin B content of AF, although, as for inhibin A, we were unable to collect fetal blood at the earlier gestation times, and we cannot exclude the fetus as a contributor of inhibin B to AF.
In cord serum from female babies, neither inhibin A nor inhibin B was
present, but in the males, we found levels of inhibin B comparable to
those in adult men (39). This differs from a previous report that no
dimeric inhibins are present in cord serum (40). That Billiar and
colleagues (40) found no inhibin B can be explained by the capture
antibody (E4) used in their ELISA-B assay, which is an antiinhibin
ßA-subunit antibody displaying no significant affinity
for the inhibin ßB-subunit. The fetal testis expresses
mRNA for the inhibin subunits, whereas the ovary does not (32), in
keeping with inhibin B being of testicular origin, rather than from
another source common to both sexes, such as the adrenals. Therefore,
these data may be novel evidence that the human fetal testis secretes
dimeric inhibin. It is likely that the majority of the inhibin detected
in cord serum by the inhibin -subunit assays used in previous
studies (5, 6, 8) is free -subunit (40). Although there are no sex
differences in immunoreactive inhibin levels in cord serum at term (6, 8, 41), at 26–28 weeks, immunoreactive inhibin levels are higher in
males than in females (38). In light of our data, the immunoreactive
inhibin findings at earlier periods of gestation may represent relative
changes in inhibin B, particularly when at these times of gestation FSH
levels in males are significantly lower than those in females (38, 42, 43). This sex difference in circulating FSH has been previously
explained by the higher circulating levels of testosterone in males
(42, 43), but it is possible that it may at least in part be due to
inhibin B, being evidence of negative feedback by inhibin B on fetal
pituitary FSH secretion at these periods of gestation.
In summary, our data offer some clarification of the biology of inhibin in early to midpregnancy, providing evidence to challenge the concept that the placenta is the only significant source of inhibin A in pregnancy. We suggest that the placenta secretes inhibin A primarily into the maternal circulation, whereas the fetal membranes are the likely major source of both inhibin A and inhibin B in AF. Our data also show for the first time that the fetal testis may secrete inhibin B.
|
Footnotes |
|---|
2 Erasmus student from the University of Leiden.
Received June 11, 1996.
Revised September 10, 1996.
Accepted September 11, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, ßA
and ßB subunits in various tissues predicts diverse
functions. Proc Natl Acad Sci USA. 85:247–251.-subunit in human serum:
implications for radioimmunoassay. J Clin Endocrinol Metab. 70:1208–1212.[Abstract]
-ßA dimer) during human pregnancy. Clin Endocrinol (Oxf). 2:391–397.
-subunit(s). J Clin Endocrinol Metab. 80:3173–3179.[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  I. Bergada, C. Milani, P. Bedecarras, L. Andreone, M. G. Ropelato, S. Gottlieb, C. Bergada, S. Campo, and R. A. Rey Time Course of the Serum Gonadotropin Surge, Inhibins, and Anti-Mullerian Hormone in Normal Newborn Males during the First Month of Life J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 4092 - 4098. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Tong, L. Rombauts, A. Mulder, B. Marjono, J. L. Onwude, and E. M. Wallace Increased Day 15-17 Serum Pro-{alpha}C Inhibin Levels Specific to Successful Pregnancy J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4464 - 4468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. G. Zeeman, J. M. Alexander, D. D. McIntire, W. Byrd, and K. J. Leveno Inhibin-A Levels and Severity of Hypertensive Disorders Due to Pregnancy Obstet. Gynecol., July 1, 2002; 100(1): 140 - 144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Thirunavukarasu, T. Stephenson, J. Forray, P. G. Stanton, N. Groome, E. Wallace, and D. M. Robertson Changes in Molecular Weight Forms of Inhibin A and Pro-{alpha}C in Maternal Serum during Human Pregnancy J. Clin. Endocrinol. Metab., December 1, 2001; 86(12): 5794 - 5804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Pautier, S. Ghione, S. Brailly-Tabard, C. Lhomme, P. Morice, and J.M. Bidart Are serum inhibin concentrations new markers of placental tumours in the course of chemotherapy? Hum. Reprod., November 1, 2001; 16(11): 2434 - 2437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Moore, D. A. Brown, W. D. Fairlie, A. R. Bauskin, P. K. Brown, M. L. C. Munier, P. K. Russell, L. A. Salamonsen, E. M. Wallace, and S. N. Breit The Transforming Growth Factor-{beta} Superfamily Cytokine Macrophage Inhibitory Cytokine-1 Is Present in High Concentrations in the Serum of Pregnant Women J. Clin. Endocrinol. Metab., December 1, 2000; 85(12): 4781 - 4788. [Abstract] [Full Text]
|
|
|
|
 |
|
 J. A. Keelan, R. L. Zhou, L. W. Evans, N. P. Groome, and M. D. Mitchell Regulation of Activin A, Inhibin A, and Follistatin Production in Human Amnion and Choriodecidual Explants by Inflammatory Mediators Reproductive Sciences, September 1, 2000; 7(5): 291 - 296. [Abstract] [PDF]
|
|
|
|
 |
|
 F. Debieve, S. Pampfer, and K. Thomas Inhibin and activin production and subunit expression in human placental cells cultured in vitro Mol. Hum. Reprod., August 1, 2000; 6(8): 743 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. Riley, R. Leask, C. Balfour, J. E. Brennand, and N. P. Groome Production of inhibin forms by the fetal membranes, decidua, placenta and fetus at parturition Hum. Reprod., March 1, 2000; 15(3): 578 - 583. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Debieve, S. Beerlandt, C. Hubinont, and K. Thomas Gonadotropins, Prolactin, Inhibin A, Inhibin B, and Activin A in Human Fetal Serum from Midpregnancy and Term Pregnancy J. Clin. Endocrinol. Metab., January 1, 2000; 85(1): 270 - 274. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. B. Billiar, M. G. Leavitt, P. Smith, E. D. Albrecht, and G. J. Pepe Functional Capacity of Fetal Zone Cells of the Baboon Fetal Adrenal Gland: A Major Source of {alpha}-Inhibin Biol Reprod, July 1, 1999; 61(1): 142 - 146. [Abstract] [Full Text]
|
|
|
|
|
|
C. Foresta, A. Bettella, M. Rossato, G. La Sala, M. De Paoli, and M. Plebani Inhibin B plasma concentrations in oligozoospermic subjects before and after therapy with follicle stimulating hormone Hum. Reprod., April 1, 1999; 14(4): 906 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Petraglia, S. Luisi, C. Benedetto, M. Zonca, P. Florio, E. Casarosa, A. Volpe, S. Bernasconi, and A. R. Genazzani Changes of Dimeric Inhibin B Levels in Maternal Serum Throughout Healthy Gestation and in Women with Gestational Diseases J. Clin. Endocrinol. Metab., September 1, 1997; 82(9): 2991 - 2995. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Wallace, N. P. Groome, S. C. Riley, A. C. Parker, and F. C. W. Wu Effects of Chemotherapy-Induced Testicular Damage on Inhibin, Gonadotropin, and Testosterone Secretion: A Prospective Longitudinal Study J. Clin. Endocrinol. Metab., September 1, 1997; 82(9): 3111 - 3115. [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 regressed
median inhibin B levels in 189 amniotic fluid samples at 14–20 weeks
gestation (
).