Mol. Hum. Reprod. Advance Access originally published online on April 20, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Human Reproduction, Vol. 10, No. 6, pp. 409-416, 2004
© European Society of Human Reproduction and Embryology 2004
Effect of ß2-glycoprotein I null mutation on reproductive outcome and antiphospholipid antibody-mediated pregnancy pathology in mice
1Department of Obstetrics and Gynaecology and Reproductive Medicine Unit, University of Adelaide, Adelaide, SA 5005 and 2Department of Immunology, Allergy and Infectious Disease and Department of Medicine, University of New South Wales, St George Hospital, NSW 2217, Australia
3 To whom correspondence should be addressed. e-mail: sarah.robertson{at}adelaide.edu.au
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
ß2-Glycoprotein I (ß2GPI) is a principal target antigen for antiphospholipid antibodies associated with recurrent pregnancy loss and fetal growth restriction in women. The significance of disrupted ß2GPI activity in contributing to pregnancy pathology in antiphospholipid syndrome (APS) is not clear. In this study the physiological requirement for functional ß2GPI in pregnancy was investigated by evaluating reproductive outcomes in ß2GPI null mutant (ß2GPI–/–) mice. ß2GPI–/– mice were fertile and carried viable fetuses to term. However, there was an 18% reduction in the number of viable implantation sites in ß2GPI–/– mice and reduced fetal weight and fetal:placental weight ratio in late gestation, suggesting compromised placental function. Placental architecture was altered in ß2GPI–/– implantation sites with a 24% increase in the junctional zone: labyrinthine ratio, but placentae showed no evidence of increased thrombosis in the absence of ß2GPI. The effect of ß2GPI genotype on pregnancy success after passive transfer of human and mouse antibodies reactive with ß2GPI was also explored. Two of five anti-ß2GPI antibodies induced pregnancy loss in ß2GPI+/+ mice but ß2GPI–/– mice were refractory to antibody-induced pregnancy failure. We conclude that functional ß2GPI is not essential for successful pregnancy in mice, but optimal placental development and fetal growth require this molecule. Together these data are consistent with pathogenic mechanisms in antiphospholipid syndrome involving both neutralization of ß2GPI function and ß2GPI–immunoglobulin complex formation.
Key words: antiphospholipid/autoantibodies/ß2-glycoprotein I/placenta/pregnancy
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Antiphospholipid syndrome (APS) is a clinical condition associated with complications of pregnancy ranging from infertility and recurrent miscarriage to placental insufficiency and fetal growth restriction. The clinical syndrome features predisposition to thrombosis, together with autoantibodies reactive with cardiolipin and other negatively charged phospholipids, or phospholipid-binding proteins. A principal antigenic target for antiphospholipid antibodies is ß2-glycoprotein I (ß2GPI) (apoliporotein H), a plasma binding protein for cardiolipin and other anionic phospholipids (Polz and Kostner, 1979; McNeil et al., 1990).
ß2GPI is a highly glycosylated single chain polypeptide of 
50 kDa (Lozier et al., 1984), consisting of five repeating units that belong to the short consensus repeat or complement control protein (CCP) superfamily (Reid and Day 1989). As well as phospholipids, ß2GPI interacts with the surface membranes of platelets (Schousboe 1980), endothelial cells (Meroni et al., 1998) and trophoblast cells (McIntyre, 1992; La Rosa et al., 1994), and is associated with lipoprotein structures, especially chylomicrons (Polz et al., 1980) and heparin (Polz et al., 1981). In vitro experiments indicate that ß2GPI can act as an inhibitor of the intrinsic blood coagulation cascade (Schousboe, 1985), platelet aggregation, and the prothrombinase activity of activated platelets (Nimpf et al., 1987). The occurrence of antibodies to ß2GPI is strongly correlated with arterial and venous thrombosis and other clinical features of APS and distinguishes between clinically significant APS and the transient occurrence of antiphospholipid antibodies secondary to infectious disease (Hunt et al., 1992).
A number of autoimmune disorders have been linked with clinical reproductive loss (Gleicher and el-Roeiy, 1988), with antiphospholipid antibodies having the strongest association. ß2GPI is implicated as the key antigenic target in APS-related reproductive disorders, with the presence of ß2GPI-dependent anticardiolipin reactivity, but not ß2GPI-independent anticardiolipin reactivity, being significantly associated with miscarriage, pre-eclampsia and fetal growth restriction in pregnant women (Katano et al., 1996), and IVF failure in infertile women (Stern et al., 1998). As a single test, ß2GPI reactivity may have greater predictive value for reproductive failure than any other single conventional antiphospholipid antibody test (Aoki et al., 1995).
The pathophysiological mechanisms by which autoantibodies exert adverse effects on pregnancy outcome remain unclear. Early hypotheses invoked thrombotic or microthrombotic disruption of placental formation or function, secondary to effects of antibodies reactive with ß2GPI or other phospholipid moieties on platelets, clotting mediators or local lipoprotein homeostasis. Antiphospholipid antibodies have been suggested to accelerate coagulation at the trophoblast cell surface through displacement of the antithrombotic molecule annexin V (Rand et al., 1997). More recently, ß2GPI reactive antibodies have been implicated in the aberrant activation of endothelial cells (Meroni et al., 1998), consequent upon binding to annexin II-anchored ß2GPI (Ma et al., 2000). A direct effect of ß2GPI-reactive antibodies on placental trophoblast cells is also hypothesized, with reported effects including inhibition of trophoblast cell proliferation (Chamley et al., 1998), interference with hCG secretion and invasiveness (Di Simone et al., 2000), and abnormal expression of adhesion molecules (Di Simone et al., 2002).
In rodents, induction of anti-ß2GPI antibodies either by active immunization or by passive transfer of human ß2GPI-reactive immunoglobulin elicits symptoms consistent with an APS-like syndrome including reduced platelet counts, prolonged clotting time, platelet and endothelial cell activation, and increased fetal resorption and fetal growth restriction (Blank et al., 1994; Garcia et al., 1997; George et al., 1998). While these experiments support a causal role for ß2GPI-reactive antibodies in pregnancy failure, they fail to discriminate between neutralization of ß2GPI activity and antibody-mediated pathology as alternative underlying mechanisms.
To gain insight into the precise function of the ß2GPI molecule in pregnancy, we have studied the fertility and fecundity of mice with a null mutation in the ß2GPI gene created by homologous recombination to replace a 4.7 kb portion of the gene containing part of exon 2 and all of exon 3 with a neomycin resistance cassette (Sheng et al., 2001a). ß2GPI-deficient animals are phenotypically normal apart from a significant reduction in in vitro thrombin generation, suggesting that ß2GPI plays a hitherto unrecognized prothrombotic role. Initially, an effect of ß2GPI status on implantation or fetal development was suggested by findings that only 8.9% of progeny bred from heterozygous parents were homozygous (–/–) for the disrupted allele. In preliminary experiments, ß2GPI–/– mice were seen to be fertile (Sheng et al., 2001a). However, the reported experiment was not designed to detect the scale of effects on fetal and placental growth and function that might be expected after disrupted trophoblast cell viability and behaviour, or perturbed clotting activity. In the current study, we have undertaken a detailed analysis of the reproductive performance of mice genetically deficient in ß2GPI, and have examined the requirement for this protein in estrous cycling, embryo implantation, and fetal and placental development. In addition, we have utilized the null mutant model to assess the importance of endogenous functional ß2GPI in mediating detrimental effects of antiphospholipid antibodies on pregnancy outcome.
|
Materials and methods |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Mice and breeding experiments
Mice homozygous for a disrupted ß2GPI gene (ß2GPI–/– mice) were generated using conventional gene targeting techniques in 129/Sv embryonic stem cells injected into C57Bl/6 blastocysts (Sheng et al., 2001a). The resulting chimeric males were inter-crossed with C57Bl/6 females to produce heterozygous mice, which were then inter-crossed to produce ß2GPI–/– and +/+ lines. ß2GPI–/– and wild-type or heterozygote control (ß2GPI+/+ or +/–) mice were bred in the University of New South Wales Animal Breeding Facility. The genotypes of progeny were confirmed by southern blot on chromosomal DNA isolated from tail biopsies using a probe targeted to intron 4 of the ß2GPI genomic sequence (Sheng et al., 2001a). Breeding experiments were conducted in an SPF facility at the University of Adelaide. Mice were provided with food and water ad libitum. All experiments were approved by the University of Adelaide Animal Ethics Committee, and carried out in accordance with the Australian code of practise for the care and use of animals for scientific purposes.
For analysis of estrous cycles, vaginal smears were prepared at 1000–1200 h daily and examined by phase contrast microscopy. Mice were allocated to one of four stages of the cycle on the basis of the cellular composition; proestrus (>50% intact, live epithelial cells), estrus (100% cornified epithelial cells), metestrus (50% leukocytes and 50% cornified epithelial cells) or diestrus (>70% leukocytes, ± cornified or intact epithelial cells). For breeding experiments, adult females (8–12 week, ß2GPI–/– or ß2GPI+/±) were housed 2:1 with adult stud males (ß2GPI–/– or +/+) and allowed to mate naturally. The day on which a copulation plug was evident was nominated day 1 of pregnancy, and the interval between placing with males and day 1 of pregnancy was noted for each female. In the first experiment, pregnant females were housed separately from males in groups of three to five and killed by cervical dislocation at 10:00–13:00 h on day 18 of pregnancy, when the number of viable and resorbing implantation sites was recorded. Viable fetuses and placentae were dissected free of decidua and fetal membranes and weighed, and the fetal:placental weight ratio (fetal weight/placental weight) was calculated. In term experiments, females were housed separately from males after detection of plugs, and pregnancy was allowed to proceed until term, when the date and time (to the nearest 0.5 day) of parturition, and the number of live pups were recorded. Pups were weighed 14–20 h after birth, and then at 8 days, 3 weeks (weaning), 6 weeks and 16 weeks after birth.
Placental histochemistry
Placentae and underlying decidual tissue were dissected from implantation sites in day 18 pregnant ß2GPI–/– and+/+ mice mated with males of the same genotypes (n = 11 from eight pregnant ß2GPI–/– females and n = 10 from seven pregnant ß2GPI+/+ females). After immersion fixation in 4% paraformaldehyde/2.5% polyvinylpyrrolidone in 70 mmol/l phosphate buffer at 4°C overnight, placentae were bisected in the mid-sagittal plane and fixed for a further 4 h. Placentae were then washed in sterile phosphate-buffered saline (PBS) four times over 2 days, processed and embedded in paraffin. Mid-sagittal 7 µm sections were cut through each placenta and stained with Masson’s trichrome, or Martius, Scarlet, Blue (MSB) using standard protocols (Bancroft and Stevens, 1990). The ability of MSB to detect fibrous deposits was confirmed in human term placental tissue. Fibrin deposition in MSB-stained mouse placental tissue was then evaluated qualitatively. The areas of the junctional zone and placental labyrinth were measured in Masson’s trichrome-stained sections by video image analysis (VIA) using Video Pro software (Leading Edge Software, Australia), using a x10 objective and x3.3 photo eyepiece lens. The video image was calibrated to micrometres with the aid of a haemocytometer. Repeated measurements of the junctional zone area in one section validated the precision of this method (<5% within-assay variation).
Passive antibody experiments
Two monoclonal autoantibodies FC1 (reactive with human ß2GPI; isotype IgG1) and FB2 (reactive with cardiolipin; isotype IgG2b) were developed from spleens of NZWxBXSBF1 mice, which spontaneously develop a systemic lupus erythematosus-like syndrome (Monestier et al., 1996). GR1D5, an IgM monoclonal antibody with specificity for ß2GPI, was obtained from a patient with APS as described previously (Ichikawa et al., 1994). Previous studies suggest that GR1D5 recognizes the same or a closely related epitope to the polyclonal human IgG anticardiolipin autoantibodies from this patient (Ichikawa et al., 1994). The mouse monoclonal antibodies were purified from hybridoma supernatants using Protein G–sepharose and the GR1D5 human IgM monoclonal antibody was purified using affinity chromatography with anti-IgM sepharose. Human antibodies obtained as serum were analysed by immunoassay for reactivity against cardiolipin using the anticardiolipin antibody enzyme-linked immunosorbent assay test kit (Medical Innovations Ltd, Australia) and ß2GPI as previously described (Sheng et al., 2001b). PA1, PA2 and PA3 polyclonal human IgG antibodies were prepared from the serum of women diagnosed with APS and a history of recurrent miscarriage. PA1 and PA2 were purified with sequential phosphatidyl serine affinity chromatography and ion exchange chromatography as previously described (McNeil et al., 1990). PA3 was purified by ammonium sulphate precipitation followed by Protein G–sepharose. A control serum (HuIgG) was obtained from a healthy volunteer and purified with ammonium sulphate precipitation followed by Protein G–sepharose. All antibody preparations were subjected to Detoxi-GelTM (Pierce, USA) to remove endotoxin and demonstrated to be free of endotoxin using an E-Toxate® Kit (Sigma, USA).
Adult females (8–16 weeks, ß2GPI–/–,+/– or +/+) were housed 3:1 with adult stud males (ß2GPI–/– or +/+) and allowed to mate naturally. Mice were administered antibodies (10 µg in 100 µl of PBS) or PBS (control) by i.v. injection (tail vein) at 1000–1200 h on days 1, 4 and 8 of pregnancy. Pregnant females were housed separately from males in groups of three to five and killed by cervical dislocation at 1000–1300 h on day 18 of pregnancy, when the numbers of viable and resorbing implantation sites were recorded. Viable fetuses and placentae were dissected free of decidua and fetal membranes and weighed, and the fetal:placental weight ratio (fetal weight/placental weight) was calculated.
Platelet counts
Blood was recovered by cardiac puncture into EDTA-coated paediatric Capiject tubes (Terumo) from female virgin, day 12 or day 18 pregnant ß2GPI+/– and ß2GPI–/– mice at 12–16 weeks of age after anaesthesia with avertin [1 mg/ml tribromoethyl alcohol in tertiary amyl alcohol (Sigma, USA) diluted to 2.5% v/v in saline; 15 µl/g body weight injected i.p.]. The concentration of platelets in blood was determined using a Technicon H2 automated haematology analyser (Bayer Diagnostics, USA).
Statistical analysis
Fertility data were analysed by independent sample t-tests, or by analysis of variance (ANOVA) followed by Bonferroni t-tests when more than two groups were compared. Placental areas were analysed by one-way ANOVA. All analyses were performed using SPSS 11.5 software (SPSS, USA). Data expressed as proportions were analysed by CHITEST and CHIDIST procedures in Excel 5.0 (Microsoft, USA). Differences between groups were considered to be significant when P < 0.05.
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Effect of ß2GPI deficiency on litter size and fetal and placental development
To investigate fertility and fecundity in mice genetically deficient in ß2GPI, fetal and placental development were examined late in gestation in pregnant ß2GPI+/+ or +/– and ß2GPI–/– females. Prior to mating, daily vaginal smears were taken from virgin ß2GPI+/+ females (n = 9) and ß2GPI–/– females (n = 12) for 2 weeks to determine the effect of ß2GPI deficiency on the estrous cycle. All control mice and all but two mice in the ß2GPI–/– group demonstrated normal cycling behaviour, as defined by at least two estrus events in the 2 week period. Mean cycle lengths were comparable to control values in ß2GPI–/– mice (mean ± SD = 5.1 ± 1.9 days versus 4.9 ± 1.4 days in ß2GPI+/+ mice).
Virgin ß2GPI+/+ or +/– and ß2GPI–/– females (n = 36, both groups) were then mated naturally with males of the same ß2GPI status (ß2GPI+/+ and ß2GPI–/– respectively) and killed on day 18 of pregnancy. The number of days between placing females with stud males and the day of detection of a copulatory plug (‘mating interval’) was comparable in ß2GPI–/– and ß2GPI+/± females mated with stud males of the same ß2GPI status (mean ± SD = 2.8 ± 1.0 in ß2GPI–/– females mated with–/– males, and 2.5 ± 1.2 in ß2GPI+/± females mated with+/+ males). The proportion of mice plugged on day 1 that were pregnant on day 18 was similar in ß2GPI–/– and ß2GPI+/± females (Table I). The number of implantation sites was influenced by ß2GPI status, with a 15% and 18% reduction in total and viable implants respectively in ß2GPI–/– females (both P < 0.02, Table I). The proportion of implantation sites undergoing resorption at day 18 was comparable (10% in ß2GPI–/– and 11% in ß2GPI+/± females, Table I).
|
To examine whether ß2GPI status influenced growth of the placenta or fetus, weights of viable fetuses and placentae were measured in ß2GPI–/– (n = 249) and ß2GPI+/± (n = 183) implantation sites. A modest but statistically significant decrease in fetal weight (3%; P = 0.02) and increase in placental weight (5%; P = 0.01) was seen in ß2GPI–/– implantation sites. Fetal:placental weight ratio, a measure of placental function, was decreased by 7% in ß2GPI–/– mice (P < 0.001; Table I).
Reproductive performance was also recorded in first pregnancies of virgin females in the breeding colony. Virgin ß2GPI+/+ and ß2GPI–/– females were mated naturally with males of the same ß2GPI status (ß2GPI+/+ and –/– respectively). In agreement with our previous findings (Sheng et al., 2001a), data from this larger data set showed that the proportion of mice plugged on day 1 which delivered viable pups at term was not affected by genotype. There was no effect of ß2GPI deficiency on the length of gestation [mean ± SD = 19.9 ± 0.4 and 20.1 ± 0.5 days in ß2GPI+/± females (n = 21) and ß2GPI–/– females (n = 27), respectively]. There was a trend towards smaller litter size in ß2GPI–/– females (mean ± SD = 8.1 ± 1.4 in ß2GPI+/+ versus 7.6 ± 1.7 pups in ß2GPI–/– litters) although the difference did not reach statistical significance. The survival of pups during the peri-natal and post-natal periods were comparable in ß2GPI-deficient and ß2GPI-replete litters, and growth trajectories of male and female pups to 16 weeks of age were not affected by genotype (n = 118–152 ß2GPI+/+ pups, and n = 158–200 ß2GPI–/– pups, data not shown).
Effect of ß2GPI deficiency on placental structure and fibrin deposition at day 18
In view of the potential role of ß2GPI in regulating the coagulation cascade and the above data suggesting subtle effects of ß2GPI gene disruption on placental size and function, the histology of ß2GPI–/– and ß2GPI+/± placentae at day 18 of pregnancy was examined at the light microscope level. Duplicate sections of placentae taken from 11 healthy ß2GPI–/– implantation sites and 10 healthy ß2GPI+/± implantation sites were stained with Masson’s trichrome to differentiate between the junctional and labyrinthine (exchange) layers in the placenta, and with MSB to detect fibrin associated with thrombotic plaques. There were significant differences in the structure of placentae from ß2GPI–/– mice, with a 16% increase in the absolute cross-sectional area of the junctional zone (P = 0.046). The proportion of the cross-sectional area of the placenta comprised by junctional zone was increased by 10% while that of the labyrinth decreased by 11% in ß2GPI–/– mice compared to ß2GPI-replete mice (both P = 0.059), leading to a 24% increase in the junctional zone:labyrinthine ratio in ß2GPI–/– mice (P = 0.036) (Table II and Figure 1). MSB staining revealed some fibrin deposition in large maternal vascular sinuses lined by spongiotrophoblasts in the junctional zone of the placenta, indicative of previous thromboses, but these were similar in size and abundance in ß2GPI–/– and ß2GPI+/+ placentae (Figure 1).
|
|
Effect of passive immunization with antiphosholipid antibodies in ß2GPI-deficient and wild-type mice
Antibody-mediated neutralization of ß2GPI function and signalling events triggered by ß2GPI–immunoglobulin complex formation are alternative mechanisms postulated to explain pregnancy pathologies in human APS and in mouse models of APS. To examine the requirement for functional ß2GPI protein in mediating pregnancy pathology elicited by ß2GPI-reactive antibodies in mice, passive immunization experiments were undertaken in wild-type and ß2GPI null mutant mice. In initial experiments, groups of 10–12 ß2GPI+/± mice were administered one of a panel of purified antibodies including mouse monoclonal antibodies FC1 (reactive with human ß2GPI) and FB2 (reactive with cardiolipin), human monoclonal antibody GR1D5 (reactive with ß2GPI), and human polyclonal IgG antibodies PA1, PA2 and PA3 prepared from women diagnosed with APS and a history of recurrent miscarriage. Additional groups of mice received control immunoglobulin or carrier alone, and all mice were treated on days 1, 4 and 8 of pregnancy. Pregnancy outcomes were affected after administration of two of the six experimental antibodies, with complete pregnancy loss evident in 6/12 (50%) of mice receiving human ß2GPI reactive monoclonal antibody GR1D5, and 4/12 (33%) of mice receiving human polyclonal antiphospholipid antibody PA3. All other groups showed normal pregnancy outcomes as judged by the proportion of mated mice pregnant at day 18 (80% or higher), with resorption rates and fetal and placental development indistinguishable from control mice receiving PBS or human IgG (not shown).
In larger experimental groups, GR1D5, PA3, human IgG or carrier (PBS) were administered to both ß2GPI+/± and ß2GPI–/– mice. Recipient mice were killed on day 18 of pregnancy and scored as pregnant or not pregnant, and when viable implantation sites were present, implantation numbers, viability and fetal and placental weights were recorded. The numbers of total and viable implantation sites were reduced by 13 and 18% respectively in ß2GPI–/– compared with ß2GPI+/± females administered PBS (both P < 0.05, Table III) and comparable in size to the previous experiment (Table I), showing that the injection regimen did not impact negatively on pregnancy success. Pregnancy outcomes in mice injected with normal human IgG were comparable to those for mice injected with carrier (PBS), in terms of proportion pregnant at day 18, number and viability of implantation sites and fetal and placental development.
|
Both experimental antibody preparations compromised pregnancy outcome as judged by the proportion of mated mice pregnant at day 18, with more profound effects in ß2GPI+/± than in ß2GPI–/– mice (Table III). Pregnancy loss was evident in 12/24 (50%) of ß2GPI+/± mice administered ß2GPI-reactive GR1D5 (significantly greater than in ß2GPI+/± administered PBS,
2-test P = 0.025). Fetal weights, placental weights and fetal:placental weight ratio were not altered in ß2GPI+/± mice that remained pregnant after administration of GR1D5. In contrast, while a higher proportion of pregnancy loss was evident in ß2GPI–/– mice administered GR1D5 than PBS (37% and 18% respectively), this effect was not statistically significant (P = 0.080). Mice receiving GR1D5 and remaining pregnant were found to have numbers of total and viable implantation sites indistinguishable from control mice of the same genotypes.
Similar results were obtained with human polyclonal antiphospholipid antibody PA3. Pregnancy loss was considerable in ß2GPI+/± mice receiving PA3 [12/24 (50%), 2-test P = 0.025] but was not different from control rates in ß2GPI–/– mice [6/27 (22%); P = 0.56] (Table III). Mice receiving PA3 and remaining pregnant had numbers of total and viable implantation sites indistinguishable from control mice of the same genotypes, and fetal and placental weights were unaltered (Table III). A reduction in fetal:placental weight ratio after administration of PA3 could not be attributed to the specificity of the antibody, since a similar effect was seen in mice administered normal human immunoglobulin.
The effect of ß2GPI deficiency and passive antibody treatment on platelet counts in pregnancy
Passive immunization with antiphospholipid antibodies has been linked with thrombocytopenia in pregnant mice. To investigate the significance of ß2GPI in regulating platelet consumption in pregnancy, platelet counts were measured using an automated flow cytometer in blood obtained at day 13 and day 18 of pregnancy in untreated and antibody-treated ß2GPI+/± and ß2GPI–/– mice. There was no effect of genotype on platelet content in either virgin or pregnant mice, and blood platelet content was increased late in gestation compared with virgin or day 13 pregnant animals (P < 0.000), irrespective of genotype (Figure 2A). Moreover, there was no effect of administration of either GR1D5 or PA3 on platelet numbers at day 18 of pregnancy in either genotype (Figure 2B).
|
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
These data indicate that reproductive performance in mice is compromised by maternal and fetal deficiency in functional ß2GPI. Pregnancy was initiated and progressed to term at normal frequencies in ß2GPI null mutant mice. However, consistent reductions in litter size were evident, along with reduced fetal weight associated with changes in placental size and structure indicative of reduced placental efficiency. Passive antibody transfer experiments implicated a role for the ß2GPI molecule in mediating the detrimental effects of antiphospholipid antibodies in pregnancy, with two antibody preparations eliciting more severe pregnancy pathologies in the presence of endogenous ß2GPI protein than in ß2GPI null mutant animals. Together these findings suggest that neutralization of ß2GPI function might contribute to placental insufficiency in APS. However, since the phenotype in mutant mice is modest and intact ß2GPI is necessary to achieve the full effect on fetal loss mediated by anti-ß2GPI-reactive antibodies, loss of ß2GPI function is unlikely to be the sole underlying pathogenic mechanism.
The lack of essential requirement for ß2GPI in murine pregnancy is unexpected given the proposed roles for this molecule in regulating the coagulation cascade and lipid metabolism. Our previous studies indicate altered in vitro thrombin generation in ß2GPI null mutant mice, although no significant differences in blood clotting time were detected by several conventional coagulation assays (Sheng et al., 2001a). While the molecular mechanisms mediating reduced litter size need further exploration, the lack of evidence of increased resorption rates at day 18 in ß2GPI null mutants indicates that any fetal loss must occur early after implantation. Alternatively, smaller litter sizes may result from implantation failure or events prior to implantation, including ovulation and early embryogenesis. Our earlier finding of smaller than expected numbers of homozygous null mutant embryos in heterozygote matings (only 8.9%) (Sheng et al., 2001a) suggested that ß2GPI deficiency might pose a selective disadvantage in early embryonic development or implantation. Since the mice used in these experiments have a mixed C57Bl/6 and 129/Sv genetic background, it is possible that there are effects of other genes on the penetrance of the phenotype arising from ß2GPI null mutation.
The alteration in placental structure seen in ß2GPI null mutant mice, specifically the relatively enlarged junctional zone, is consistent with aberrant and possibly delayed placental morphogenesis and suggests that fetal growth restriction results from impaired placental nutrient transfer function. The placental junctional zone in the murine placenta is the site of trophoblast cell proliferation, differentiation and hormone synthesis, while the labyrinthine layer is the site of exchange of nutrients, gases and wastes between the maternal and fetal circulations (Redline and Lu, 1989; Georgiades et al., 2002). The junctional zone is often called the spongiotrophoblast layer, but since it is comprised of both spongiotrophoblast and glycogen cells and forms the feto-maternal interface in the latter third of pregnancy, it is more correctly referred to as the junctional zone (Georgiades et al., 2002; Coan et al., 2004). Glycogen cells comprise an invasive population of trophoblasts that migrate deeper into the decidua from day 13 of gestation and are analogous to invasive extravillous cytotrophoblasts of the human placenta (Redline and Lu, 1989; Robertson et al., 1999; Georgiades et al., 2002). The placental junctional zone, derived from ectoplacental cone, grows more slowly than the placental labyrinth but nevertheless doubles in volume between days 12.5 and 16.5 of gestation, after which its volume and proportion declines. From days 12.5 to 18.5, the volume of the labyrinth increases 4-fold as trophoblasts, maternal blood space and fetal capillaries grow and remodel, supporting increased fetal demand on labyrinthine exchange function in the mature placenta (Coan et al., 2004).
The cause of perturbed placental structure in the absence of ß2GPI is unclear, with no histological evidence for thrombotic plaques and no change in circulating platelet numbers to support a coagulation-related aetiology. Interestingly, genetic deficiencies in growth factors required for optimal placental development, including granulocyte-macrophage colony-stimulating factor, result in similar placental phenotypes (Robertson et al., 1999). This is consistent with the possibility that ß2GPI deficiency limits placental development indirectly through other means, perhaps involving endothelial cell activation and establishment of the placental vasculature, or through interfering with lipid availability.
The effects of human antiphospholipid antibodies on pregnancy in mice were less severe in the current study than in previously reported studies. Each of the two antibodies found to influence pregnancy appeared to have an ‘all-or-nothing’ effect, acting either to induce 100% pregnancy loss in 50% of mice and, as inferred from largely unchanged resorption rates and fetal growth parameters, failing to elicit any impact on pregnancy outcome in the remainder. Placental development and circulating platelet numbers were also normal in pregnant antibody-treated animals. This contrasts with reports from other groups of fetal resorption, severe fetal growth restriction and placental pathologies associated with evidence of thrombocytopenia and placental thrombosis and necrosis in murine models of APS using antibodies reactive with ß2GPI or cardiolipin or other phospholipids (Blank et al., 1991, 1994; Girardi et al., 2003; Piona et al., 1995). It is reasonable to expect that differences in the ligand and epitope specificities, concentration, and avidity of human antiphospholipid sera could account for variation between clinical preparations. Indeed, other authors report considerable variation in the ability of anticardiolipin antibodies to induce pregnancy loss in mice (Sthoeger et al., 1993; Chamley et al., 1994). Moreover, species-related distinctions in the biochemistry of mouse and human ß2GPI and its association with other molecular moieties would be expected to affect the downstream consequences of ß2GPI–immunoglobulin interaction at the trophoblast or endothelial cell surface, such that antibodies might be pathogenic in humans but innocuous in mice. A further distinction between the human clinical condition and mouse models is the concentration of circulating antibodies, with pathologies in mice reported after a single administration of 10 µg of antibody (corresponding to 300 ng/ml in blood) (Sthoeger et al., 1993; Blank, 1994; Piona et al., 1995), whereas APL antibodies may reach considerably higher concentrations in women (Branch et al., 1990). Higher doses of passively administered antibody thus might be expected to induce more severe pathologies of pregnancy in mice.
There was rarely evidence of fetal resorption or any conceptus-derived tissue in uteri of non-pregnant, antibody-treated mice. Thus, in our hands, pregnancy failure manifested as loss in the pre- or early post-implantation period, presumably reflecting a role for target antigens in events of early pregnancy. This result is consistent with reported effects of antiphospholipid antibodies on development of the preimplantation embryo and maternal tract receptivity (Sthoeger et al., 1993; Tartakovsky et al., 1996). Both pathology-inducing antibody preparations elicited more pronounced pregnancy loss in ß2GPI-replete mice, supporting a central role for ß2GPI protein in the underlying mechanisms. This is not unexpected in the case of the monoclonal anti-ß2GPI-reactive GR1D5, and is also reasonable for polyclonal PA3, where immunoglobulins specific for ß2GPI are likely to be present amongst a range of phospholipid and co-factor reactivities represented.
It is difficult to explain our observation of differential susceptibility of individual animals to treatment with ß2GPI-reactive antibodies. It has been reported that endothelial damage resulting from mechanical injury (Pierangeli et al., 1994) or phytochemical insult (Jankowski, et al., 2002) can potentiate the prothrombotic consequences of ß2GPI-reactive antibody administration, indicative of a ‘double hit’ phenomenon (Jankowski et al., 2002). Early pregnancy is extremely sensitive to environmental stressors acting on the nervous, endocrine and immune systems (Arck 2001), all of which can elevate TNF
expression to promote endothelial damage and stimulate thrombotic processes (Clark et al., 1998). It is reasonable to speculate that environmental stressors interacting with genetic modifiers and immune status influence both the ability of individual mice to compensate for genetic ß2GPI deficiency and the pathological effects of ß2GPI-reactive antibodies.
In summary, these experiments show that genetic ß2GPI deficiency can interfere with optimal placental morphogenesis, indicating that neutralization of ß2GPI activity may contribute to the placental insufficiency underlying miscarriage and fetal growth restriction in women. Since thrombosis was not evident in ß2GPI null placentas, any detrimental effect of neutralizing ß2GPI activity in the human placenta is likely to be mediated via interfering with a function other than anticoagulant activity. Indeed our finding of dysregulated placental morphogenesis in the absence of thrombosis in mice is consistent with findings in placental tissue recovered from women experiencing antiphospholipid antibody syndrome-associated early pregnancy failure, where defective decidual endovascular trophoblast invasion, rather than excessive intervillous thrombosis, is the most frequent histological abnormality (Sebire et al., 2002). However, neutralization of ß2GPI function is unlikely to be the sole explanation for the pathogenic effects of ß2GPI-reactive antibodies in vivo. The refractoriness of ß2GPI null mutants to antiphospholipid syndrome-like pathology supports a central role for the ß2GPI molecule and for ß2GPI-immunoglobulin complex formation in the signalling cascade mediating pregnancy disruption. This interpretation is consistent with data linking ß2GPI-reactive antibodies to aberrant endothelial cell activation via binding to annexin II-anchored ß2GPI (Ma et al., 2000) or other signalling pathways such as Toll-like receptors (Raschi et al., 2002), as well as indications from complement factor-deficient mice that fetal injury results from antiphospholipid antibody-mediated triggering of the complement cascade (Girardi et al., 2003). We conclude that both neutralization of ß2GPI function as well as ß2GPI–immunoglobulin complex formation are implicated in the pathological mechanisms of pregnancy disruption in APS in women.
|
Acknowledgements |
|---|
This study was supported by NHMRC (Australia) and Women’s and Children’s Hospital (Adelaide, Australia) project grants and an NHMRC Research Fellowship (S.A.R).
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Aoki K, Dudkiewicz AB, Matsuura E, Novotny M, Kaberlein G and Gleicher N (1995) Clinical significance of beta 2-glycoprotein I-dependent anticardiolipin antibodies in the reproductive autoimmune failure syndrome: correlation with conventional antiphospholipid antibody detection systems. Am J Obstet Gynecol 172,926–931.[CrossRef][ISI][Medline]
Arck PC (2001) Stress and pregnancy loss: role of immune mediators, hormones and neurotransmitters. Am J Reprod Immunol 46,117–123.[CrossRef][ISI][Medline]
Bancroft J and Stevens A (eds) (1990) Theory and Practice of Histological Techniques 3rd edn, Churchill Livingstone, Edinburgh, UK.
Blank M, Cohen J, Toder V and Shoenfeld Y (1991) Induction of anti-phospholipid syndrome in naive mice with mouse lupus monoclonal and human polyclonal anti-cardiolipin antibodies. Proc Natl Acad Sci USA 88,3069–3073.
Blank M, Tincani A and Shoenfeld Y (1994) Induction of experimental antiphospholipid syndrome in naive mice with purified IgG antiphosphatidylserine antibodies. J Rheumatol 21,100–104.[ISI][Medline]
Branch DW, Dudley DJ, Mitchell MD, Creighton KA, Abbott TM, Hammond EH and Daynes RA (1990) Immunoglobulin G fractions from patients with antiphospholipid antibodies cause fetal death in BALB/c mice: a model for autoimmune fetal loss. Am J Obstet Gynecol 163,210–216.[ISI][Medline]
Chamley LW, Pattison NS and McKay EJ (1994) The effect of human anticardiolipin antibodies on murine pregnancy. J Reprod Immunol 27,123–134.[CrossRef][ISI][Medline]
Chamley LW, Duncalf AM, Mitchell MD and Johnson PM (1998) Action of anticardiolipin and antibodies to beta2-glycoprotein-I on trophoblast proliferation as a mechanism for fetal death. Lancet 352,1037–1038.[CrossRef][ISI][Medline]
Clark DA, Chaouat G, Arck PC, Mittruecker HW and Levy GA (1998) Cytokine-dependent abortion in CBA x DBA/2 mice is mediated by the procoagulant fgl2 prothombinase. J Immunol 160,545–549.
Coan PM, Ferguson-Smith AC and Burton GJ (2004) Developmental dynamics of the definitive mouse placenta assessed by stereology. Biol Reprod Feb 18 [E-pub].
DiSimone N, Caliandro D, Castellani R, Ferrazzani S and Caruso A (2000) Interleukin-3 and human trophoblast: in vitro explanations for the effect of interleukin in patients with antiphospholipid antibody syndrome. Fertil Steril 73,1194–200.[CrossRef][ISI][Medline]
DiSimone N, Castellani R, Caliandro D and Caruso A (2002) Antiphospholid antibodies regulate the expression of trophoblast cell adhesion molecules. Fertil Steril 77,805–811.[CrossRef][ISI][Medline]
Garcia CO, Kanbour-Shakir A, Tang H, Molina JF, Espinoza LR and Gharavi AE (1997) Induction of experimental antiphospholipid antibody syndrome in PL/J mice following immunization with beta 2 GPI. Am J Reprod Immunol 37,118–124.[Medline]
George J, Afek A, Gilburd B, Blank M, Levy Y, Aron-Maor A, Levkovitz H, Shaish A, Goldberg I, Kopolovic J, Harats D and Shoenfeld Y (1998) Induction of early atherosclerosis in LDL-receptor-deficient mice immunized with beta2-glycoprotein I. Circulation 98,1108–1115.
Georgiades P, Ferguson-Smith AC and Burton GJ (2002) Comparative developmental anatomy of the murine and human definitive placentae. Placenta 23,3–19.[CrossRef][ISI][Medline]
Girardi G, Berman J, Redecha P, Spruce L, Thurman JM, Kraus D, Hollmann TJ, Casali P, Caroll MC, Wetsel RA, Lambris JD, Holers VM and Salmon JE (2003) Complement C5a receptors and neutrophils mediate fetal injury in the antiphospholipid syndrome. J Clin Invest 112,1644–1654.[CrossRef][ISI][Medline]
Gleicher N and el-Roeiy A (1988) The reproductive autoimmune failure syndrome. Am J Obstet Gynecol 159,223–227.[ISI][Medline]
Hunt JE, McNeil HP, Morgan GJ, Crameri R M and Krilis SA (1992) A phospholipid-beta 2-glycoprotein I complex is an antigen for anticardiolipin antibodies occurring in autoimmune disease but not with infection. Lupus 1,75–81.
Ichikawa K, Khamashta MA, Koike T, Matsuura E and Hughes GR (1994) beta 2-Glycoprotein I reactivity of monoclonal anticardiolipin antibodies from patients with the antiphospholipid syndrome. Arthritis Rheum 37,1453–1461.[ISI][Medline]
Jankowski M, Vreys I, Wittevrongel C, Boon D, Vermylen J, Hoylaerts MF and Arnout J (2002) Thrombogenicity of {beta}2-glycoprotein I-dependent antiphospholipid antibodies in a photochemically-induced thrombosis model in the hamster. Blood 5,5.
Katano K, Aoki A, Sasa H, Ogasawara M, Matsuura E and Yagami Y (1996) beta 2-Glycoprotein I-dependent anticardiolipin antibodies as a predictor of adverse pregnancy outcomes in healthy pregnant women. Hum Reprod 11,509–512.[ISI][Medline]
LaRosa L, Meroni PL, Tincani A, Balestrieri G, Faden D, Lojacono A, Morassi L, Brocchi E, Del Papa N, Gharavi A et al (1994) Beta 2 glycoprotein I and placental anticoagulant protein I in placentae from patients with antiphospholipid syndrome. J Rheumatol 21,1684–1693.[ISI][Medline]
Lozier J, Takahashi N and Putnam FW (1984) Complete amino acid sequence of human plasma beta 2-glycoprotein I. Proc Natl Acad Sci USA 81,3640–3644.
Ma K, Simantov R, Zhang JC, Silverstein R, Hajjar KA and McCrae KR (2000) High affinity binding of beta 2-glycoprotein I to human endothelial cells is mediated by annexin II. J Biol Chem 275,15541–15548.
McIntyre JA (1992) Immune recognition at the maternal–fetal interface: overview. Am J Reprod Immunol 28,127–131.[ISI][Medline]
McNeil HP, Simpson RJ, Chesterman CN and Krilis SA (1990) Anti-phospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibitor of coagulation: beta 2-glycoprotein I (apolipoprotein H). Proc Natl Acad Sci USA 87,4120–4124.
Meroni PL, Del Papa N, Raschi E, Panzeri P, Borghi MO, Tincani A, Balestrieri G, Khamashta MA, Hughes GR, Koike T and Krilis SA (1998) Beta2-glycoprotein I as a ‘cofactor’ for anti-phospholipid reactivity with endothelial cells. Lupus 7,S44–47.[Medline]
Monestier M, Kandiah DA, Kouts S, Novick KE, Ong GL, Radic MZ and Krilis SA (1996) Monoclonal antibodies from NZW x BXSB F1 mice to beta2 glycoprotein I and cardiolipin. Species specificity and charge-dependent binding. J Immunol 156,2631–2641.[Abstract]
Nimpf J, Wurm H and Kostner GM (1987) Beta 2-glycoprotein-I (apo-H) inhibits the release reaction of human platelets during ADP-induced aggregation. Atherosclerosis 63,109–114.[CrossRef][ISI][Medline]
Pierangeli SS, Barker JH, Stikovac D, Ackerman D, Anderson G, Barquinero J, Acland R and Harris EN (1994) Effect of human IgG antiphospholipid antibodies on an in vivo thrombosis model in mice. Thromb Haemost 71,670–674.[ISI][Medline]
Piona A, La Rosa L, Tincani A, Faden D, Magro G, Grasso S, Nicoletti F, Balestrieri G and Meroni PL (1995) Placental thrombosis and fetal loss after passive transfer of mouse lupus monoclonal or human polyclonal anti-cardiolipin antibodies in pregnant naive BALB/c mice. Scand J Immunol 41,427–432.[CrossRef][ISI][Medline]
Polz E and Kostner GM (1979) The binding of beta 2-glycoprotein-I to human serum lipoproteins: distribution among density fractions. FEBS Lett 102,183–186.[CrossRef][ISI][Medline]
Polz E, Wurm H and Kostner GM (1980) Investigations on beta 2-glycoprotein-I in the rat: isolation from serum and demonstration in lipoprotein density fractions. Int J Biochem 11,265–270.[CrossRef][ISI][Medline]
Polz E, Wurm H and Kostner GM (1981) Studies on the composition of the protein part of triglyceride rich lipoproteins of human serum: isolation of polymorphic forms of beta 2-glycoprotein-I. Artery 9,305–315.[ISI][Medline]
Rand JH, Wu XX, Andree HA, Lockwood CJ, Guller S, Scher J and Harpel PC (1997) Pregnancy loss in the antiphospholipid-antibody syndrome—a possible thrombogenic mechanism. N Engl J Med 337,154–160.
Raschi E, Testoni C, Bosisio D, Boghi M, Koike T, Mantovani A and Meroni P (2002) Role of the MYD88 transduction signalling pathway induced by anti-beta2 glycoprotein I antibodies in endothelial cells. 10th International Congress on the Antiphospholipid Antibodies, Sicily, Italy. Abstract 186.
Redline RW and Lu CY (1989) Localization of fetal major histocompatibility complex antigens and maternal leukocytes in murine placenta. Implications for maternal–fetal immunological relationship. Lab Invest 61,27–36.[ISI][Medline]
Reid KB and Day AJ (1989) Structure-function relationships of the complement components. Immunol Today 10,177–180.[CrossRef][ISI][Medline]
Robertson SA, Roberts CT, Farr KL, Dunn AR and Seamark RF (1999) Fertility impairment in granulocyte-macrophage colony-stimulating factor-deficient mice. Biol Reprod 60,251–161.
Schousboe I (1980) Binding of beta 2-glycoprotein I to platelets: effect of adenylate cyclase activity. Thromb Res 19,225–237.[CrossRef][ISI][Medline]
Schousboe I (1985) beta 2-Glycoprotein I: a plasma inhibitor of the contact activation of the intrinsic blood coagulation pathway. Blood 66,1086–1091.
Sebire NJ, Fox H, Backos M, Rai R, Paterson C and Regan L (2002) Defective endovascular trophoblast invasion in primary antiphospholipid antibody syndrome-associated early pregnancy failure. Hum Reprod 17,1067–1071.
Sheng Y, Reddel SW, Herzog H, Wang YX, Brighton T, France MP, Robertson SA and Krilis SA (2001a) Impaired thrombin generation in beta 2-glycoprotein I null mice. J. Biol. Chem. 276,13817–13821.
Sheng Y, Hanly JG, Reddel SW, Kouts S, Guerin J, Koike T, Ichikawa K, Sturgess A, Krilis SA (2001b) Detection of ‘antiphospholipid’ antibodies: a single chromogenic assay of thrombin generation sensitively detects lupus anticoagulants, anticardiolipin antibodies, plus antibodies binding beta(2)-glycoprotein I and prothrombin. Clin Exp Immunol 124,502–508.[CrossRef][ISI][Medline]
Stern C, Chamley L, Hale L, Kloss M, Speirs A and Baker HW (1998) Antibodies to beta2 glycoprotein I are associated with in vitro fertilization implantation failure as well as recurrent miscarriage: results of a prevalence study. Fertil Steril 70,938–944.[CrossRef][ISI][Medline]
Sthoeger ZM, Mozes E and Tartakovsky B (1993) Anti-cardiolipin antibodies induce pregnancy failure by impairing embryonic implantation. Proc Natl Acad Sci USA 90,6464–6467.
Tartakovsky B, Bermas BL, Sthoeger Z, Shearer GM and Mozes E (1996) Defective maternal-fetal interaction in a murine autoimmune model. Hum Reprod 11,2408–2411.
Submitted on January 9, 2004; resubmitted on March 15, 2004; accepted on March 21, 2004.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Blank, L. Anafi, G. Zandman-Goddard, I. Krause, S. Goldman, E. Shalev, R. Cervera, J. Font, M. Fridkin, H.-J. Thiesen, et al. The efficacy of specific IVIG anti-idiotypic antibodies in antiphospholipid syndrome (APS): trophoblast invasiveness and APS animal model Int. Immunol., July 1, 2007; 19(7): 857 - 865. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Giannakopoulos, F. Passam, S. Rahgozar, and S. A. Krilis Current concepts on the pathogenesis of the antiphospholipid syndrome Blood, January 15, 2007; 109(2): 422 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Tian, P. Jackson, C. Gunter, J. Wang, C. O. Rock, and S. Jackowski Placental Thrombosis and Spontaneous Fetal Death in Mice Deficient in Ethanolamine Kinase 2 J. Biol. Chem., September 22, 2006; 281(38): 28438 - 28449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y Shoenfeld and M Blank Autoantibodies associated with reproductive failure Lupus, September 1, 2004; 13(9): 643 - 648. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||