| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Features |
Reproductive Medicine Unit, Department of Obstetrics and Gynecology, The University of Adelaide, The Queen Elizabeth Hospital, Adelaide, South Australia 5011, Australia
Address all correspondence and requests for reprints to: Mr. Fred Amato, Reproductive Medicine Unit, Department of Obstetrics and Gynecology, The Queen Elizabeth Hospital (DX465216), Woodville Road, Woodville, South Australia 5011, Australia.
Abstract
An anti-hCG autoantibody was found in a patient with a 9-yr history of secondary infertility. Although the patient had regular menstrual cycles, had conceived spontaneously, and had good hormonal and follicular responses to gonadotropic stimulation regimens during the in vitro fertilization work-up, she presented with apparent recurrent pregnancy loss associated with prolonged raised hCG levels. Initially the presence of a high mol wt hCG complex was demonstrated in the serum by gel chromatography. The binding of [125I]recombinant hCG to a serum sample and subsequently to the affinity-purified IgG from the same sample revealed the presence of an hCG antibody. The antiserum was shown to be specific, with a low affinity (Ka, 1.4 x 106 liters/mol), but a high capacity (418 nmol/liter), for hCG. Cross-reaction with recombinant human FSH, recombinant human LH, hCG
, and hCGß were low (<0.019%, 0.021%, 0.039%, and 0.006%, respectively). In addition, heat-inactivated serum and the affinity-purified IgG were shown to inhibit the action of hCG in an in vitro bioassay. We suggest that the persisting titer of the antibody to be responsible for the patient’s infertility.
HCG IS PRODUCED mainly by the syncytiotrophoblast cells of the periimplantation embryo and, to a lesser extent, by the pituitary gland and is a heterodimer composed of a noncovalently linked -subunit (hCG) and a ß-subunit (hCGß). It is structurally related to the pituitary hormones FSH, LH, and TSH, with a common -subunit and a specific ß-subunit. hCG is considered to play an essential role in the establishment and maintenance of early pregnancy.
hCG antibodies have been found under a number of circumstances. 1) Vaccination against hCG has been shown to block fertility in the marmoset, the baboon, and the rhesus monkey (1). In addition, clinical trials have demonstrated that immunization of women with either hCGß, a heterospecies dimer of hCGß associated with the -subunit of ovine LH, or the C-terminal region of hCGß resulted in the development of contraceptive levels of anti-hCG antibodies (2, 3, 4). The study of Talwar et al. (3) also provided evidence that the circulating anti-hCG antibodies prevented pregnancy in 1224 cycles, and that booster injections were required to maintain antibody titers. Fertility was restored when titers (neutralization capacity of the antibody) dropped below 35 ng hCG/ml of serum. 2) hCG antibodies have been detected in young males who had been treated with exogenous urinary hCG (5, 6). 3) The presence of naturally occurring antibodies to hCG/LH has also been reported in young women. Wass et al. (7) described the presence of antibodies to hCG/LH in normal sera, but these affinities were too low to affect biological activity. Subsequently, a woman has been reported with a history of spontaneous abortion, followed by an anovular period and then normal ovulatory cycles and variations in hCG/LH titers in her serum (8). The patient had no previous exposure to hormone therapy.
We describe a case of a 30-yr-old woman with unexplained infertility who has developed a neutralizing antibody to hCG that appeared to be maintained by repeated exposure to exogenous and endogenous hCG. This interfered with in vitro fertilization (IVF) treatment by blocking the action of exogenous hCG.
Case Report
The patient, at the age of 16 yr, had a spontaneous pregnancy that was uneventful, with a normal vaginal delivery. After a 3-yr period of contraception, the patient tried over several years to become pregnant. During infertility investigations she was shown to be ovulating and to have a normal uterus and tubal patency, and her partner had a satisfactory semen analysis. The patient underwent three cycles of stimulated IVF using GnRH agonist, human menopausal gonadotropin, and hCG (10,000 IU). On each occasion large numbers of follicles developed with appropriate serum concentrations of E2. No oocytes were ever obtained, and the aspirates consisted of only clear fluid with healthy granulosa cell clumps. The patients were advised that they should consider the use of donor oocytes. This was not acceptable, and they contacted the Reproductive Medicine Unit at The University of Adelaide.
After the IVF cycles there was documented evidence of at least seven occasions when the hCG test for pregnancy had been positive. At no stage was a fetus seen, and the patient was considered to have had recurrent miscarriages. However, hCG levels were reported to have persisted for several months postmiscarriage. Further analyses revealed a low titer of anticardiolipin IgM, a normal level of activated protein C, and a borderline level of 64% for protein S. The total protein electrophoretic pattern appeared normal by visual assessment, and only the IgM level was found to be elevated (3.61 g/liter; normal range, 0.5–3.0 g/liter). The patient was followed over one menstrual cycle and was shown to ovulate on the 19th d of the cycle with an excellent progesterone response, but elevated levels of LH (21–26 IU/liter). On the next cycle a single large follicle was seen, and hCG (10,000 IU) was administered to the patient. The pregnancy test was positive 14 d later, but hCG slowly declined over several weeks. The patient subsequently underwent several intrauterine insemination cycles with mild stimulation of the ovaries with recombinant FSH (Puregon, 50 IU; N.V. Organon, Oss, The Netherlands). On each occasion hCG (5,000 IU) was injected, and a positive pregnancy test was obtained, but with declining hCG levels over several weeks. On all occasions the patient had a menstrual period at the expected time. In subsequent intrauterine insemination cycles no hCG was used, and spontaneous ovulation occurred. Pregnancy tests were always negative. At this stage the patient was investigated for hCG antibodies.
Materials and Methods
Reagents
The transformed Leydig cell line, MA10, was donated by Dr. M. Ascoli (University of Iowa, Ames, IA). Recombinant human CG (r-hCG; Ovidrel) and recombinant human LH (r-hLH, Luveris) were donated by Serono Australia Pty. Ltd. (Frenchs Forrest, Australia). Recombinant human FSH (r-hFSH, Gonal-F) was purchased from Serono Australia Pty. Ltd. hCGß, First International Reference Preparation 75/551, was obtained from the National Institute for Biological Standards (London, UK). The free - and ß-subunits of hCG were provided by the National Hormone and Pituitary Program of the NIDDK, NIH, University of Maryland School of Medicine (Baltimore, MD). [125I]r-hCG was prepared by the chloramine-T method, with resultant specific activities of 121–150 kBq/pmol.
hCG immunoassay
hCG was measured in serum or plasma with a total hCGß immunometric assay, using the Vitros Total ß-hCG Reagent Pack (Ortho-Clinical Diagnostics, Amersham Pharmacia Biotech, Little Chalfont, UK) on the automated Vitros Immunodiagnostic System. According to the manufacturer, cross-reactivities of TSH (250 mIU/liter), LH (400 IU/liter), and FSH (400 IU/liter) were less than 5%, less than 5%, and less than 10%, respectively. Urinary hCG was measured with the same assay after the specimen was concentrated 30-fold by ultrafiltration (3-kDa molecular mass cut-off Omega membrane, Pall Gelman Laboratory, Sydney, Australia). Results were confirmed by the Abbott ARCHITECT total ß-hCG assay (Abbott Laboratories, Chicago, IL).
IgG purification
Serum (3 ml), with hCG levels less than 5 IU/liter, was loaded onto a protein G column (1 ml; HiTrap, Amersham Pharmacia Biotech, Uppsala, Sweden), preequilibrated and run with 20 mM sodium phosphate, pH 7.0, at a flow rate of 0.5 ml/min. The column was washed with the same buffer (5 ml), and subsequently the IgG was eluted with 100 mM glycine-HCl, pH 2.7. Fractions with IgG were pooled, concentrated, and desalted using a Macrosep centrifugal concentrator, with a 10-kDa molecular mass cut-off membrane (Pall Filtron, Northborough, MA). IgG was quantitated with the Micro BCA Protein Assay (Pierce Chemical Co., Rockford, IL), using BSA as standard.
Gel filtration
Serum (1 ml) with endogenous hCG was subjected to gel filtration on a Superdex 75 column (1.6 x 60 cm; Amersham Pharmacia Biotech), preequilibrated and run with 150 mM ammonium bicarbonate at a flow rate of 1 ml/min. One-milliliter fractions were collected, desalted, lyophilized, and reconstituted in 250 µl 50 mM phosphate buffer, pH 7.4, with 150 mM NaCl and 0.5% BSA. Fractions were assayed for hCG immunoreactivity. Sera (200 µl) or IgG (200 µl; 8.5 mg), incubated with 12 µl [125I]r-hCG (700,000 cpm) for 18 h at 4 C, were subjected to gel filtration on a Superose 12 column (1 x 30 cm; Amersham Pharmacia Biotech), preequilibrated and run with the same buffer, as described above, at a flow rate of 0.5 ml/min. Fractions were subsequently assessed for [125I]r-hCG activity.
Electrophoresis and Western blotting
Serum proteins and IgG, purified from the same serum were resolved by discontinuous SDS-PAGE in 6% gels under nonreducing conditions by the method of Laemmli (9). Proteins were then transferred to nitrocellulose membrane (0.4 µm; Bio-Rad Laboratories, Inc., Regent Park, Australia) using the procedure described by Towbin et al. (10). Mol wt was determined by comparison with prestained Precision Protein Standards (Bio-Rad Laboratories, Inc.). The IgG bands were identified by incubating the membrane (preblocked with 5% BSA) with antihuman IgG, conjugated to horseradish peroxidase (1:20,000; Amersham Pharmacia Biotech) for 1 h at room temperature, treated with enhanced chemiluminescence (ECL) reagents (Amersham Pharmacia Biotech), and exposed to Hyperfilm ECL (Amersham Pharmacia Biotech). The ability of the resolved proteins to bind [125I]r-hCG was assessed by incubating the nitrocellulose membrane with [125I]r-hCG (80,000 cpm/ml) in PBS-BSA for 18 h at 4 C, followed by exposure to Hyperfilm MP (Amersham Pharmacia Biotech) for 72 h at -80 C.
Bioassay
The capacity of the patient’s serum or IgG to neutralize the hCG bioactivity was determined by an in vitro bioassay using the transformed murine Leydig cell line MA10 (11), as previously described by Simula et al. (12). Briefly, standards (r-hCG; 100 µl) or r-hCG, pretreated with various amounts of serum or IgG, were incubated with 100 µl cell suspension (0.5 x 105 cells/well) in a microtiter plate (Costar, Cambridge, MA) for 2 h at 37 C in 5% CO2/95% air. Progesterone produced and secreted into the medium was measured using an RIA kit from Diagnostics Systems Laboratories, Inc. (Webster, TX). The bioactivities of all samples were assessed within the same assay to avoid any interassay variation.
Antiserum characterization
The titer of the antibody was assessed by incubating serum dilutions (50 µl) for 18 h at 4 C with [125I]r-hCG (30,000 cpm; SA, 121 kBq/pmol) and 0.015 M phosphate, pH 7.4, with 0.15 M NaCl, 0.05% BSA, and 0.05% sodium azide in a total assay volume of 200 µl. IgG-bound [125I]r-hCG was precipitated with 25% polyethylene glycol 6000 in PBS (500 µl) and separated by centrifugation, followed by aspiration of supernatant. The amount of antisera that bound 50% of the [125I]r-hCG was used in the RIA to determine the antisera specificity and affinity. The specificity of the antisera was assessed by comparison of the displacement of [125I]r-hCG by increasing doses of r-hCG, free hCG, free hCGß, r-hFSH, and r-hLH. Data from r-hCG dose-response curve were used for determining the binding capacity and affinity of the antiserum by Scatchard analysis (13).
Results
hCG was shown to be present in the serum, as judged by the detection in two different total hCGß assays (the Vitros and the Abbott ARCHITECT) and parallelism of standard and sample dilutions in the assay. Although hCG was detected in serum, none was detected in a 24-h urine specimen collected at the same time.
The elution profile from Superdex 75 of hCG immunoreactivity of serum from the patient was compared with that of a normal pregnant subject with an equivalent level of hCG. The results illustrated in Fig. 1
demonstrate that the hCG immunoreactive peak from the patient serum eluted in fractions 41–51, with an apparent molecular mass greater than 70 kDa (beyond the separating range of the column) and greater than the hCG immunoreactivity of the control subject, which eluted in fractions 52–65 (apparent molecular mass, 67 kDa). To examine the possibility that hCG was bound to a serum protein, sera collected from both the patient and a control subject were incubated with [125I]r-hCG for 18 h at 4 C and chromatographed on a Superose 12 column. The results in Fig. 2A show that the radioactive peak of [125I]r-hCG eluted earlier (peak of activity in fraction 23) when incubated with the patient’s serum compared with when incubated with control serum (peak of activity in fractions 28–29). Similar results were obtained when IgG purified from the same sera was treated the same way (Fig. 2B), suggesting the presence of an anti-hCG antibody. This was also supported by the results of the SDS-PAGE and Western blot analysis. The resolution of the sera proteins and their corresponding affinity-purified IgG are shown in Fig. 3A. Immunodetection of the IgG bands in the sera and the purified IgG after SDS-PAGE and transfer to a nitrocellulose membrane showed them to have molecular masses ranging from 112–190 kDa (Fig. 3B). Bands with molecular masses ranging from 112–138 kDa, corresponding to a portion of the IgG bands in the serum and the purified IgG of the patient, were shown to bind [125I]r-hCG (Fig. 3C, lanes 1 and 3, respectively). In comparison, neither control serum (Fig. 3C, lane 2) nor control IgG (Fig. 3C, lane 4) bound to [125I]r-hCG.
|
|
|
) or IgG purified from the patient’s serum (Fig. 4B). The heat-inactivated serum and IgG were added to r-hCG (final concentration, 20,680 IU/liter) before addition to MA10 cells. Complete inhibition was achieved with less than 12.5 µl serum or 880 µg/ml (176 µg/well) total IgG. However, heat-inactivated serum and IgG from a control subject had no inhibitory effect.
|
, and hCGß. The cross-reactivities of r-hFSH, r-hLH, hCG, and hCGß were less than 0.019%, 0.021%, 0.039%, and 0.006% on a molar basis, respectively. Data from the r-hCG dose-response binding curve, when plotted according to the Scatchard method (Fig. 5), indicated an antiserum with a low affinity for r-hCG (Ka, 1.4 x 106 liter/mol), but a high binding capacity of 418 nmol/liter.
|
Conflicting reports have described the presence (7) and absence (14) of hCG antibodies in normal human sera. The study by Johnson et al. (14) also demonstrated the absence of such antibodies in sera from a group of infertile women. Although several researchers have suggested that autoimmunity to gonadotropins may be the cause of gonadal failure (15), we are aware of only one case that describes the presence of autoantibodies to hCG/LH that were suggested to be the cause of infertility in a 32-yr-old woman (8). In this case the patient had no previous exposure to hormone therapy and, after a spontaneous abortion, presented with antibodies of high affinity for hCG (Ka, >2 x 1010 liter/mol) and LH (Ka, >0.5 x 1010 liter/mol) as well as high hCG binding capacity (maximum, 2000 ng [125I]hCG bound/ml serum). When ovulatory cycles resumed, the binding capacity for [125I]hCG had decreased to 1 ng/ml serum, and the affinity for LH had decreased. We have detected the presence of a hCG antibody in the serum of a patient with 9 yr of unexplained secondary infertility. The titer remained steady for the period monitored (6 months) after the last detected hCG (results not shown). Unlike the previous case, the antiserum had a lower affinity (Ka, 1.4 x 106 liter/mol), but a higher binding capacity (418 nmol/liter serum; 348,333 IU/liter). This binding capacity was sufficient to neutralize the estimated endogenous hCG level of 135 IU/liter at the time of periimplantation (4) and even the maximum serum hCG levels of 187 IU/liter reported after sc injection of exogenous hCG (5,000 IU), normally used for ovulation induction (16). The failure to recover any oocytes from multiple follicles in three IVF cycles was suggestive of an immature cumulus-oocyte complex, probably due to blocking of the administered exogenous hCG by the antibody. The ability of the patient’s serum to neutralize the bioactivity (in vitro) of hCG at levels greater than those found at the time of periimplantation provided further evidence that the hCG antibody was the prime cause of the patient’s spontaneous abortions. The fact that the antibody was specific for the intact heterodimeric hCG molecule, with low cross-reaction with the - and ß-subunits, suggests that the antibody must be binding to a site encompassing both subunits, which is essential for receptor recognition. The absence of detectable hCG in the patient’s urine provided further evidence that the immunoreactive serum hCG was bound to the antibody. The formation of such a high mol wt complex would prevent filtration through the kidney, which is the major organ responsible for the degradation and/or excretion of hCG (17) and would explain its persistence in the serum. The low cross-reaction with LH or FSH allowed the patient to have normal ovular cycles and to initiate pregnancies.
Clinical trials of hCG vaccination have shown that contraceptive hCG antibody levels decline in the absence of booster injections, indicating that the process is reversible (1, 2). Although the antigen source in our study is not known, the long period of infertility suggests that hCG administered during the various treatments and/or hCG produced by the spontaneous pregnancies have been the immunogenic boosters that have maintained the antibody titer. Alternatively, the patient’s endogenous hCG may have been rendered immunogenic by structural alteration, possibly caused by a mutation in one of the hCGß genes. However, we have not sequenced the CGß gene clusters. Therefore, it was concluded that the high binding capacity of the anti-hCG antibody was the major cause of the patient’s infertility, and the only feasible option for the patient was gestational surrogacy.
Acknowledgments
We thank Ms. Michele Kolo and Mr. Thomas A Gilmore for their technical support.
Footnotes
Abbreviations: ECL, Enhanced chemiluminescence; hLH, human LH; IVF, in vitro fertilization; r-hCG, recombinant hCG.
Received August 8, 2001.
Accepted December 6, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  H. Wan, M. A. Versnel, L. M. E. Leijten, C. G. van Helden-Meeuwsen, D. Fekkes, P. J. M. Leenen, N. A. Khan, R. Benner, and R. C. M. Kiekens Chorionic gonadotropin induces dendritic cells to express a tolerogenic phenotype J. Leukoc. Biol., April 1, 2008; 83(4): 894 - 901. [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 |
