ARTICLE IN PRESS

Acta histochemica 112 (2010) 34—41

0065-1281/$ - sdoi:10.1016/j.

�Correspondfax: +381 11 61

E-mail addr

www.elsevier.de/acthis

Effects of anti-phospholipid antibodies on a humantrophoblast cell line (HTR-8/SVneo)

Milica Jovanovica, Milica Bozica, Tamara Kovacevica, Ljiljana Radojcicb,Milos Petronijevicc, Ljiljana Vicovaca,�

aInstitute for Application of Nuclear Energy – INEP, University of Belgrade, Banatska 31b, 11080 Belgrade, SerbiabMilitary Medical Academy, Belgrade, SerbiacMedical Faculty, University of Belgrade, Belgrade, Serbia

Received 23 January 2008; received in revised form 1 July 2008; accepted 17 July 2008

KEYWORDSExtravilloustrophoblast;HTR-8/SVneo;Anti-phospholipidsyndrome;Integrins;Human

ee front matter & 2008acthis.2008.07.001

ing author. Tel.: +381 118 724.ess: vicovac@inep.co.y

SummaryAntibodies to phospholipids (aPL) have been shown to adversely affect trophoblastinvasion in vivo and in vitro. HTR-8/SVneo cells derived from first trimester ofpregnancy extravillous trophoblast were studied. Matrigel invasion assay, cytochem-istry and cell-based enzyme-linked immunosorbant assay (ELISA) with aPL or normalIgG was used. Our data show that aPL at 100 mg/ml decrease invasiveness of HTR-8/SVneo cells to 60% of control (po0.01), and this was also shown for primarycytotrophoblast (to 15.5% of control, po0.001). aPL treatment caused a significantdecrease in integrin a1, a5, and b1 proteins (86%, 84%, and 87%, respectively). Weconclude that HTR-8/SVneo cell culture is a suitable model to study mechanisms ofaction of aPL on trophoblast, which in HTR-8/SVneo cells inhibit invasion bydecreasing integrins a5, a1, and b1.& 2008 Elsevier GmbH. All rights reserved.

Introduction

Several autoimmune disorders have been linkedwith infertility and adverse pregnancy outcomes(Gleicher and El-Roeiy, 1988; Rote and Stetzer,2003). Anti-phospholipid syndrome (APS) is aclinical condition characterized by predisposition

Elsevier GmbH. All rights rese

619 252;

u (L. Vicovac).

to thrombosis, the presence of autoantibodiesreactive with cardiolipin or other negativelycharged phospholipids and phospholipid-bindingproteins such as b2-glycoprotein I (b2-GPI). Repro-ductive repercussions range from infertility andrecurrent miscarriage to placental insuffciency,fetal growth restriction, and pregnancy-inducedhypertension (Branch et al., 1985; Radojcic et al.,2004). It was originally assumed that the mainreproductive complications in APS were caused byincreased predisposition to thrombosis. However,

rved.

ARTICLE IN PRESS

Effects of anti-phospholipid igG on trophoblast cells 35

more recent evidence suggests direct effects of aPLantibodies on trophoblast function.

Implantation of the blastocyst into the uterinewall and development of a functional placenta arecrucial events for the establishment of pregnancy. Innormal placentation, extravillous trophoblast (EVT)cells of the human placenta invade the uterinestroma and develop vascular connections betweenmother and fetus by replacing endothelial cellsand mural vascular smooth muscle cells (Pijnenborget al., 1983). It has been well established that fornormal invasion to occur, integrin switch andadherence to a physiological pattern of expressionof adhesion molecules is critical (Damsky et al.,1994). Integral membrane proteins – integrin a5b1,acting as fibronectin receptor, and integrin a1b1 ascollagen/laminin receptor – of the invasive tropho-blast have been found to be particularly relevant(Damsky et al., 1992). Deviations from the normalpattern have been linked to serious conditions suchas pre-eclampsia (Zhou et al., 1993).

Direct binding of antibodies to phospholipids(aPL) to trophoblast cell membranes has beendemonstrated (Chamley, 1997; Di Simone et al.,2005). This has been shown to induce inhibitionof human chorionic gonadotropin and decreasetrophoblast fusion (Quenby et al., 2005). Cellinvasion was also inhibited by aPL (Katsuragawaet al., 1997; Di Simone et al., 1999), which inisolated term trophoblast was accompanied bymodulation of cell adhesion molecules (Di Simoneet al., 2002). aPL were recently hypothesized tostimulate premature onset of cytotrophoblast pro-liferation and syncytial fusion (Bose et al., 2006)which could adversely affect EVT invasion andplugging of spiral arteries, causing miscarriage.

This study was intended to explore the directeffects of aPL on first trimester human trophoblastusing an immortalized extravillous first trimestercell line HTR-8/SVneo (Graham et al., 1993).Effects of aPL on in vitro trophoblast invasion, cellproliferation, and expression of integrins a1, a5,and b5 by HTR-8/SVneo cell are documented hereindicating that integrins may be involved in theinhibition of trophoblast invasion induced by aPL.This demonstrates the suitability of this cell line forfurther studies of aPL-induced pathophysiology ofthe human trophoblast.

Material and methods

Reagents and antibodies

RPMI 1640 medium and fetal calf serum (FCS)were purchased from PAA Laboratories (Linz,

Austria). Matrigel was purchased from BD Bios-ciences (USA). The following mouse monoclonalantibodies were used: anti-integrin a1 and anti-integrin b1 (Santa Cruz, CA); anti-integrin a5antibody (R&D Systems, UK & Europe); anti-cyto-keratin-18 (Sigma-Aldrich, St. Louis, MO, USA);anti-proliferation marker Ki-67 (Dako, Denmark),anti-cytokeratin 18 fragment M30, as marker ofapoptosis (Roche, Germany). Medium DMEM/F12,trypan blue, thiazolyl blue (MTT), gentamycin,were from Sigma-Aldrich, St. Louis, MO, USA.Biotinylated horse anti-mouse IgG, avidin–biotin-peroxidase complex (ABC) and diaminobenzidine(DAB) substrate kit for peroxidase were from VectorLaboratories, Burlingame, CA, USA. All otherreagents were of the highest commercial gradeavailable. IgG antibodies were isolated in ourlaboratory from plasma of 13 patients with APSpositive for anti-cardiolipin IgG as determined bysolid-phase enzyme-linked immunosorbant assay(ELISA) and 10 healthy normal non-pregnant sub-jects. This was achieved by ammonium-sulfateprecipitation and affinity chromatography on aSepharose 4B column (Pharmacia, Uppsala, Swe-den), as previously described (Bollag and Edelstein,1991).

Cells

The effects of aPL were studied on primarycytotrophoblast and/or HTR-8/SVneo cells. HTR-8/SVneo cells were kindly provided by Dr. CharlesH. Graham (Queen’s University, Kingston, ON,Canada). This cell line was obtained from humanfirst trimester placenta explant cultures immorta-lized by SV40 large T antigen (Graham et al., 1993;Irving et al., 1995). Cytotrophoblast cells wereisolated from first trimester of pregnancy placentasfrom legal abortions undertaken for non-medicalreason at Institute of Obstetrics and Gynaecology,Clinical Center of Serbia, Belgrade in accordancewith the local ethical standards. Primary cytotro-phoblast cells were isolated from first trimesterhuman placentas (6–12 weeks) by trypsin/DNAsedigestion, purified on a Percoll gradient, charac-terized by immunolabelling for cytokeratin-18and cultured as previously reported (Vicovacet al., 2007). HTR-8/SVneo cells were culturedon glass coverslips for immunocytochemistry, in96-well plates for cell-based ELISA and on Matrigel-coated inserts for invasion assay. In each experi-ment, cells were treated with 100 mg/ml aPL orcontrol IgG in DMEM/F12 for primary cytotropho-blast and RPMI 1640 for HTR-8/SVneo cells, all with10% FCS.

ARTICLE IN PRESS

M. Jovanovic et al.36

Invasion assay

An in vitro invasion assay was performed aspreviously described by Librach et al. (1991) withminor modifications. Briefly, transwell inserts(polyethylene terephthalate – PET, track-etchedmembrane-diameter 6.4mm, with 8 mm porosity,Falcon, BD Labware, USA) were coated with 20 mlof growth factor-reduced Matrigel (1:2 dilution)with control or aPL IgG (100 mg/ml) in serum-free medium and incubated at 37 1C for 30min.Cytotrophoblasts (2� 105) or HTR-8/SVneo cells(5� 104) were seeded into transwell chambersin 200 ml of medium with control or aPL IgG. Thesame media (500 ml) were added to the lowerchambers. Cultures were incubated at 37 1C ina humidified atmosphere (5% CO2 in air) for 24 h,then media were removed and cells were washedgently twice with 0.05M phosphate-buffered saline(PBS), pH 7.2. Cells that remained on the upperside of the filters were gently removed with acotton swab. Cells from the lower side of the filtermembrane were fixed with 4% paraformalde-hyde–0.05% Triton X-100 for 30min at roomtemperature. The membranes were cut out fromthe inserts using a scalpel blade and the cells werelabelled for the binding of monoclonal anti-cyto-keratin-18 antibody, as described for immunocyto-chemistry. Membranes were mounted with lowersurface facing up and cytokeratin-18-labelled cellbodies and processes counted using a light micro-scope (Reichert-Jung with Leica DC150 DigitalCamera System, Wetzlar, Germany) in 60 repre-sentative fields. The results were expressed as apercent of control. Experiments were performedthree times in duplicate.

Immunocytochemistry

Cell cultures were washed with PBS, air-driedand fixed with 4% paraformaldehyde with 0.05%Triton X-100. Endogenous peroxidase activity wasblocked with 1% hydrogen peroxide for 30min andnon-specific binding of proteins with 10% non-immune horse serum in PBS for 10min. Cells werethen incubated with primary antibody to cyto-keratin-18 (diluted 1:6000), Ki-67 (diluted 1:10),M30 (diluted 1:25), or anti-integrin a1, a5, and b1antibodies (diluted 1:100) for 1 h. This was followedby incubation with biotinylated horse anti-mouseIgG secondary antibody (diluted 1:200) for 30minand avidin-biotinylated peroxidase complex (ABCVector kit, prepared according to the manufac-turer’s instruction) for further 30min. All stepswere performed at room temperature. All dilutions

and thorough washes between steps were per-formed using PBS, unless otherwise specified.Labelling was visualized using DAB (prepared asspecified by the manufacturer of the kit) for10min. Omission of the primary antibody and useof non-immune serum in place of specific anti-body was incorporated as controls and resultedin complete absence of labelling. Slides werelightly counterstained (for 15 s) using hematoxylin(Vector). Slides were then mounted in Vectamount(Vector) mounting medium and examined using aReichert–Jung microscope with a Leica DC150Digital Camera System (Wetzlar, Germany). Forassessment of proliferating or apoptotic cells,respectively, anti-Ki-67 or M30 labelled cells werecounted per 3000 cells in non-overlapping fields bytwo observers.

Cell-based ELISA (CELISA)

HTR-8/SVneo fixed cell monolayers were rehy-drated with PBS (3� 5min), blocked with 1% bovineserum albumin (BSA, 200 ml/well) for 1 h, andincubated for 1 h with anti-integrin a1, a5, and b1antibodies, 50 ml/well (diluted 1:200 in PBS-1%BSA). Plates were incubated for 30min with50 ml/well biotinylated horse anti-mouse IgG(diluted 1:1000), then with 50 ml/well of avidin–

biotin-peroxidase complex (ABC Vector kit, pre-pared according to the manufacturer’s instructions)for another 30min. Wells were incubated with50 ml/well substrate (0.003% H2O2) and 50 ml/wellof chromogen (0.05% 3,30, 5,50-tetramethylbenzi-dine, TMB) for 5min. Wells were washed (5� 5min)with PBS containing 0.05% Tween-20 between eachstep, and all stages were performed at roomtemeprature. The reaction was stopped with 0.2MH2SO4 (100 ml/well). Absorbance was measured at450 nm using a Microplate reader (LKB). Absor-bances measured were normalized to cell number.For assessment of non-specific binding, cells wereincubated in non-immune serum in place of specificantibodies. The absorbance for non-specific bindingwas substracted from all absorbances. All experi-ments were repeated at least three times, nX6.

Determination of viable cell number

MTT ([3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyl-tetrazolium bromide; thiazolyl blue]) test was usedas an indicator of cell number (Hanisch et al.,1993). HTR-8/SVneo cells were plated in 96-wellplates and incubated with control or aPL IgG, asdescribed above. One hundred microliters of MTT in10% FCS/PBS (1mg/ml) was then added to each

ARTICLE IN PRESS

Effects of anti-phospholipid igG on trophoblast cells 37

well. After incubation for 2 h at 37 1C, medium wasreplaced by 1-propanol (100 ml/well) and the plateswere vigorously shaken to ensure complete solubi-lization of the blue formazan. Absorbance wasmeasured at 570 nm using a microplate reader(LKB) and cell numbers were determined using astandard curve obtained with 5� 103, 1� 104,2� 104, 4� 104, 6� 104 or 8� 104 cells/well.

Statistics

Statistical analysis of data was carried out usingthe Statistical Software Program version 5.0(Primer of Biostatistic, McGraw-Hill Companies, Inc.,New York, NY, USA) using non-parametric Mann–

Whitney Rank Sum Test with values consideredsignificantly different when po0.05.

Figure 1. The effect of aPL IgG (100 mg/ml) on cellinvasion by (A) freshly isolated cytotrophoblast from firsttrimester of pregnancy placenta and (B) HTR-8/SVneocells. Pores occupied by cells and cell processes werecounted after 24 h culture. The data are expressed aspercent of pores occupied in control, values are given asmean7SEM, differences significant at po0.001 forprimary trophoblast (n ¼ 4), and po0.05 for HTR-8/SVneo cells (n ¼ 4).

Results

For the purpose of this study, preparation of IgGwith aPL was made from sera of patients with APSpreviously found positive for anti-cardiolipin anti-bodies (aCL). Prepared IgG was used on isolatedfirst trimester of pregnancy trophoblast and HTR-8/SVneo cells (Figure 1). Cell purity of isolatedtrophoblast determined by immunolabelling withanti-cytokeratin-18 was 490%. Treatment with aPLat 100 mg/ml for 24 h induced inhibition of Matrigelinvasion by cultured first trimester cytotrophoblastdown to 15.6% (po0.001) of control incubated withIgG from pooled normal at 100 mg/ml (Figure 1A).To ascertain the suitability of HTR-8/SVneo cells asa model for studying the effects of aPL on earlypregnancy trophoblast, a Matrigel invasion experi-ment was performed. Under the same conditions,IgG with aPL at 100 mg/ml, inhibited an invasion byHTR-8/SVneo cell line significantly (po0.05) downto 60.2% of control (Figure 1B). Integrin subunits a5,a1, and b1 were labeled in HTR-8/SVneo cellsby immunocytochemistry, and all were detected.Immunolabeling for a1 and b1 is shown in Figure 2Aand B, respectively. When cultured in medium withaPL (100 mg/ml), less integrin a1 and b1 labeling wasseen (Figure 2C and D, respectively). In order toquantify the effects of aPL IgG on the level ofintegrin subunit a1, a5, and b1, a cell-based ELISAtest was performed. Absorbance values werenormalized for cell numbers and the result ex-pressed as a percent of control (Figure 3). The cell-based ELISA test was typically run after 48 h ofculture, but comparable results were obtainedafter 24 h (data not shown). The data obtainedshow that the integrins are inhibited when cells are

cultured in the presence of aPL IgG at 100 mg/ml,for subunit a1 to 86.6% (Figure 3A) of control(po0.005), and for subunits a5 and b1 to 83.9%(po0.001) and 86.8% (po0.001), respectively(Figure 3B and C).

Since HTR-8/SVneo cells divide in culture,possible effects of aPL IgG on cell proliferationwere determined based on the MTT test (Figure 4).Results show that aPL induced an increase in HTR-8/SVneo cell proliferation, which was small (108.1%of control), but significant (po0.001). In order toexclude the possibility that this increase was due toincreased enzyme activity measured by MTT test,immunocytochemical identification of proliferatingcells using Ki-67, as a marker, was performed oncontrol and aPL-treated cell cultures. Data shownin Figure 5C and D illustrates a higher proportionof Ki67 immunopositive cells after treatment withaPL IgG (increased by 38% over control upon cellcounting). A possibility that aPL could induceapoptosis was studied by assessment of caspase-cleaved epitope of cytokeratin-18 cytoskeletal

ARTICLE IN PRESS

Figure 2. Immunolocalization of integrin subunits a1 (A, C), and b1 (B, D) in HTR-8/SVneo cell control (A, B) andcultures treated with aPL IgG (C, D) at 100 mg/ml. Non-specific control (E). Scale bar ¼ 3 mm.

M. Jovanovic et al.38

protein by cytochemistry for M30 (Figure 5A and B).No apoptotic effect was seen upon treatment withaPL IgG when compared to the control.

Discussion

Direct effects of aPL on EVT were studied here,using a the HTR-8/SVneo cell line which wasderived from human invasive EVT (Graham et al.,1993). The observed phenotype and physiologicalcharacteristics are consistent with that of normal

first trimester invasive trophoblast cells (Irvinget al., 1995). HTR-8/SVneo cells have been previouslyused in models of EVT migration (Gleeson et al.,2001; Chakraborty et al., 2003) and invasion (Huberet al., 2006). In this study, we confirmed that thecell line shows an extravillous phenotype withrespect to expression of the integrin subunits, a1,a5, and b1. Anti-phospholipid antibodies, used inthis study, were isolated from sera positive for aCL.Other studies (McNeil et al., 1990; Hunt and Krilis,1994) have established that antibodies detectedwith aCL assays in APS patients consist largely ofantibodies against b2-glycoprotein I (b2-GPI). It was

ARTICLE IN PRESS

Figure 3. The effect of aPL IgG (100 mg/ml) on integrina1 (A), a5 (B) and b1 (C) subunit levels in cell-based ELISA.Data are expressed as percent of untreated control,values given as mean7SD (n ¼ 10), differences signifi-cant at po0.01 for a1 and po0.001 for a5 and b1subunits.

Figure 4. Effect of aPL IgG (100 mg/ml) on viable HTR-8/SVneo cell number determined by MTT test after 48 hof culture. Data are expressed as percent of controlrepresented as mean7SD (n ¼ 36), significant atpo0.001.

Effects of anti-phospholipid igG on trophoblast cells 39

therefore assumed that the IgG preparation used inthis study was a mixture of anti-phospholipid/anti-protein cofactor antibodies. The IgG isolated in ourlaboratory was shown effective in inhibition ofcytotrophoblast invasion in vitro, as reported in theliterature. Our observations demonstrate, for thefirst time, that the invasion of HTR-8/SVneo cellsin the presence of aPL is also inhibited. This isconsistent with data shown here and in theliterature for isolated cytotrophoblast using aMatrigel invasion test (Di Simone et al., 1999). Itshould be noted, however, that invasion by freshlyisolated cytotrophoblast cells of first trimester ofpregnancy placentas is reduced more than in theHTR-8/SVneo cell line. Isolated trophoblast hasbeen previously shown to bind exogenous aPL IgG invitro. This binding was linear and was not saturatedin the concentration range 3–400 mg/ml (Di Simoneet al., 1999). The binding capacity for aPL in theHTR-8/SVneo cell line has not been studied.

Immunocytochemistry and cell-based ELISAshowed that the integrins relevant for cell adhesionand migration in HTR-8/SVneo cells, subunits a1, a5,and b1, are inhibited at the protein level, whencells are cultured with IgG with aPL reactivity.Consequently, reduction in expression of integrinsinvolved in attachment to fibronectin and laminin,extracellular proteins present at implantation sitesin vivo and in the Matrigel invasion model, couldcontribute to the observed inhibition of invasion invitro. This finding is in keeping with the datareported for primary culture of term cytotropho-blast cells that aPL antibodies modulate integrinsand expression of other cell adhesion molecules,which may lead to inadequate invasion (Di Simoneet al., 2002). These authors have also establishedthat aPL significantly reduced the expression of a1integrin at the protein and RNA level.

It has been recently hypothesized that aPLantibodies may stimulate the premature onset ofcytotrophoblast proliferation and syncytial fusion(Bose et al., 2006), which could result in exhaustionof cytotrophoblast ‘stem’ cells, adversely affectingboth villous syncytiotrophoblast function and EVTinvasion. Our model based on proliferative tropho-blast cells, as opposed to invasive trophoblast cellsin vivo or isolated cytotrophoblast in vitro thatdo not proliferate, was well suited to test thishypothesis. In this study, treatment of HTR8/SVneocells with aPL was found to induce an increase incell proliferation with no change in the rate ofapoptosis, as evidenced by caspase-cleaved epi-tope of cytokeratin-18 (M30). Studies utilizingprimary cultures of trophoblast have also foundno change in apoptosis, as evidenced by DNA frag-mentation or M30 cell positivity (Di Simone et al.,

ARTICLE IN PRESS

Figure 5. The effect of aPL IgG (100 mg/ml) on apoptosis (A, B) and cell proliferation (C, D) as evidenced byimmunocytochemistry using antibodies against M30 and Ki 67, respectively. Arrows in A and B point to an apoptotic cell.Scale bar ¼ 10 mm.

M. Jovanovic et al.40

2006), even though Bcl-2/Bax ratio was reduced inaPL-treated trophoblast. The mechanisms respon-sible for inhibition of integrin production in thepresence of aPL are currently not known. However,in endothelial cell models, human monoclonal aCL/b2-GPI was shown to induce cellular effects viaphosphorylation of p38, and translocation of NF-kBinto the nucleus and gene regulation (Bohgakiet al., 2004). The data presented here confirmthat the inhibition of integrins is one of the mecha-nisms in aPL-induced inhibition of trophoblastinvasion, and that the HTR-8/SVneo cell line couldbe used to facilitate understanding of placenta-related aspects of pathophysiology of APS.

Acknowledgment

This work was supported by the research grant143059 of the Ministry of Science, Republic ofSerbia.

References

Bohgaki M, Atsumi T, Yamashita Y, Yasuda S, Sakai Y,Furusaki A, et al. The p38 mitogen-activated protein

kinase (MAPK) pathway mediates induction of thetissue factor gene in monocytes stimulated withhuman monoclonal anti-beta 2 glycoprotein I anti-bodies. Int Immunol 2004;16:1633–41.

Bollag DM, Edelstein SJ. Concentrating protein solutions.In: Bollag DM, Edelstein SJ, editors. Protein methods.New York, Chichester, Brisbane, Toronto, Singapore:Wiley; 1991. p. 71–92.

Bose P, Kadyrov M, Goldin R, Hahn S, Backos M, Regan L,et al. Aberrations of early trophoblast differentiationpredispose to pregnancy failure: lessons from the anti-phospholipid syndrome. Placenta 2006;27:869–75.

Branch DW, Scott JR, Kochenour NK, Hershgold E.Obstetric complications associated with the lupusanticoagulant. N Engl J Med 1985;313:1322–6.

Chakraborty C, Barbin YP, Chakrabarti S, Chidiac P, DixonSJ, Lala PK. Endothelin-1 promotes migration andinduces elevation of [Ca2+]i and phosphorylation ofMAP kinase of a human extravillous trophoblast cellline. Mol Cell Endocrinol 2003;201:63–73.

Chamley LW. Antiphospholipid antibodies or not? The roleof beta-2 glycoprotein I in autoantibody-mediatedpregnancy loss. J Reprod Immunol 1997;36:123–42.

Damsky CH, Fitzgerald ML, Fisher SJ. Distribution patternsof extracellular matrix components and adhesion re-ceptors are intricately modulated during first trimestercytotrophoblast differentiation along the invasive path-way, in vivo. J Clin Invest 1992;89:210–22.

ARTICLE IN PRESS

Effects of anti-phospholipid igG on trophoblast cells 41

Damsky CH, Librach C, Lim KH, Fitzgerald ML, McMasterM, Janatpour M, et al. Integrin switching regulatesnormal trophoblast invasion. Development 1994;120:3657–66.

Di Simone N, Caliandro D, Castellani R, Ferrazzani S, DeCarolis S, Caruso A. Low molecular weight heparinrestores in vitro trophoblast invasiveness and differ-entiation in presence of immunoglobulin G fractionsobtained from patients with antiphospholipid syn-drome. Hum Reprod 1999;14:489–95.

Di Simone N, Castellani R, Caliandro D, Caruso A.Antiphospholipid antibodies regulate the expressionof trophoblast cell adhesion molecules. Fertil Steril2002;77:805–11.

Di Simone N, Raschi E, Testoni C, Castellani R, D’Asta M,Shi T, et al. Pathogenic role of anti-beta 2-glycopro-tein I antibodies in phospholipid associated fetal loss:characterisation of anti-beta 2-glycoprotein I bindingto trophoblast cells and functional effects of anti-beta2-glycoprotein I antibodies in vitro. Ann Rheum Dis2005;64:462–7.

Di Simone N, Castellani R, Raschi E, Borghi MO, MeroniPL, Caruso A. Anti-beta-2 glycoprotein I antibodiesaffect Bcl-2 and Bax trophoblast expression withoutevidence of apoptosis. Ann N Y Acad Sci 2006;1069:364–76.

Gleeson L, Chakraborty C, McKinnon T, Lala PK. Insulin-like growth factor-1 stimulates human trophoblastmigration by signaling through alpha5 beta1 integrinvia mitogen activated protein kinase pathway. J ClinEndocrinol Metab 2001;86:2484–93.

Gleicher N, El-Roeiy A. The reproductive autoimmunefailure syndrome. Am J Obstet Gynecol 1988;159:223–7.

Graham CH, Hawley TS, Hawley RC, MacDougall JR,Kerbel RS, Khoo N, et al. Establishment and char-acterization of first trimester human trophoblast cellswith extended lifespan. Exp Cell Res 1993;206:204–11.

Hanisch FG, Dressen F, Uhlenbruck G. Quantitative micro-adhesion assay on polystyrene matrices. In: Gabius HJ,Gabius S, editors. Lectins and glycobiology. Berlin,Heidelberg: Springer; 1993. p. 411–7.

Huber V, Saleh L, Bauer S, Husslein P, Knofler M.TNFalpha-mediated induction of PAI-1 restricts inva-

sion of HTR-8/SVneo trophoblast cells. Placenta 2006;27:127–36.

Hunt J, Krilis S. The fifth domain of beta 2-glycoprotein Icontains a phospholipid binding site (Cys281-Cys288)and a region recognized by anticardiolipin antibodies.J Immunol 1994;152:653–9.

Irving JA, Lysiak JJ, Graham CH, Hearn S, Han VK, LalaPK. Characteristics of trophoblast cells migrating fromfirst trimester chorionic villus explants and propa-gated in culture. Placenta 1995;16:413–33.

Katsuragawa H, Kanzaki H, Inoue T, Hirano T, Mori T, RoteNS. Monoclonal antibody against phosphatidylserineinhibits in vitro human trophoblastic hormone produc-tion and invasion. Biol Reprod 1997;56:50–8.

Librach CL, Werb Z, Fitzgerald ML, Chiu K, Corwin NM,Esteves RA, et al. 92-kD type IV collagenase mediatesinvasion of human cytotrophoblasts. J Cell Biol 1991;113:437–49.

McNeil HP, Hunt JE, Krilis SA. New aspects of anti-cardiolipin antibodies. Clin Exp Rheumatol 1990;8:525–7.

Pijnenborg R, Bland JM, Robertson WB, Brosens I.Uteroplacental arterial changes related to interstitialtrophoblast migration in early human pregnancy.Placenta 1983;4:397–413.

Quenby S, Mountfield S, Cartwright JE, Whitley GS,Chamley L, Vince G. Antiphospholipid antibodiesprevent extravillous trophoblast differentiation. FertilSteril 2005;83:691–8.

Radojcic L, Marjanovic S, Vicovac L, Kataranovski M.Anticardiolipin antibodies in women with unexplainedinfertility. Physiol Res 2004;53:91–6.

Rote NS, Stetzer BP. Autoimmune disease as a cause ofreproductive failure. Clin Lab Med 2003;23:265–93.

Vicovac L, Bozic M, Bojic-Trbojevic Z, Golubovic S.Carcinoembryonic antigen and related molecules innormal and transformed trophoblast. Placenta 2007;28:85–96.

Zhou Y, Damsky CH, Chiu K, Roberts JM, Fisher SJ.Preeclampsia is associated with abnormal expressionof adhesion molecules by invasive cytotrophoblasts.J Clin Invest 1993;91:950–60.