1 Institute of Reproduction and Development, Monash University, Clayton, Victoria 3168, Australia and 2 School of Biological and Molecular Sciences, Oxford Brookes University, Oxford, UK
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: activin A/follistatin/human/pregnancy/radioimmunoassay
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The involvement of the activins in pregnancy emerged from the identification of mRNA for the , ßA and ßB subunits and dimeric proteins in the human placenta (Meunier et al., 1988
; Petraglia et al., 1991
; de Kretser et al., 1994
; Yokoyama et al., 1995
) and fetal membranes (Petraglia et al., 1993a
). High concentrations of activin A have been shown in ovine (Wongprasartsuk et al., 1994
) and human (Petraglia et al., 1993b
) amniotic fluid. With the development of a variety of more sensitive and specific assays for the measurement of activin A, elevated concentrations in human pregnancy were demonstrated (Petraglia et al., 1993b
, 1994
; Harada et al., 1996
; Woodruff et al., 1997
). A subsequent study with a larger cohort of women confirmed these findings (Muttukrishna et al., 1996
). In early pregnancy, activin A concentrations have been shown to be higher than those of cycling women, and studies in women with non-functional ovaries have suggested a feto-placental origin for activin A (Birdsall et al., 1997
; Lockwood et al., 1997
; Muttukrishna et al., 1997a
). Moreover, higher than normal activin A concentrations have been demonstrated in gestational diseases (Petraglia et al., 1995a
,b
), premature labour (Petraglia et al., 1997
) and in pre-eclampsia (Muttukrishna et al., 1997b
).
The role of activin in various biological processes is modulated by follistatin, a high-affinity activin-binding protein, which can neutralize the majority of the actions of activin (Nakamura et al., 1990; Kogawa et al., 1991
). Follistatin exists as two forms termed follistatin 288 (FS288) and the larger form, FS315, which arise through an alternative splicing action (Shimasaki et al., 1988
) and a number of other forms which arise from proteolytic cleavage and glycosylation variants (Sugino et al., 1993
).
Follistatin has been isolated from human placenta (de Kretser et al., 1994) and is also found in fetal membranes, decidua and amniotic fluid (Petraglia et al., 1994
; Wongprasartsuk et al., 1994
). Recently, serum concentrations of follistatin have been shown to rise throughout pregnancy, although the numbers of subjects in the studies were limited (Wakatsuki et al., 1996
; Woodruff et al., 1997
).
The high-affinity binding of activin by follistatin can affect the accuracy of the assays for either protein since the formation of the complex between activin and follistatin can interfere in the binding of the proteins to the antisera (de Kretser et al., 1994; McFarlane et al., 1996
). This has led to the use of a variety of methods to remove this interference such as analyte denaturation (Knight et al., 1996
), dissociating agents or detergents (McFarlane et al., 1996
; Evans et al., 1998
) and sample extraction (Harada et al., 1996
) to enable the measurement of total activin or follistatin.
This study utilizes two assays which address the complexities of the measurement of total activin and follistatin, to determine the concentrations of these proteins in a large cohort of pregnant women in whom the outcome of the pregnancy was normal.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Longitudinal study:
Sequential samples were taken at 2- to 4-weekly intervals from pregnant women who volunteered at their first ante-natal clinic visit. Nine women whose pregnancy satisfied the above criteria were included in the study.
Hormone assays
Activin A enzyme-linked immunosorbent assay (ELISA):
Total serum activin A concentrations were measured using a two-site enzyme immunoassay specific for activin A as previously described (Knight et al., 1996). Human recombinant (hr) activin A, purified as described previously (Robertson et al., 1992
) from material provided by Biotech Australia Pty Ltd (East Roseville, NSW, Australia), was used as standard. The mean intra- and inter-assay coefficients of variation (CV, %) were 9.8% and 12.3% respectively. The minimum detection limit was 0.05 ng/ml.
Follistatin radioimmunoassay:
Total follistatin concentrations were measured using a heterologous, discontinuous radioimmunoassay. The assay used a rabbit antiserum (#204) raised against purified 35 kDa bovine follistatin and human recombinant follistatin 288 (hrFS288; provided by the National Institute of Diabetes, Digestive and Kidney Disease, Bethesda, MD, USA) was used as standard. [125I]-hrFS288 was used as tracer, iodinated using Iodogen reagent (Pierce, Rockford, IL, USA). Dissociating reagents were used to remove the interference of activin in the assay. A combination of reagents (termed TDS reagent) at final concentrations of 6.67% Tween-20 (Sigma, St Louis, MO, USA), 3.33% sodium deoxycholate (BDH, Poole, Dorset, UK) and 0.13% sodium dodecyl sulphate (SDS; BioRad, Hercules, CA, USA) was used (modified from McFarlane et al., 1996). Immune complex precipitation was achieved by the use of a goat anti-rabbit second antibody (GAR#11; IRD, Monash University). The total serum concentration per tube was equalized using normal human serum (provided by the Australian Red Cross Blood Bank, South Melbourne, Victoria, Australia) immediately after addition of the second antibody.
Radioimmunoassay procedure
The assay buffer used was 0.1 M phosphate-buffered saline (PBS), pH 7.4 containing 0.5% bovine serum albumin (BSA; Sigma). Standard and samples (100µl) were diluted in assay buffer and incubated with 100 µl antiserum #204 (1/8000 diluted in assay buffer containing 1/1400 normal rabbit serum, NRS; IRD, Monash University) and 100 µl TDS reagent overnight at room temperature. Tracer was added at 10 000 c.p.m./100 µl of assay buffer containing 0.1% Triton X-100 (v/v) (Sigma) on day 2 and incubation continued overnight at room temperature. On day 3, 100 µl of second antibody GAR#11 (1/90) in assay buffer containing 0.05 M EDTA (BDH) was added to precipitate immune complexes and the total serum concentration per tube was equalized at the same time by adding 100 µl of serum/buffer to each tube to give a final concentration of 100 µl serum per tube. Following incubation overnight at 4°C, 2 ml cold 0.9% saline was added and the tubes were centrifuged at 4000 g for 45 min at 4°C. The supernatant was decanted and the pellets counted on a gamma counter.
The sensitivity of the radioimmunoassay, based on a +2 logit value, was 1.4 ng/ml with a working range of 1.4 to 110 ng/ml. The intra-assay coefficient of variation based on the mean CV from six assays was 5.7% and the inter-assay CV based on a quality control pool of normal human serum was 9.5%.
Radioimmunoassay validation:
Samples of male and female human serum were serially diluted and assessed for their dose-dependence and parallelism to the standard. Cross-reactivity of hrFS315 was determined using material kindly provided by Dr H.Sugino, Institute for Enzyme Research, University of Tokushima, Japan. The accuracy of the assay was tested by measuring serum samples `spiked' with known amounts of hrFS288 (3.125100 ng/ml) to assess quantitative recoveries. Interference in the radioimmunoassay from activin was assessed by pre-incubating increasing amounts (050 ng/tube) of hr activin A with normal human serum and assaying these with and without the TDS reagent in the radioimmunoassay.
Data analysis
Doseresponse curves were linearized using logit log-dose transformation and analysed by parallel line statistics. Data from the cross-sectional study were grouped into 2-week intervals and represented as arithmetic mean ± SEM. For analysis, the data were log-transformed and analysed using a one-way analysis of variance (ANOVA) with Bonferroni post test to determine significant differences in concentrations between different stages of gestation. Correlations were calculated using Pearson correlation coefficients. Analyses were carried out using the GraphPad software package (GraphPad Software Inc., San Diego, CA, USA).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
In the absence of the dissociating reagent TDS, follistatin was undetectable in male human serum (Figure 2). Without dissociation, only free follistatin, and a small amount of complexed follistatin possibly where the epitope is exposed, can potentially be measured. It therefore appears that the majority of follistatin in serum exists in a complex with activin and as such is undetectable. However, in the presence of the TDS reagent, a level of 7.1 ng/ml of follistatin was measured in the serum. This was due to the dissociation of follistatin from the complex with activin, thereby allowing the measurement of `total' follistatin in serum.
|
Activin A and follistatin in pregnancy
Cross-sectional study
Mean serum activin A concentrations rose 69-fold (P < 0.001) throughout pregnancy from 0.07 ± 0.02 ng/ml at weeks 67 to a peak of 4.59 ± 0.54 ng/ml at 3839 weeks (Figure 3). Activin A concentrations remained detectable but stable from 6 to 14 weeks (P > 0.05), but then showed a small but significant rise at 2223 weeks (P < 0.05). Subsequently, there was a 6.1-fold increase (P < 0.001) from weeks 2425 to 3031 weeks where concentrations plateaued slightly and then rose steeply from weeks 3233 to a peak at 3839 weeks (3.0-fold, P < 0.001).
|
Longitudinal study
In all nine women followed longitudinally, activin and follistatin concentrations rose in parallel, beginning in the third trimester and rising to a peak towards term (Figure 4ai). The activin A and follistatin concentrations in each woman were highly correlated across pregnancy in all nine women (r = 0.8670.998, P < 0.05 to < 0.0001). The concentrations of follistatin and activin A during pregnancy in these nine women fell within the normal range established by the cross-sectional study (Figure 5a and b
), except for one woman whose activin A concentrations were higher than the normal range until week 30 when the concentrations remained within the normal range until term, while her follistatin concentrations fell within the normal range.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The assays used in this study have been carefully characterized and provide measurements of the total activin A and total follistatin. While activin does not interfere in the follistatin radioimmunoassay, it is recognized that the concentrations of follistatin may be an underestimate since cross-reactivity of hrFS315, which probably represents the major circulating form, is only 36% with reference to the hrFS288 assay standard. However, the results described here in the last trimester of pregnancy are in fact only slightly lower than the results of Wakatsuki et al. (1996) who appear to have the only assay reported which is directed specifically to the measurement of FS315.
The activin A ELISA used in this study is not compromised by the presence of follistatin (Knight et al., 1996), in contrast to a number of other assays reported in the literature. However, the activin A concentrations are significantly less than those shown in the study by Muttukrishna et al. (1996) using the same ELISA with a different standard preparation. They report concentrations of 25 ng/ml (compared with 4.6 ng/ml) at the end of the third trimester, but the patterns of activin A concentrations across pregnancy are similar.
There is a need for the establishment of international standards for the measurement of these proteins in conditions where the concentrations may be of clinical importance. This is of particular importance for the measurement of activin A since our studies of this protein in allantoic fluid have shown that preparations of activin A may run with different mobilities on high-performance liquid chromatography and reflect significantly differing potencies when measured in bioassays, ELISA and radioimmunoassay, all of which were proven to be specific for this protein (Foulds et al., 1998).
The principal finding in this study is a parallel rise in the circulating concentrations of total follistatin and activin A as pregnancy progresses, confirming and extending earlier data (Petraglia et al., 1993b; Harada et al., 1996
; Muttukrishna et al., 1996
; Woodruff et al., 1997
). The peripheral concentrations of follistatin and activin A are highly correlated throughout pregnancy, but there is a greater increase in follistatin concentrations in the second trimester and a steeper rise in activin A in the third trimester. The significance of this observation is unclear, as are the mechanisms through which the increases occur. The principal source of activin A and follistatin during pregnancy is likely to be the feto-placental unit, since these proteins have been isolated from the term placenta (de Kretser et al., 1994
; Yokoyama et al., 1995
) and activin A concentrations in maternal blood decline rapidly in the puerperium (Harada et al., 1996
; Petraglia et al., 1997
). Further, activin A concentrations were significantly higher in multiple pregnancies, wherein the placental mass is greater, compared with singleton pregnancies (Lockwood et al., 1997
). Total activin A concentrations show a rapid decline 4 h after surgical termination of pregnancy (Muttukrishna et al., 1997a
). However, the activin A concentrations had not returned to luteal phase concentrations, suggesting the ovary as a possible source for some of the activin in early pregnancy, or a low clearance rate of activin.
The specific function of the elevated concentrations of activin A during pregnancy remains a matter of conjecture. Given the capacity for activin to cause immunomodulation, the elevated concentrations arising from a placental source may form part of the immune mechanisms protecting the fetus. In this regard, there is evidence that the early pregnancy trophoblast and fetal membranes express the ßA subunit and may be related to the establishment of the placenta (Petraglia et al., 1993a). In this context, there are increasing data to support the view that follistatin may be elevated as part of the body's acute-phase reaction as it can be controlled by the cytokine interleukin-1ß (Phillips et al., 1996
) and can be stimulated by lipopolysaccharide (Klein et al., 1996
; Michel et al., 1996
). Given the very significant stress and tissue remodelling that occurs in the postpartum period, the continued elevation of follistatin concentrations in this period (Wakatsuki et al., 1996
) may well be driven by these mechanisms. The greater elevation of both proteins in the second and third trimesters probably reflects placental secretion, but whether they have a role to play in the mother is unclear. Given that the rise in serum follistatin concentrations occurs earlier than activin, this may represent a protective measure since the pleiotrophic actions of activin could affect a large number of physiological processes throughout the body. The capacity of follistatin to bind and neutralize the actions of activin make such a role possible, and recent data indicate that this binding may direct the follistatinactivin complex into the cell and to a lysosomal pathway (Hashimoto et al., 1997
). Additionally, previous studies have shown that the fast form of
2-macroglobulin can bind both activin A and follistatin (Phillips et al., 1997
) and may serve to direct the complex into a degradative pathway through the lipoprotein receptor pathway.
Depending on the balance between the relative concentrations of activin and its binding protein, follistatin, it is possible that differences in free activin concentrations between women may be reflected in differences in the timing of the cascade of events that initiate parturition. In this context, Draper et al. (Draper et al., 1997) have shown that uterine smooth muscle represents a binding site for labelled activin A injected into the circulation of rats in late gestation, and they suggested that this may be involved in the initiation of uterine contractility. Further studies are necessary to determine the role of these proteins in pregnancy.
In summary, the results presented in this study provide total follistatin and activin A concentrations in the circulation of a large cohort of patients during a normal pregnancy and provide the basis for the exploration of patterns of secretion of these proteins during abnormal pregnancy.
![]() |
Acknowledgments |
---|
![]() |
Notes |
---|
4 Present address: Pfizer Pharmaceuticals Inc., Shinjuku-ku, Tokyo 163-0461, Japan
5 To whom correspondence should be addressed
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
de Kretser, D.M., Foulds, L.M., Hancock, M. et al. (1994) Partial characterization of inhibin, activin, and follistatin in the term human placenta. J. Clin. Endocrinol. Metab., 79, 502507.[Abstract]
Draper, L.B., Chong, H., Wang, E. et al. (1997) The uterine myometrium is a target for increased levels of activin A during pregnancy. Endocrinology, 138, 30423046.
Evans, L.W., Muttukrishna, S. and Groome, N.P. (1998) Development, validation and application of an ultra-sensitive two-site enzyme immunoassay for human follistatin. J. Endocrinol., 156, 275282.
Foulds, L.M., de Kretser, D.M., Farnworth, P. et al. (1998) Ovine allantoic fluid contains high concentrations of activin A: partial dissociation of immunoactivity and bioactivity. Biol. Reprod., 59, 233240.
Harada, K., Shintani, Y., Sakamoto, Y. et al. (1996) Serum immunoreactive activin A levels in normal subjects and patients with various diseases. J. Clin. Endocrinol. Metab., 81, 21252130.[Abstract]
Hashimoto, O., Nakamura, T., Shoji, H. et al. (1997) A novel role of follistatin, an activin-binding protein, in the inhibition of activin action in rat pituitary cells. J. Biol. Chem., 272, 1383513842.
Klein, R., Clarke, I.J., Hedger, M.P. et al. (1996) Plasma follistatin concentrations increase following lipopolysaccharide administration in sheep. Clin. Exp. Pharm. Physiol., 23, 754755.[ISI][Medline]
Knight, P.G., Muttukrishna, S. and Groome, N.P. (1996) Development and application of a two-site enzyme immunoassay for the determination of `total' activin-A concentrations in serum and follicular fluid. J. Endocrinol., 148, 267279.[Abstract]
Kogawa, T., Nakamura, T., Sugino, H. et al. (1991) Activin-binding protein is present in pituitary. Endocrinology, 128, 14341440.[Abstract]
Ling, N., Ying, S.-Y., Ueno, N. et al. (1986) Pituitary FSH is released by a heterodimer of the ß-subunits from the two forms of inhibin. Nature, 321, 779782.[ISI][Medline]
Lockwood, G.M., Ledger, W.L., Barlow, D.H. et al. (1997) Measurement of inhibin and activin in early human pregnancy: demonstration of fetoplacental origin and role in prediction of early-pregnancy outcome. Biol. Reprod., 57, 14901494.[Abstract]
Mason, A.J., Hayflick, J.S., Ling, N. et al. (1985) Complementary DNA sequences of ovarian follicular fluid inhibin show precursor structure and homology with transforming growth factor-ß. Nature, 315, 659663.[ISI]
Mather, J.P., Moore, A. and Li, R.-H. (1997) Activins, inhibins, and follistatins: further thoughts on a growing family of regulators. Proc. Soc. Exp. Biol. Med., 215, 209222.[Abstract]
McFarlane, J., Foulds, L., Piscotta, A. et al. (1996) Measurement of activin in biological fluids by radioimmunoassay, utilizing dissociating agents to remove the interference of follistatin. Eur. J. Endocrinol., 134, 481489.[ISI][Medline]
Meunier, H., Rivier, C., Evans, R.M. et al. (1988) Gonadal and extragonadal expression of inhibin, ßA and ßB subunits in various tissues predicts diverse functions. Proc. Natl Acad. Sci. USA, 85, 247251.[Abstract]
Michel, U., Schneider, O., Kirchho, F. et al. (1996) Production of follistatin in porcine endothelial cells: differential regulation by bacterial compounds and the synthetic glucocorticoid RU 28362. Endocrinology, 137, 49254934.[Abstract]
Muttukrishna, S., Fowler, P.A., George, L. et al. (1996) Changes in peripheral serum levels of total activin A during the human menstrual cycle and pregnancy. J. Clin. Endocrinol. Metab., 81, 33283334.[Abstract]
Muttukrishna, S., Childs, T.G., Groome, N.P. et al. (1997a) Source of circulating levels of inhibin A, pro alpha C-containing inhibins and activin A in early pregnancy. Hum. Reprod., 12, 10891093.[ISI][Medline]
Muttukrishna, S., Knight, P.G., Groome, N.P. et al. (1997b) Activin A and inhibin A as possible endocrine markers for pre-eclampsia. Lancet, 349, 12851288.[ISI][Medline]
Nakamura, T., Takio, K., Eto, Y. et al. (1990) Activin-binding protein from rat ovary is follistatin. Science, 247, 836838.[ISI][Medline]
Petraglia, F., Garuti, G.C., Calza, L. et al. (1991) Inhibin subunit in human placenta: localization and messenger ribonucleic acid levels during pregnancy. Am. J. Obstet. Gynecol., 165, 750758.[ISI][Medline]
Petraglia, F., Anceshi, M.M., Calza, L. et al. (1993a) Inhibin and activin in human fetal membranes: evidence for a local effect on prostaglandin release. J. Clin. Endocrinol. Metab., 77, 542548.[Abstract]
Petraglia, F., Garg, S., Florio, P. et al. (1993b) Activin A and activin B measured in maternal serum, cord blood serum, and amniotic fluid during pregnancy. Endocrine J., 1, 323327.
Petraglia, F., Gallinelli, A., De Vita, D. et al. (1994) Activin and parturition: changes of maternal serum levels and evidence for binding sites in placenta and fetal membranes. Obstet. Gynecol., 84, 278282.[Abstract]
Petraglia, F., Aguzzoli, L., Gallinelli, A. et al. (1995a) Hypertension in pregnancy: changes in activin A maternal serum concentration. Placenta, 16, 447454.[ISI][Medline]
Petraglia, F., De Vita, D., Gallinelli, A. et al. (1995b) Abnormal concentration of maternal serum activin A in gestational diseases. J. Clin. Endocrinol. Metab., 80, 558561.[Abstract]
Petraglia, F., Di Blasio, A.A., Florio, P. et al. (1997) High levels of fetal membrane activin ßA and activin receptor IIB mRNAs and augmented concentration of amniotic fluid activin A in women in term or preterm labor. J. Endocrinol., 154, 95101.[Abstract]
Phillips, D.J., Hedger, M.P., McFarlane, J.R. et al. (1996) Follistatin concentrations in male sheep increase following sham castration/castration or injection of interleukin-1ß. J. Endocrinol., 151, 119124.[Abstract]
Phillips, D.J., McFarlane, J.R., Hearn, M.T. et al. (1997) Inhibin, activin and follistatin bind preferentially to the transformed species of -2 macroglobulin. J. Endocrinol., 155, 6571.
Robertson, D.M., Foulds, L.M., Prisk, M. et al. (1992) Inhibin/activin ß-subunit monomer: isolation and characterization. Endocrinology, 130, 16801687.[Abstract]
Shimasaki, S., Koga, M., Esch, F. et al. (1988) Primary structure of the human follistatin precursor and its genomic organization. Proc. Natl Acad. Sci. USA, 85, 42184222.[Abstract]
Sugino, K., Kurosawa, N., Nakamura, T. et al. (1993) Molecular heterogeneity of follistatin, an activin-binding protein: higher affinity for the carboxyl-terminal truncated forms for heparan sulfate proteoglycans on the ovarian granulosa cell. J. Biol. Chem., 268, 1557915587.
Vale, W., Rivier, C., Vaughan, J. et al. (1986) Purification and characterization of a FSH releasing protein from porcine ovarian fluid. Nature, 321, 776779.[ISI][Medline]
Wakatsuki, M., Shintani, Y., Abe, M. et al. (1996) Immunoradiometric assay for follistatin: serum immunoreactive follistatin levels in normal adults and pregnant women. J. Clin. Endocrinol. Metab., 81, 630634.[Abstract]
Wongprasartsuk, S., Jenkin, G., McFarlane, J.R. et al. (1994) Inhibin and follistatin concentrations in fetal tissues and fluids during gestation in sheep: evidence for activin in amniotic fluid. J. Endocrinol., 141, 219229.[Abstract]
Woodruff, T.K., Sluss, P., Wang, E. et al. (1997) Activin A and follistatin are dynamically regulated during human pregnancy. J. Endocrinol., 152, 167175.[Abstract]
Yohkaichiya, T., Polson, D., O'Connor, A. et al. (1991) Concentrations of immunoactive inhibin in serum during human pregnancy: evidence for an ovarian contribution. Reprod. Fertil. Steril., 3, 671678.
Yokoyama, Y., Nakamura, T., Nakamura, R. et al. (1995) Identification of activins and follistatin proteins in human follicular fluid and placenta. J. Clin. Endocrinol. Metab., 80, 915921.[Abstract]
Submitted on June 8, 1998; accepted on December 7, 1998.