localized in the human placenta, the major expression being from the ... gestations were all confirmed by ultrasound scan before commence- ment of the ...
Human Reproduction Vol.18, No.4 pp. 744±748, 2003
DOI: 10.1093/humrep/deg187
Changes in circulating concentrations of inhibins A and pro-a C during ®rst trimester medical termination of pregnancy S.Lahiri, C.J.Anobile, P.Stewart and W.L.Ledger1 Academic Unit of Obstetrics and Gynaecology, University Department of Reproductive and Developmental Medicine, The Jessop Wing, Tree Root Walk, Shef®eld S10 2SF, UK 1
To whom correspondence should be addressed. E-mail: W.Ledger@Shef®eld.ac.uk
Key words: inhibin A/pro-a C/pregnancy/medical termination
Introduction Inhibins are non-steroidal glycoproteins thought to have important roles in reproductive physiology. Inhibin A has a molecular weight of 34 kDa, and comprises an a-subunit linked by a disulphide bond to a highly homologous bAsubunit. Inhibin B is a similar dimeric glycoprotein with a and bB-subunits. Non-bioactive forms of the a-subunit include the amino-terminally extended product named inhibin pro-a C. Inhibin A is the major circulating bioactive inhibin found in early pregnancy. Inhibin B is not detectable in early pregnancy in the human (Illingworth et al., 1996; Fowler et al., 1998). Circulating levels of inhibin A and pro-a C have been implicated in the process of implantation and early pregnancy development (Muttukrishna et al., 1995). They have also been proposed as markers of fetal viability (Norman et al., 1993). In the non-pregnant female, inhibins are secreted and synthesized by both the developing Graa®an follicle and corpus luteum (Groome et al., 1994; Muttukrishna et al., 1994). The corpus luteum has been shown to be the major site of 744
inhibin production during the luteal phase in a normal menstrual cycle (Illingworth et al., 1991) and production of inhibin A continues within the corpus luteum as pregnancy is established (Illingworth et al., 1996; Rombauts et al., 1996). During early pregnancy, mRNA for a, bA and bB have been demonstrated in the corpus luteum (Roberts et al., 1993). However, inhibins are also synthesized and secreted by the developing human placenta (McLachlan et al., 1986). Both a and bA-subunit mRNAs (Meunier et al., 1988; Baird and Smith et al., 1993) and proteins (Petraglia et al., 1991) have been localized in the human placenta, the major expression being from the syncytiotrophoblast. The local actions during placental growth and differentiation are mirrored by changes in the circulating levels of dimeric inhibins and inhibin pro-a C as pregnancy progresses (Ledger, 2001). Concentrations of inhibin A in the circulation increase progressively in early pregnancy (Illingworth et al., 1996). There is a transient fall in circulating concentration at approximately 12 weeks gestation, followed by further ã European Society of Human Reproduction and Embryology
Downloaded from http://humrep.oxfordjournals.org/ by guest on June 11, 2013
BACKGROUND: Both inhibin A and inhibin pro-a C are detectable in the circulation in increasing amounts during establishment of pregnancy. However, their origins and functions remain to be elucidated. We have studied levels of inhibin A and inhibin pro-a C in serum samples collected at various stages during medical termination of pregnancy with consecutive use of mifepristone and misoprostol. METHODS: Samples were collected from three groups of patients at different weeks of gestation (group A: 6±7 weeks, n = 6; group B: 7±8 weeks, n = 6; group C: 8±9 weeks, n = 6) at the time of administration of oral mifepristone, 48 h later just before administration of vaginal misoprostol and again soon after expulsion of the products of conception. Plasma concentrations of inhibin A and pro-a C were assayed using speci®c and sensitive enzyme-linked immunosorbent assays. Results were correlated with concentrations of hCG and progesterone. RESULTS: We observed a signi®cant fall in plasma concentration of inhibin pro-a C following administration of mifepristone, which continued after administration of misoprostol. In contrast mifepristone had no effect on plasma levels of inhibin A, which fell steeply only after administration of misoprostol. CONCLUSIONS: These results suggest dissociation between major sources of inhibin A and inhibin pro-a C in early pregnancy. Treatment with mifepristone, a competitive antagonist of the progesterone receptor, resulted in a signi®cant and rapid fall in concentrations of inhibin pro-a C, identifying a link between production of pro-a C and luteal steroidogenesis. In contrast, concentrations of inhibin A did not fall after mifepristone, identifying a predominantly feto-placental origin in early human pregnancy.
Inhibins A and pro-aC during pregnancy termination Table I. Mean (6 2 SEM) hormone concentrations at the three different times of collection
Inhibin A (pg/ml) Pro-a-C (pg/ml) hCG (IU/l) Progesterone (nmol/l)
Gestation in weeks
Pre-mifepristone mean (6 2 SEM)
Pre-misoprostol mean (6 2 SEM)
Post-termination mean (6 2 SEM)
6±7 7±8 8±9 6±7 7±8 8±9 6±7 7±8 8±9 6±7 7±8 8±9
42.64 6 37.1 86.8 6 45.9 120 6 28.3 1109 6 231 776 6 375 972 6 478 24117 6 14302 61463 6 37830 95546 6 25080 60 6 12.6 45.2 6 16.2 60.4 6 15
47.49 6 26.4 85.7 6 29.6 117.6 6 43.8 1046 6 284 715 6 254 911 6 415 31464 6 13896 55403 6 25325 94279 6 29705 60.1 6 14.5 47.8 6 13.8 69.2 6 31.6
17.9 6 11.4 43.7 6 40.1 41.2 6 12.1 932 6 209 570 6 231 717 6 342 21187 6 9314 54805 6 41328 77505 6 26381 50.6 6 11.6 38 6 14.3 43.4 6 16.8
acts predominantly on steroidogenic activity within the corpus luteum. Misoprostol, a prostaglandin analogue, produces uterine contractions, interrupts placental blood ¯ow and causes expulsion of the feto-placental unit. Observation of the effects of the two agents on concentrations of inhibins allows an assessment of the relative importance of the two sources of inhibin A in early pregnancy. Materials and methods Patients and sample collection Women with singleton intrauterine pregnancies undergoing medical termination were invited to participate in this study. Informed consent was obtained from all patients. Three groups of patients were studied, each consisting of six women (n = 18): group 1, 6±7 weeks gestation; group 2, 7±8 weeks gestation; group 3, 8±9 weeks gestation. The gestations were all con®rmed by ultrasound scan before commencement of the procedure. The sample protocol was approved by the North and South Shef®eld Ethics Committees. Medical termination of pregnancy was initiated by oral administration of mifepristone (200 mg) followed 48 h later by the administration of vaginal misoprostol (200 mg) at 4 h intervals until expulsion of the fetus. Only the women who did not exhibit signi®cant vaginal bleeding after mifepristone alone were included in the study. During the procedure, three blood samples were collected: sample A, immediately before administration of mifepristone: sample B, immediately before administration of misoprostol; and sample C, soon after the expulsion of the products of conception. Blood was collected in EDTA and plasma was separated and stored at ±40°C for hormone assay. Measurement of inhibin A and pro-a C in serum Quanti®cation of inhibin A and pro-a C concentrations was performed by using commercial speci®c two-site enzyme immunoassay kits developed from the assay described by Groome et al. (1995) (dimeric inhibin A ELISA MCA1273KZZ, pro-a-C ELISA MCA1254KZZ; Oxford Bio Innovation Ltd, Upper Heyford, Oxfordshire, UK). Samples (100 ml) were treated and assayed in duplicate for dimeric inhibin A content according to the manufacturer's instructions. The kit was supplied with lyophilised inhibin A puri®ed from human follicular ¯uid, which has been calibrated against recombinant human 32 kDa inhibin. This was used as the assay standard by reconstitution followed by serial dilution with fetal calf serum to
745
Downloaded from http://humrep.oxfordjournals.org/ by guest on June 11, 2013
increases in concentration from mid-gestation onwards (Muttukrishna et al., 1995). Studies demonstrating lower levels of inhibin A in failing pregnancies have implicated inhibin A in the processes of successful implantation and early pregnancy development (Florio et al., 1995). Inhibin A may also have a role as a marker of fetal viability in IVF pregnancies (Lockwood et al., 1997). There is, however, little information about the possible functions of the inhibins during early pregnancy. Such information will follow from studies aiming to identify the source of inhibin in early pregnancy, and the effects of pregnancy disruption on inhibin production. To date, studies of early pregnancy in patients undergoing IVF with donor oocytes in whom the corpus luteum was either inactive following treatment with a GnRH analogue, or after early menopause, showed increases in circulating inhibin A (Birdsall et al., 1997; Lockwood et al., 1997) indicating the placenta as a source of inhibin A. However, although concentrations of inhibin A are increased in both normal pregnancies and in women pregnant after donor oocyte IVF, concentrations were higher in pregnancies in women with functional ovaries (Treetampinich et al., 2000), suggesting an ovarian origin for inhibin A in addition to the placenta in early pregnancy. In order to explore the relative contribution of the corpus luteum and feto±placental unit to the circulating pool of inhibin A in early pregnancy, changing concentrations of inhibin A and B during surgical termination of pregnancy have been studied. This study showed rapid decline in concentrations of inhibins following surgical evacuation of the uterus, despite maintenance of luteal progesterone secretion (Muttukrishna et al., 1997a). Circulating concentrations of inhibin A declined to ~30% of pre-termination levels by 24 h after surgery. This background concentration of inhibin A is thought to represent the contribution of the corpus luteum to the overall pool. The present study was designed to investigate the production of the corpus luteum and feto-placental unit in early pregnancy using medical termination of pregnancy as a means of dissociating luteal function from that of the feto-placental unit. Medical termination of pregnancy involves consecutive use of two agents, mifepristone and misoprostol. Mifepristone
S.Lahiri et al.
achieve a concentration range of 3.9±500 pg/ml. Quality control samples (20, 120 and 300 pg/ml) were prepared by dilution of the stock inhibin A preparation, and were employed in each assay performed. The assay sensitivity was 3.9 pg/ml and the intra- and inter-assay coef®cients of variation were
