pathophysiology of autoimmune-associated congenital heart ... · applied cardiopulmonary...

96 Y. Qu, M. Boutjdir

Applied Cardiopulmonary Pathophysiology 16: 96-112, 2012

Pathophysiology of autoimmune-associated congenitalheart block

Yongxia Qu1,2, Mohamed Boutjdir1,2,3

1VA New York Harbor Healthcare System, Brooklyn, NY; 2State University of New YorkDownstate Medical Center, Brooklyn, NY and 3New York University School of Medicine,New York, NY, USA

Abstract

Congenital heart block (CHB), detected at or before birth, in a structurally normal heart, isstrongly associated with autoantibodies to SSA/Ro-SSB/La ribonucleoproteins. The historicalhallmark of CHB is complete atrioventricular block (AVB) which is irreversible. CHB occurs in2% of primigravid mothers with anti-Ro/La antibodies, and in 20% of women who had a pre-vious CHB offspring. Although not consistent, emerging clinical research expanded the spec-trum of CHB to include lesser degrees of AVB, sinus bradycardia, QT prolongation, and lateonset cardiomyopathy. Despite over a century since the discovery of CHB, the wide spectrumof CHB manifestation has posed a challenge in the elucidation of the true underlying patho-genesis and thus effective therapy. This review focuses on the recent knowledge and some hy-potheses proposed for CHB pathogenesis, mainly the apoptosis and the Ca channel hypothe-ses. Experimentally, the first challenge in establishing the association between anti-Ro/La anti-bodies and CHB was overcome by the development of in vivo animal models for CHB. The in-cidence of conduction abnormalities in these models was fairly similar to that clinically, indi-cating that anti-Ro/La antibodies are essential but not sufficient for the disease expression. Sub-sequent in vitro experiments demonstrated that maternal anti-Ro/La antibodies directly crossreact and inhibit Ca channels. This inhibition of Ca channels provides a logical explanation forthe AVB, sinus bradycardia and perhaps even the cardiomyopathy because Ca channels playa vital role in the action potential genesis and conduction at the atrioventricular node and si-nus node; and in excitation-contraction coupling of the developing heart, respectively. BecauseRo/La antigens are intracellularly located, apoptosis was proposed as a process that helpsRo/La antigens translocate to the fetal cell surface membrane where they will be accessible tocirculating maternal antibodies. In this regard, extensive evidence showed that in experimen-tally induced apoptotis in human fetal ventricular myocytes, the intracellular Ro/La antigenstranslocate to the sarcolemma, interact with anti-Ro/La antibodies, divert normal clearance ofapoptotic cardiac myocytes by healthy cardiac myocytes toward clearance by professionalmacrophages with the release of the inflammatory and/or fibrosing cytokines which leads tofibrosis and permanent CHB. Despite the significant progress made in the last decade in theunderstanding of the CHB pathogenesis, the wide spectrum of CHB expression point to themultifactorial nature of CHB pathogenesis and to the need for continuous and joint interna-tional efforts to dissect the complex pathways involved in CHB.

Key words: atrioventricular block, neonatal lupus, anti-Ro/SSA antibodies, Ca channels, apop-tosis

97Pathophysiology of autoimmune-associated congenital heart block

Abbreviations

5-HT4: Serotonin 4 receptorsAVB: Atrioventricular blockCHB: Congenital heart block CHB-IgG: Maternal IgG containing anti-

Ro/La antibody from motherswith CHB children

CM: Cardiomyopathy ECG: Electrocardiogram EFE: Endocardial fibroelastosis ET: Endothelin Fc R: Fc gamma receptorGST: Glutathione S-transferase IC: Immune complexes IKs: The delayed rectifier K current INa: The fast Na current Ito : Transient outward K currentLV: Left ventricle NLS: Neonatal Lupus SyndromesPRIDE: PR Interval and Dexamethasone

EvaluationQTc: Corrected QT intervalRV: Right ventricleSA : Sinoatrial siRNA: Small interfering RNAssRNA: Single strand RNATGFβ: Transforming growth factor betaTLR7: Toll-like receptor 7uPA: Urokinase plasminogen activatoruPAR: Urokinase plasminogen activator

receptor

Introduction

Autoimmune-associated congenital heartblock (CHB) which occurs in the absence ofstructural abnormalities is a passively ac-quired autoimmunity consequent to placen-tal transport of antibodies reactive withSSA/Ro-SSA/La ribonucleoprotein complex(1, 2). CHB is a manifestation of the so calledneonatal lupus syndromes (NLS). Third de-gree atrioventricular block (AVB) is the hall-mark of CHB and carries a significant mortal-ity (20-30%) and morbidity (67% of survivingaffected children require permanent pace-maker before adulthood (2-8). Other mani-

festations of NLS include cutaneous, liverand hematological disease, all of which gen-erally disappear with the clearance of mater-nal antibodies from the neonatal circulationat about 6-8 months (2). The most commonwindow period for CHB detection is be-tween 20 and 24 weeks gestation, which isrelevant for timing of fetal surveillance (2).Emerging clinical studies show that conduc-tion abnormalities seen in CHB have expand-ed beyond AVB to include sinus bradycardia(9-11) although controversial (12), QT ab-normalities (13), and late onset fatal car-diomyopathy (14-16) as late as 10 years ofage. The incidence of CHB is low in the gen-eral population (1 of every 15,000) but in-creases to 5% in women with Lupus (2).However, the recurrence rate for CHB in asubsequent pregnancy is alarming and ap-proaches 20% (2), affecting the decision ofthe mother to have another child. A networkof research registries worldwide is needed toenable the acquisition of these rare diseasecases and to facilitate further the investiga-tion of epidemiology, clinical and basic re-search aspects of CHB. To date, and despitethe significant progress made in the lastdecade at all fronts, CHB continues to pose achallenge to both basic scientists and clini-cians in the understanding of the true patho-genesis underlying the disease.

Spectrum of CHB

First degree AVB and progression ofconduction abnormalities

Historically, the signature lesion of CHB iscomplete AVB. Over the years, as the tech-nology to diagnose CHB improved, a lesserdegrees of AVB were readily detectable inutero. Existing data regarding the true inci-dence of first degree AVB and its potential topredict later progression to more advancedAVB is conflicting (17, 18, 15, 19, 20). A highincidence of first degree AVB was observedby Sonesson et al. (20), in a prospectivestudy. Twenty-four Ro 52 seropositive


women were followed up weekly between18 and 24 weeks of gestation, with twoDoppler echocardiographic methods de-signed to estimate the time delay betweenhemodynamic events caused by atrial andventricular depolarization. Two hundredeighty-four women with normal pregnanciesserved as controls. A PR interval >135 mswas considered abnormal. In Ro 52 seropos-itive women, eight of 24 fetuses had signs offirst degree AVB, one of which progressed tocomplete AVB and six spontaneously revert-ed to normal conduction before or shortly after birth. The authors concluded that Ro 52seropositive pregnant women frequently car-ry fetuses with first degree AVB and progres-sion to a more severe degree of block mayoccur in some. This observation was not re-produced by Freidman et al., in the PRIDEstudy (19). Ninety five Ro 52 seropositivewomen had fetal echocardiograms per-formed weekly from 16 to 26 weeks gesta-tion and biweekly from 26 to 34 week in 98pregnancies. PR intervals of >150 ms wasused as cutoff for first degree AVB. Of 98 fe-tuses, only two fetuses had first degree AVBdetected at or before 22 weeks, and each re-verted within 1 week with 4 mg dexametha-sone. Ninety-two fetuses had normal PR in-tervals. Three fetuses had complete AVB,none of whom had a preceding abnormal PRinterval. The conclusion was that prolonga-tion of the PR interval was uncommon anddid not precede more advanced block.

Re-evaluation of the PRIDE study usingthe cut off value of 135 ms as PR interval pro-longation revealed consistency between thetwo studies, with about one-third of the fetus-es in PRIDE having a prolonged PR interval.Interestingly, in the PRIDE study, all fetuseswith PR interval of 135-150 ms spontaneous-ly reversed by the next echocardiogram. InSonesson’s study (20), however, only two fe-tuses had PR prolongation as defined byPRIDE criteria. In summary, the discrepancybetween the two studies was mostly relatedto the difference in the definition of PR pro-longation. Mild PR prolongation might becommon and can revert spontaneously. A

universal cutoff for a pathogenic PR-intervalto predict disease progression and standardi-zation of the diagnostic approach for fetalAVB is warranted.

Sinus bradycardia

Sinus bradycardia unrelated to AVB was firstreported in animal models of CHB (21-23),and in the in vitro experiments using Langen-dorff-perfused isolated hearts (21, 22, 24).These observations were later confirmed insome clinical cases by Brucato et al. (11), andHamilton’s group (10). Brucato and col-leagues (11) reported a significant transientsinus bradycardia in 4 infants among 24 oth-erwise healthy children from anti-Ro anti-body positive mothers whose ECGs were ob-tained within the first 3 days of life. In all cas-es, sinus bradycardia disappeared within 10days after birth, with no sequelae. An 11% ofchronotropic incompetence of the sinusnode was found in infants with AVB frommothers with anti-Ro antibodies by Menon etal. (10). However, Costedoat-Chalumeau didnot find any significant difference in meanheart rate when comparing the ECGs of 58anti-Ro-positive children with those of 85 an-ti-Ro-negative children of the same age (25).Although sinus bradycardia has been consis-tently demonstrated in animal models, theclinical sinus bradycardia may be a rare andseemingly reversible part of the spectrum ofanti-Ro/La antibody related cardiac disease.Its prediction of further cardiac conductionabnormalities in CHB has not been fully illu-cidated.

Prolongation of QT interval

QT prolongation has been reported in infantsfrom mothers with anti-Ro/La antibodies (13,26), but these observations have not beenconfirmed by others (19, 25). QT prolonga-tion in adults positive for anti-Ro/La antibod-ies has also been reported (27-29). Cimazand colleagues (13) reported a mean QTc


prolongation in the absence of CHB in 21children born to anti-Ro-positive motherswhen compared to 7 children born to anti-Ro-negative mothers. QTc prolongation wasresolved during the first year of life. Coste-doat-Chalumeau et al. (25), have addressedthe same issue in a study that comparedECGs in 58 consecutive children aged 0 to 2months and born to anti-Ro-positive motherswith a carefully defined control group of 85infants aged 0 to 2 months born to anti-Ro-negative mothers with connective tissue dis-ease. No difference was found for QTc be-tween these two groups. Interestingly, themean QTc interval recorded during the peri-od from 2 to 4 months showed a significantlengthening in comparison with those ob-tained during the period from 0 to 2 monthsin both anti-Ro-positive and anti-Ro-negativegroups. In agreement with this, Schwartz andcolleagues (30) have shown that there was aphysiological lengthening of the QTc intervalat the second month of life in a prospectivestudy involving 4205 healthy newborns. Thiscould explain why Cimaz and colleagues(13) found a QTc prolongation in the anti-SSA/Ro-positive group in which ECGs wererecorded at 90 days, compared with 7 daysfor the anti-SSA/Ro-negative controls.

Two prospective studies of QTc prolonga-tion have been recently published. Motta etal. (12) compared the ECGs obtained from51 infants born to anti-Ro positive motherswith those from 50 control infants frommothers with connective tissue disease butwere negative for anti-Ro antibodies. MeanQTc of infants born from anti-Ro positivemothers was slightly prolonged but did notreach the statistical significance when com-pared with controls. In another study byGerosa et al. (31), 60 anti-Ro positive and36 anti-Ro negative mothers were prospec-tively followed before/during pregnancy andunderwent weekly fetal echocardiographyfrom 18th to 26th week gestation. Infants’ ECGand/or ECG-holter were performed at 1, 3, 6and 12 months. ECGs of 200 consecutiveneonates were used as a healthy controlgroup. No differences in the prevalence of

QTc interval prolongation (>440 ms) was ob-served between the anti-Ro-positive and neg-ative groups. ECG-holter showed QTc prolon-gation >440ms in 59% infants of anti-Ro pos-itive mothers, 60% in controls, QTc>470msin four infants of anti-Ro positive group andtwo infants in controls. Genetic causes ofQTc prolongation were excluded. These stud-ies showed that QTc prolongation is frequentin infants with mothers with autoimmune dis-ease, independent of maternal antibody pro-file. Thus, the current data do not show con-vincing evidence for the association of anti-Ro/La with QTc prolongation in the infants.Multinational studies that will systematicallyinvestigate the QTc prolongation as anotherpotential spectrum of conduction abnormali-ties in infants with CHB are warranted.

Cardiomyopathy

Approximately 15-20% of CHB affected fe-tuses develop more diffuse myocardial dis-ease manifested as cardiomyopathy (CM)usually associated with endocardial fibroelas-tosis (EFE) (32-34), with or without clinicalconduction abnormalities. The prognosis forfetuses and infants with diffuse CM/EFE isgenerally poor, with death or need for car-diac transplantation in 85% of the cases de-spite pacemaker therapy (14, 33, 34, 35).Nield et al. (34), described 13 children withCHB, associated with EFE predominantly in-volving the left ventricle. Severe ventriculardysfunction was present in all cases and ledto death in nine and cardiac transplantationin two. The same year, those authors also re-ported 3 cases of severe EFE, mainly ventric-ular, in children without CHB born to moth-ers with anti-Ro antibodies (33). Guettrot-Im-bert et al. (36), in contrast, recently reported5 cases of mild EFE from 4 mothers with an-ti-Ro antibodies, detected at ages between22-24 week gestation. The main EFE affectedsites were the left atrium, aortic and tricuspidannuli. Only one case had left ventricular in-volvement. The appearance of EFE on fetalechocardiography remained stable, regard-


less of treatment (2 cases received be-tamethasone). One pregnancy was mediallyaborted at 30 weeks. During a 5 year followup, one case had most severe EFE with signif-icant right ventricular involvement, requiringsurgery at age 6. The remaining 3 casesshowed satisfactory development. The pre-sentations in this case report contrast withprevious reports about the severity of EFE(32-34). Two similar cases were reported byRaboisson et al. (37) and Pises et al. (38).Roboisson et al. (37) reported one case RVEFE associated with first degree AVB and asecond case of isolated LV/RV EFE withoutany AV conduction prolongation from infantsborn to mothers with anti-Ro/La antibodies.In both cases, the fetal LV function was nor-mal and EFE was resolved after dexametha-sone treatment. Pises et al. (38) reported aprenatal diagnosis of right ventricular EFEwithout left ventricular involvement and with-out complete AVB in a fetus of a motherwith anti-Ro/La antibodies. Improvement ofEFE with maternal dexamethasone therapywas similar to three patients reported byRaboisson et al. (37). Of note is that none ofthe cases from Guettrot-Imbert’s study wereassociated with CHB (36). The finding of EFEin the absence of AVB and the evolution oflate CM/EFE despite adequate pacing sug-gest that CM/EFE and AVB may be two sepa-rate disease manifestations in NLS but acausal relationship cannot be ruled out. Insummary, despite early institution of cardiacpacing, some infants with CHB will developcardiomyopathy. Patients with CHB requireclose follow-up, not only for their cardiac rateand conduction abnormalities, but also fortheir ventricular function.

Pathogenesis of CHB

Role of anti-Ro/La antibodies in thepathogenesis of CHB

While almost all CHB diagnosed before theage of 26 weeks gestation were from moth-ers with anti-Ro/La antibodies, the risk of

CHB is only 2% in mother known to have an-ti-Ro/La antibodies without prior childrenwith CHB or rash (2). It is thus logical to ques-tion whether the anti-Ro/La antibody is justan innocent bystander or is there any causalrelationship between anti-Ro/La antibodiesand CHB. In this regard, a number of in vivoand ex vivo animal models were developedto investigate the potential role of anti-Ro/Laantibodies in the pathogenesis of CHB.

Passive mice model of CHB

To test a potential association between thepresence of anti-Ro/La antibodies and theelectrocardiographic abnormalities in CHB,a passive model of CHB was developed inmice (23). Timed pregnant mice were inject-ed with anti-Ro/La antibodies purified frommothers with CHB children. ECG screeningof the pups from the injected mothersshowed sinus bradycardia and AVB whencompared to controls. First-degree AVB wasobserved in 88% (14/16), 90% (9/10) and47% (14/30), and sinus bradycardia was al-so present in 44% (7/16), 70% (7/10) and33% (10/30) of pups from mothers injectedat 8, 11 or 16 days gestation, respectively. In-terestingly, no complete AVB was observed.The greater percentage and degree of sinusbradycardia and PR prolongation in the 11-day group correlates with the „ window peri-od“ of susceptibility observed in humans.The high incidence of sinus bradycardia sug-gests possible sinoatrial (SA) node involve-ment. No follow-up ECGs were performed toexamine the natural progression of the ob-served lesser degrees of AV block and sinusbradycardia in this study.

Active mice model of CHB

Unlike the passive model of CHB where micewere directly injected with anti-Ro/La anti-bodies, the active model of CHB was devel-oped by immunizing female mice and rabbitwith human or murine Ro 52, Ro 60 or La 48


antigens (21, 39-41). The average incidenceof complete AVB was 2.5% which is similarto the incidence reported in humans. Specif-ically, in the first study by Boutjdir et al. (21),female mice were immunized with Ro 52 re-combinant protein and ECG was recordedfrom pups at birth. Nine out of 20 pups fromthe immunized mothers had conduction sys-tem abnormalities but none of the 22 controlpups exhibited conduction abnormalities.Four pups had sinus bradycardia, three pupshad PR prolongation and two pups had com-plete AV dissociation. Subsequently, Miran-da-Carus et al. (40), also identified a spec-trum of AV nodal conduction abnormalitieson the ECG tracings obtained from pupsborn to mice immunized with Ro/La com-plex but not in the control groups. PR pro-longation was observed in 7% of pups bornto mothers immunized with 48 kDa La, in6–7% of pups born to 52α Ro and 52β Ro-immunized mothers, 20% of pups born to 60kDa Ro-immunized mothers, and 9% of pupsborn to murine 52 kDa Ro-immunized moth-ers. Second degree AVB was observed in 2%of offspring from 52α Ro-immunized moth-ers. Complete AVB was observed in 2–6% ofthe pups from 52α Ro and 52β Ro-immu-nized mothers. Suzuki et al. (41) demonstrat-ed that maternal immunization with 60kDaRo, 48 kDa La or recombinant calreticulin inmice resulted in 9-18% second degree AVBand 0% in controls.

Salomonsson et al. (42) were using a pep-tide from the Ro 52 spanning amino-acids200-238 referred to as p200 to immunize fe-male rats. PR prolongation was induced in19% (10 ⁄ 52) of rat pups. No second- or thirddegree AVB was reported in this rat model(42). Xiao et al. (39) immunized seven femalerabbits with human Ro 52 antigen. Of 152pups, 31 pups (20.4%) were born dead, 1pup (0.7%) had second degree AVB, 7 pups(4.6%) had sinus bradycardia, 8 pups (5.3%)had PR prolongation, and 5 pups (3.3%) hadboth sinus bradycardia and AVB. The remain-ing 100 pups had normal sinus rhythm andnormal PR interval.

Altogether, the results from both the pas-sive and active animal models of CHB pro-vide evidence for a necessary pathogenicrole of maternal anti-Ro/La antibodies in thedevelopment of CHB.

Maternal anti-Ro/La antibodies

induce sinus bradycardia and AVB in

isolated Langendorff-perfused hearts

While the consequence of chronic exposureof fetal heart to anti-Ro/La antibodies wereexamined in animal models as describedabove, the acute effects of anti-Ro/La anti-bodies on the isolated completely denervat-ed hearts were tested on Langendorf-per-fused human fetal (21) and rat hearts (22).Perfusion of these hearts with IgG containinganti-Ro/La antibody from mothers with CHBchildren (CHB-IgG) resulted first in sinusbradycardia, followed by first and second de-gree AVB that subsequently degenerated intocomplete AVB within 15-20 min of CHB-IgGperfusion. In contrast, perfusion of the heartwith normal IgG devoid of anti-Ro/La anti-bodies from healthy mothers with healthychildren did not alter ECG parameters (21,22). The sinus bradycardia and/or AVB weresimilarly demonstrated in Langendorff-per-fused hearts by others (9, 24, 43, 44) indi-cating that the electrocardiographic abnor-malities seen in CHB can be induced by anacute initial effect which may be distinctmechanistically from the chronic effect asso-ciated with long term exposure of the heartto maternal antibodies.

Collectively, animal models and in vitroexperiments provide strong evidence for apathogenic role of anti-Ro/La antibodies inthe development of CHB. The spectrum ofconduction abnormalities was varied and in-cluded a greater incidence of first degreeAVB than that reported in humans. The rateof advanced degree of AVB approximatedthe 1–5% risk for a mother with anti-Ro/Laantibodies to have a child with CHB and sug-gests that additional factors are required topromote disease expression.


Apoptosis-inflammation-fibrosis hypothesis

Maternal anti-La antibodies were identifiedon the surface of fetal myocardial fibers (45),and anti-Ro antibodies have been elutedfrom an affected fetal heart (46). Because theRo/La antigens are localized inside the celland maternal antibodies (IgG) cannot crossthe sarcolemma of a healthy cardiac myo -cyte, apoptosis was proposed as a mecha-nism by which the intracellular Ro/La anti-gens are translocated to the cell surface to beaccessible to maternal antibodies.

Translocation of intracellular

autoantigens to the cell membrane

Substantial experimental evidence has beenproposed to account for the translocation ofRo/La antigen to the cell surface, including vi-ral infection, UV light and IFNg treatment(47, 48). The limitation of the above studies isthe use of non-cardiac myocytes. Miranda etal. (49), using human fetal ventricular myo -cytes in culture from 16- to 24-week heartsdemonstrated surface expression of Ro/Laprotein in early and late apoptotic cells in-duced by staurosporine but not in non-apop-totic cardiac myocytes. It was suggested thatexperimental induction of apoptosis led tothe translocation of Ro/La antigens to cellsurface membrane to subsequently be acces-sible to binding by circulating anti-Ro/La anti-bodies, triggering the inflammation cascadeand eventually fibrosis. Perhaps the strongestevidence supporting the role of apoptosis inprovoking cardiac injury is the immunohisto-chemical analysis of hearts from several fetus-es dying with CHB (50, 51). Remarkably, notonly is apoptosis detectable, but exaggerated30-fold in septal tissue of CHB affectedhearts compared to age-matched controls.These histologic clues suggest either anunchecked exuberance of apoptosis or a po-tential defect in clearance.

Phygocytosis of cardiac myocytes

The above observations point to an exagger-ated apoptosis as the initial link between ma-ternal anti-Ro/La antibodies and tissue injury(52, 53). Under physiologic conditions,apoptosis is a distinct form of cell death inwhich the cell commits to a suicide program,leading to rapid elimination without inflam-mation (54, 55). Thus, the pathologic cas-cade from apoptosis to inflammation and fi-brosis initiated by autoantibodies in CHB hasbeen an active area of research. Clancy et al.(56) used phase-contrast and confocal mi-croscopy on co-cultured healthy and apop-totic human fetal ventricular myocytes todemonstrate that the healthy cardiac myo -cytes are capable of phagocytosing apoptot-ic cardiac myocytes. Experiments were thenperformed with apoptotic cardiac myo cytespreincubated with CHB-IgG. These treatedcells were co-cultured with healthy human fe-tal cardiac myocytes. Pretreatment of apop-totic cells with CHB-IgG inhibits the normalphagocytosis of the apoptotic cells. In con-trast, preincubation of apoptotic cells withcontrol IgG had no effect on engulfment.Given the role of the cardiac myocyte in thephysiologic removal of apoptotic cells, per-turbation of this developmental function bymaternal anti-Ro/La antibodies could ac-count for the exaggerated apoptosis ob-served in the autopsy sections from fetusesdying with CHB. Blocking physiologic apop-totic cell removal by inadvertent antibodybinding to apoptotic cells would be expectedto direct the pool of IgG-apoptotic cell com-plexes toward proinflammatory clearance byinfiltrating professional scavengers, themacrophages which may lead to fibrosis andpermanent heart block. Support for this mod-el are data demonstrating that supernatantsfrom macrophages co-cultured with op-sonized apoptotic human fetal cardiac my-ocytes transdifferentiate cardiac fibroblasts tomyofibroblasts, a scarring phenotype (57).Prolonged secretion of cytokines such asTGFβ may contribute to the exuberant scar-ring seen in CHB (58).


Mechanism of decreased clearance

of apoptotic cells

Given recent evidence implicating the uroki-nase plasminogen activator receptor (uPAR)as a “don’t eat me” signal during efferocyto-sis, experiments were performed to addresswhether surface bound anti-Ro antibody in-hibits apoptotic cell removal via an effect onthe expression/function of the urokinase-type plasminogen activator protease uPA/uP-AR system. As assessed by flow cytometryand confocal microscopy, uPAR colocalizesand interacts with Ro 60 on the surface ofapoptotic human fetal ventricular cardiacmyo cytes. Blocking of uPAR enhancesphagocytosis of apoptotic cardiac myocytesby healthy cardiac myocytes and reverses theanti-Ro 60 antibody dependent impairedclearance of apoptotic cardiac myocytes.Binding of anti-Ro 60 antibody to apoptoticcardiac myocytes results in increased uPARexpression, as well as enhanced uPA activity.It was suggested that increased uPAR expres-sion and uPA activity induced by anti-Ro an-tibody 60 binding to the apoptotic fetal car-diac myocyte provides a molecular basis bywhich these antibodies inhibit efferocytosis,resulting in exacerbated apoptosis and ulti-mately to scaring of the fetal conduction sys-tem and the working myocardium (59). Inter-estingly, only Ro 60, but not Ro 52 or La 48interacts with uPAR. This finding further illus-trates the complexity of the pathogenesis ofCHB as anti-Ro 52 antibody is known to be amajor player in the development of CHB (13,21, 42, 60-62).

Cascade linking inflammation to

fibrosis

Alvarez et al. (63) continued the efforts in in-vestigating downstream crosstalk betweeninflammatory and fibrosis pathways initiatedby binding anti-Ro/La antibodies to their cog-nitive antigens on the apoptotic cells. Incuba-tion of macrophages with immune complex-es (IC) comprised of Ro 60, hY3 ssRNA and

anti-Ro 60 antibody induces the Toll-like re-ceptor 7 (TLR7)-dependent generation of su-pernatants which provoked a fibrosing phe-notype in human fetal cardiac fibroblasts. Su-pernatants from macrophages incubatedwith IC induced the fibroblast secretion ofTGF, which was inhibited by an antagonist ofTLR7. Under the same conditions, the in-duced fibroblast secretion of TGFβ was de-creased by inhibitors of the endothelin (ET)-1receptors ETa or ETb, or by an anti-ET-1 anti-body, but not control antibody. ExogenousET-1 induced a profibrosing phenotype whilefibroblasts transfected with either ETa or ETbsiRNA were unresponsive to the profibrosingeffects of the IC-generated macrophage su-pernatants. Immunohistochemistry of thehearts from two fetuses dying with CHB re-vealed the presence of ET-1-producingmononuclear cells in the septal region in ar-eas of calcification and fibrosis. The authorsconcluded that these cells are a major sourceof TGFβ, and that ET-1 is one of the key com-ponents responsible for the profibrosing ef-fects generated by stimulated macrophages.A novel role of ET-1 in linking TLR7 inflamma-tory signaling to subsequent fibrosis was sug-gested.

In summary, the apoptosis hypothesisstipulates that in apoptotic human fetal ven-tricular myocytes, the intracellular autoanti-gens to maternal anti-Ro/La antibodiestranslocate to the cardiac myocyte sarcolem-ma where they are accessible to circulatinganti-Ro/La antibodies which might inadver-tently divert normal clearance of apoptoticcardiac myocytes by healthy cardiac myo -cytes thus increasing uPAR expression anduPA activity, toward clearance by profession-al macrophages (via FcγR) with the release ofthe inflammatory and/or fibrosing cytokines.ET-1 is one of the key components responsi-ble for the profibrosing effects generated bystimulated macrophages. Components of theproposed cascade include macrophages(representing the inflammatory component)and fibroblasts (representing the scarringcomponent).


Calcium channel hypothesis

As mentioned above, maternal antibodiescan also interact and recognize proteins oth-er than their own autoantigens. Anti-Roand/or anti-La antibodies were reported tocross-react with laminin B-1 chain, with hu-man cardiac myosin heavy chain (64), calreti-culin (41, 65), 5-HT4 serotoninergic receptor(62), and alpha-enolase (66) and Ca channels(39, 67, 68).

Calcium channel

The formulation of the Ca channel hypothe-sis is driven by the fact that AV node electro-genesis and action potential propagation tothe ventricle is under the control of L-type Cachannel. L-type Ca channel is a protein com-plex which consists of an α1 pore formingsubunit, β, and α2δ accessory subunits. The1 subunit is composed of four homologous(I-IV) domains containing six transmembranesegments. The Ca channel hypothesis pro-poses that circulating maternal antibodies di-rectly bind to specific epitopes of the Cachannel pore-forming subunit and inhibit Caentry to the cell. This inhibition of the Cachannel function by anti-Ro/La antibodies isper se sufficient to cause abnormal electro-cardiographic abnormalities similar to thoseseen in CHB.

Acute effect of anti-Ro/La antibodies

We have previously established that CHB-IgGstained the sarcolemma of non-premeabi-lized fetal cardiac myocytes using confocalimmunostaining, directly cross-reacts with L-type Ca channel α subunit by Western blot,functionally block the channel activity by thepatch-clamp techniques and cause conduc-tion abnormalities including sinus bradycar-dia and AVB by surface ECG (21, 22, 67-69).

At the whole heart level, perfusion of iso-lated beating hearts with CHB-IgG resulted inthe development of AVB as documented by

ECG recordings (21,22). Using optical map-ping technique which has the unique abilityto simultaneously record action potentialsfrom multiple sites, perfusion of CHB-IgG onisolated young rabbit hearts resulted in sinusbradycardia and various degrees of AVB in-cluding complete AV dissociation (24). Acti-vation maps revealed marked conduction de-lay at the sino-atrial junction but only minorchanges in overall atrial and ventricular acti-vation patterns. No conduction disturbanceswere seen in the presence of control IgGfrom mothers with healthy children (24). TheAVB and sinus bradycardia were similarlydemonstrated in Langendorff-perfused hu-man fetal (21) and animal hearts (9, 24, 43,44).

At the tissue level, using the standard dou-ble microelectrode technique, CHB-IgG butnot control IgG superfusion of dissectedhearts to specifically isolate and expose thesino-atrial and AV nodal areas resulted in ac-tion potential conduction block between theatria and AV node followed by complete inhi-bition of AV nodal action potential and con-sequently conduction to the ventricle (21).

At the single cell level, CHB-IgG, but notcontrol IgG, selectively inhibits L- and T-typeCa currents in freshly isolated cardiac my-ocytes from SA node, AV node, and the ven-tricle (21, 67-69). CHB-IgG had no effect onthe transient outward K current (Ito) (22), thedelayed rectifier K current (IKs) (22) and thefast Na current (INa) (22) indicating specificityfor Ca channels. Furthermore, CHB-IgG alsoinhibited 1C L-type Ca channel expressedsolely in Xenopus oocytes and in mammaliantsA201 cells to eliminate contamination byother ion channels present in the native car-diac cells (69). Altogether, the data demon-strate that acute exposure of cardiac my-ocytes to CHB-IgG functionally inhibits L-type Ca current resulting in conduction ab-normalities including sinus bradycardia, PRprolongation and complete AVB.


Anti-Ro/La antibodies and the newly

discovered α1D L-type Ca channel in

the heart

Because AV block has been the hallmarkphenotype for CHB, the AV node, ratherthan the sinoatrial node, has been the mainfocus of previous publications (2-4), and dur-ing clinical diagnosis of CHB (2-4, 6, 70). Si-nus bradycardia unrelated to AV block wasfirst reported in animal models of CHB (21-23, 39), which was later confirmed clinicallyin patients by Brucato et al. (11) and Menonet al. (10). These observations indicate thatthe spectrum of conduction abnormalities inCHB extends beyond the AV node to also af-fect the SA node. Up until recently, the ionicand molecular basis of this sinus bradycardiawas not known. Several currents have beenimplicated in the sinus node function (71-74).The major currents involved in the diastolicdepolarization and pacemaker activity in-clude the hyperpolarization-activated funnycurrent, If; the delayed rectifier, IK and the T-type Ca current. While there is a consensusthat the α1C L-type Ca current plays only aminimal role in the diastolic depolarization,the α1D isoform of L-type Ca channelemerged as a key player in the sinus nodepacemaker. The expression of α1D L-typeCa channel has always been thought to be re-stricted to the neuroendocrine tissue until thepublication of the knockout mouse model forcongenital deafness (75). Interestingly, thepups born to mice with this α1D L-type Cachannel knockout also exhibited a cardiacphenotype, sinus bradycardia and AV blockreminiscent of CHB (75). This finding trig-gered further research aimed at first investi-gating the potential presence and later thefunction of α1D L-type Ca channel in theheart (75-78). Indeed, studies performed inthe mouse revealed that α1D L-type Cachannels is expressed in the heart with (75-77) unique expression in the supraventriculartissue, especially the SA node (75-77) whereit was confirmed that α1D L-type Ca cur-rent plays a major role in phase 4 diastolicslope which controls sinus rate. Our group

was first to characterize the expression ofthis channel in the human fetal heart (67) andacross other species (67,79). We found thatthis channel has differential developmentalexpression profile, including a more abun-dant expression level, and universal localiza-tion in fetal and neonatal stages compared toadult heart (79). To test the potential role ofα1D L-type Ca channel in the sinus bradycar-dia observed in animal models of CHB, theeffects of CHB-IgG on this channel were in-vestigated. CHB-IgG, but not control IgG, didin fact inhibit α1D L-type Ca current (67).This inhibition may explain the sinus brady-cardia seen in CHB given that α1D L-type Cacurrent contributes to phase 4 diastolic depo-larization (10, 11, 21-23, 39).

Cross-reactivity and binding site of

anti-Ro/La antibodies on L-type Ca

channel

Because CHB-IgG functionally inhibits bothα1C and α1D L-type Ca channels, we soughtto determine whether CHB-IgG cross-reactand physically bind to the Ca channel. Usingimmunoprecipitation and Western blot, ourprevious experiments showed that CHB-IgGrecognizes and cross-reacts with both theα1C and α1D pore forming protein subunit(39, 67, 68). Next, we generated GST fusionproteins corresponding to the extracellularloop S5–S6 of each of the four domains thatform the pore of the Ca channel α1D subunitand tested the reactivity to CHB sera frommothers with CHB infants. A fraction (14.4%)of these sera reacted with the extracellularloop of S5–S6 of the first but not the second,third or fourth domains of the α1D subunit asdemonstrated by both ELISA and Westernblots (80).

The consequences of the inhibition ofboth α1C and α1D L-type Ca channels byCHB-IgG in the fetal ventricular myocytescould also account for the observation thataround 20% of children with CHB developlate onset cardiomyopathy despite appropri-ate pacing (14, 33, 34, 35). Because both


α1C and α1D L-type Ca channels are ex-pressed in the fetal heart ventricles and mayparticipate in the excitation-contraction cou-pling especially during fetal stages where themyocyte relies heavily on Ca entry from thesarcolemma. Inhibition of these channelsmay exert negative inotropic effect and/orelectro-mechanical uncoupling leading tocontractile dysfunction. To date, experimen-tal data supporting this proposition are stilllacking.

Consequence of chronic exposure of

Ca channels to anti-Ro/La antibodies

The above electrophysiological and biochem-ical findings clearly indicate that maternal an-ti-Ro/La antibodies interact directly with L-type Ca channels to acutely inhibit their func-tion. To investigate the consequence ofchronic exposure of pups to maternal autoan-tibodies as it occurs during pregnancy, weused our previously established murine mod-el for CHB (21, 68) where mice were immu-nized with recombinant Ro 52 protein andECG measured in pups at birth. The hypoth-esis tested was that chronic prolonged expo-sure of pups heart to maternal antibodiesduring pregnancy could lead to reduced Cachannels density. Patch-clamp techniqueswere used to record whole cell L-type Ca cur-rents in pups born to immunized mothersand controls. L-type Ca current density in car-diac myocytes from pups in Ro 52-immu-nized group was reduced by 38.6 ± 4.5% (P < 0.02, n = 17) (68). Similarly, in the rabbitCHB model immunized with Ro 52 protein(39), a 31 ± 3.4% (P = 0.02, n = 24) reductionof L-type Ca current density and a 19 ± 1.6%(P = 0.031 ) decrease in Ca channel proteinas well were observed in the pups born to im-munized mothers compared to controls.These observations indicate that duringchronic exposure in utero to maternal anti-bodies, the reduced L-type Ca current densi-ty could be attributed to a decrease in thenumber of functional Ca channels probablydue to internalization and degradation, set-

ting the stage for irreversible AVB at birthwhere proliferation of cardiac myocytes isminimal.

Possible reversal of electrocardio-

graphic abnormalities by

up-regulation of Ca channels

Now that reduced Ca channel density by ex-posure to maternal antibodies is demonstrat-ed, we hypothesized that if inhibition of theCa current is critical in cardiac conductiondisorders seen in CHB, then up-regulation ofthe Ca channels should rescue or reverse theelectrocardiographic abnormalities seen inCHB. Transgenic (TG) mice overexpressingthe α1C subunit of the L-type Ca channelwere used. The cardiac-specific over -expression of the α1C subunit of the L-typeCa channel and the detailed molecular, he-modynamic and electrophysiological charac-teristics of these TG mice are reported else-where (81). Briefly, the α1C transcript was in-creased by 2.8 fold. Similarly, the density ofL-type Ca current was increased by 44% to52% in cardiac myocytes from the TG mice.This percent increase is ideal because CHB-IgG inhibits Ca current within the samerange, 40–60% (21, 22, 67). Therefore, wepostulated that immunization of the TG miceoverexpressing α1C Ca channel proteinshould give birth to pups with no or fewerelectrocardiographic abnormalities. Indeed, alesser degree of sinus bradycardia and fewerAV conduction abnormalities were observedin TG pups overexpression α1C Ca channelfrom immunized mothers compared to non-TG pups (82). These findings demonstratethat up-regulation of L-type Ca channels canbe used as potential therapeutic approach toCHB by the development of cardiac specificCa channel agonists.

All together, the various available dataclearly and unambiguously indicate that anti-Ro/La antibodies from mothers with CHBchildren directly cross-react with both α1Cand α1D Ca channel protein on the cell sur-face membrane and inhibit the channel activ-


ity in both the SA and AV nodes which couldaccount for sinus bradycardia and AV block.The inhibition of L-type Ca channels in theventricles might account for the ventriculardysfunction seen in CHB. A proposed molec-ular basis of the Ca channel hypothesis is de-picted in the Figure: Two distinct conse-quences of Ca channel blockade by maternalantibodies can be identified: Acute (minutes,Step 1) and chronic (weeks-months, Steps 2-4) effects. In the acute phase (Step 1), circu-lating maternal anti-Ro/La antibodies bind toCa channels at the cell membrane, and inhib-it Ca current which per se is sufficient to

cause sinus bradycardia and AVB as demon-strated on the ECG of ex-vivo hearts perfusedwith CHB-IgG (21,22). Chronic exposure ofCa channels to circulating maternal antibod-ies, as it is the case during pregnancy, couldlead to a few scenarios: Binding of anti-Ro/Laantibodies to Ca channels, leads to cross-link-ing (Step 2) followed by internalization anddegradation of the channel protein (Step 3)as demonstrated by a decrease of Ca chan-nel current densities and proteins from thehearts of pup chronically exposed to mater-nal anti-Ro/La antibodies (68,69). Cross-link-ing of adjacent ion channels by the two Fab

Proposed Ca Channel Hypothesis for the Development of Autoimmune Associated Congeni-tal Heart Block.Acute (minutes, Step 1) and chronic (weeks to months, Steps 2-4) phases of CHB development. Inthe acute phase (Step 1), circulating maternal anti-Ro/La antibodies bind to Ca channels at the cellmembrane, and inhibit Ca current. In the chronic phase, long term exposure of Ca channels to cir-culating maternal antibodies during pregnancy, leads to binding of anti-Ro/La antibodies to Ca chan-nels, cross-linking (Step 2) followed by internalization and degradation of the channel protein (Step3). AV nodal cells with sufficient amount of the Ca channel reserve will survive (Step 4a). Cells withlesser Ca channel density reserves (step 4b) will eventually die leading to fibrosis and ultimately cal-cification. Alternatively, internalization and degradation of Ca channels affects the dynamic intracellu-lar Ca homeostasis which may lead to apoptosis (Step 4c), the translocation of intracellular Ro/La an-tigens to cell surface, the binding of the circulating anti-Ro/La antibodies and initiation of the inflam-mation, fibrosis and calcification.


arms of IgG increases the rate of normal in-ternalization of the target protein/antibodycomplex and thereby decreases the channeldensity on the cell surface as has been re-ported previously (83). Depending on the Cachannel density of the individual fetal cardiacmyocyte, AV nodal cells with sufficientamount of the remaining Ca channel reservewill survive (Step 4a) and continue to con-duct the impulse to the ventricle. Cells withlesser Ca channel density reserves (step 4b)will eventually die leading to fibrosis and ulti-mately calcification. Alternatively, internaliza-tion and degradation of Ca channels couldalso affect the dynamic intracellular Cahomeostasis which may lead to apoptosis(Step 4c), the translocation of intracellularRo/La antigens to cell surface, the binding ofthe circulating anti-Ro/La antibodies and initi-ation of the inflammation and fibrosis. Collec-tively, ample experimental evidence showthat anti-Ro/La antibodies and Ca channelsare causally related to the development ofCHB but the range and frequency of conduc-tion defects suggest that additional factorsmust be necessary to explain the full spec-trum of CHB.

References

1. Leach JL, Sedmak DD, Osborne JM, Rahill B,Lairmore MD, Anderson CL. Isolation from hu-man placenta of the IgG transporter, FcRn,and localization to the syncytio trophoblast:implications for maternal-fetal antibody trans-port. J Immunol 1996; 157 (8): 3317-3322

2. Buyon JP, Hiebert R, Copel J, Craft J, FriedmanD, Katholi M, Lee LA, Provost TT, Reichlin M,Rider L, Rupel A, Saleeb S, Weston WL,Skovron ML. Autoimmune-associated congen-ital heart block: demographics, mortality, mor-bidity and recurrence rates obtained from anational neonatal lupus registry. J Am Coll Car-diol 1998; 31 (7): 1658-1666

3. Waltuck J, Buyon JP. Autoantibody-associatedcongenital heart block: outcome in mothersand children. Ann Intern Med 1994; 120 (7):544-551

4. Silverman ED. Congenital heart block andneonatal lupus erythematosus: prevention isthe goal. J Rheumatol 1993; 20 (7): 1101-1104

5. Julkunen H, Kurki P, Kaaja R, Heikkila R, Immo-nen I, Chan EK, Wallgren E, Friman C. Isolatedcongenital heart block. Long-term outcome ofmothers and characterization of the immuneresponse to SS-A/Ro and to SS-B/La. ArthritisRheum. 1993; 36 (11): 1588-1598

6. Schmidt KG, Ulmer HE, Silverman NH, Klein-man CS, Copel JA. Perinatal outcome of fetalcomplete atrioventricular block: a multicenterexperience. J Am Coll Cardiol 1991; 17 (6):1360-1366

7. Buyon JP, Clancy RM, Friedman DM. Cardiacmanifestations of neonatal lupus erythemato-sus: guidelines to management, integratingclues from the bench and bedside. Nat ClinPract Rheumatol 2009; 5 (3): 139-148

8. Buyon JP, Clancy RM, Friedman DM. Autoim-mune associated congenital heart block: inte-gration of clinical and research clues in themanagement of the maternal/foetal dyad atrisk. J Intern Med 2009; 265 (6): 653-662

9. Hamilton RM, Lee-Poy M, Kruger K, SilvermanED. Investigative methods of congenital com-plete heart block. J Electrocardiol 1998; 30Suppl: 69-74

10. Menon A, Silverman ED, Gow RM, HamiltonRM. Chronotropic competence of the sinusnode in congenital complete heart block. AmJ Cardiol 1998; 82 (9): 1119-1121, A1119

11. Brucato A, Cimaz R, Catelli L, Meroni P. Anti-Ro-associated sinus bradycardia in newborns.Circulation 2000; 102 (11): E88-89

12. Motta M, Rodriguez-Perez C, Tincani A, Loja-cono A, Chirico G. Outcome of infants frommothers with anti-SSA/Ro antibodies. J Perina-tol 2007; 27 (5): 278-283

13. Cimaz R, Stramba-Badiale M, Brucato A, Catel-li L, Panzeri P, Meroni PL. QT interval prolon-gation in asymptomatic anti-SSA/Ro-positiveinfants without congenital heart block. ArthritisRheum 2000; 43 (5): 1049-1053

14. Moak JP, Barron KS, Hougen TJ, Wiles HB, Bal-aji S, Sreeram N, Cohen MH, Nordenberg A,Van Hare GF, Friedman RA, Perez M, CecchinF, Schneider DS, Nehgme RA, Buyon JP. Con-genital heart block: development of late-onsetcardiomyopathy, a previously underappreciat-ed sequela. J Am Coll Cardiol 2001; 37 (1):238-242


15. Askanase AD, Friedman DM, Copel J, DischeMR, Dubin A, Starc TJ, Katholi MC, Buyon JP.Spectrum and progression of conduction ab-normalities in infants born to mothers with an-ti-SSA/Ro-SSB/La antibodies. Lupus 2002; 11(3): 145-151

16. Saleeb S, Copel J, Friedman D, Buyon JP. Com-parison of treatment with fluorinated glucocor-ticoids to the natural history of autoantibody-associated congenital heart block: retrospec-tive review of the research registry for neona-tal lupus. Arthritis Rheum 1999; 42 (11): 2335-2345

17. Rein AJ, Mevorach D, Perles Z, Gavri S, Nad-jari M, Nir A, Elchalal U. Early diagnosis andtreatment of atrioventricular block in the fetusexposed to maternal anti-SSA/Ro-SSB/La anti-bodies: a prospective, observational, fetal kine-tocardiogram-based study. Circulation 2009;119 (14): 1867-1872

18. Vesel S, Mazi U, Blejec T, Podnar T. First-de-gree heart block in the fetus of an anti-SSA/Ro-positive mother: reversal after a short courseof dexamethasone treatment. Arthritis Rheum2004; 50 (7): 2223-6

19. Friedman DM, Kim MY, Copel JA, Davis C,Phoon CK, Glickstein JS, Buyon JP. Utility ofcardiac monitoring in fetuses at risk for con-genital heart block: the PR Interval and Dexa -methasone Evaluation (PRIDE) prospectivestudy. Circulation 2008; 117 (4): 485-493

20. Sonesson SE, Salomonsson S, Jacobsson LA,Bremme K, Wahren-Herlenius M. Signs of first-degree heart block occur in one-third of fetus-es of pregnant women with anti-SSA/Ro 52-kdantibodies. Arthritis Rheum 2004; 50 (4):1253-1261

21. Boutjdir M, Chen L, Zhang ZH, Tseng CE, Di-Donato F, Rashbaum W, Morris A, el-Sherif N,Buyon JP. Arrhythmogenicity of IgG and anti-52-kD SSA/Ro affinity-purified antibodies frommothers of children with congenital heartblock. Circ Res 1997; 80 (3): 354-362

22. Boutjdir M, Chen L, Zhang ZH, Tseng CE, El-Sherif N, Buyon JP. Serum and immunoglobu-lin G from the mother of a child with congeni-tal heart block induce conduction abnormali-ties and inhibit L-type calcium channels in a ratheart model. Pediatr Res 1998; 44 (1): 11-19

23. Mazel JA, El-Sherif N, Buyon J, Boutjdir M. Elec-trocardiographic abnormalities in a murinemodel injected with IgG from mothers of chil-dren with congenital heart block. Circulation1999; 99 (14): 1914-1918

24. Restivo M, Kozhevnikov DO, Boutjdir M. Opti-cal mapping of activation patterns in an animalmodel of congenital heart block. Am J PhysiolHeart Circ Physiol 2001; 280 (4): H1889-1895

25. Costedoat-Chalumeau N, Amoura Z, Lu-poglazoff JM, Huong DL, Denjoy I, Vauthier D,Sebbouh D, Fain O, Georgin-Lavialle S,Ghillani P, Musset L, Wechsler B, Duhaut P,Piette JC. Outcome of pregnancies in patientswith anti-SSA/Ro antibodies: a study of 165pregnancies, with special focus on electrocar-diographic variations in the children and com-parison with a control group. Arthritis Rheum2004; 50 (10): 3187-3194

26. Gordon PA, Khamashta MA, Hughes GR,Rosenthal E. Increase in the heart rate-correct-ed QT interval in children of anti-Ro-positivemothers, with a further increase in those withsiblings with congenital heart block: commenton the article by Cimaz et al. Arthritis Rheum2001; 44 (1): 242-243

27. Lazzerini PE, Acampa M, Guideri F, CapecchiPL, Campanella V, Morozzi G, Galeazzi M,Marcolongo R, Laghi-Pasini F. Prolongation ofthe corrected QT interval in adult patients withanti-Ro/SSA-positive connective tissue dis-eases. Arthritis Rheum 2004; 50 (4): 1248-1252

28. Lazzerini PE, Capecchi PL, Acampa M, Mo-rozzi G, Bellisai F, Bacarelli MR, Dragoni S,Fineschi I, Simpatico A, Galeazzi M, Laghi-Pasi-ni F. Anti-Ro/SSA-associated QTc interval pro-longation in the adults: The role of antibodylevel and specificity. Arthritis Care Res 2011.

29. Lazzerini PE, Capecchi PL, Laghi-Pasini F. The„ Invulnerability“ of the Adult Conduction Sys-tem to Anti-Ro/SSA Antibodies? A Matter ofCalcium Channel Expression on the Cardio -myocyte. J Cardiovasc Electrophysiol 2011; 22(8): E88

30. Schwartz PJ, Montemerlo M, Facchini M, Sal-ice P, Rosti D, Poggio G, Giorgetti R. The QTinterval throughout the first 6 months of life: aprospective study. Circulation 1982; 66 (3):496-501

31. Gerosa M, Cimaz R, Stramba-Badiale M, Gou-lene K, Meregalli E, Trespidi L, Acaia B, Catta-neo R, Tincani A, Motta M, Doria A, Zulian F,Milanesi O, Brucato A, Riboldi P, Meroni PL.Electrocardiographic abnormalities in infantsborn from mothers with autoimmune diseases– a multicentre prospective study. Rheumatol-ogy (Oxford) 2007; 46 (8): 1285-1289


32. Jaeggi ET, Hamilton RM, Silverman ED, Zamo-ra SA, Hornberger LK. Outcome of childrenwith fetal, neonatal or childhood diagnosis ofisolated congenital atrioventricular block. Asingle institution’s experience of 30 years. JAm Coll Cardiol 2002; 39 (1): 130-137

33. Nield LE, Silverman ED, Smallhorn JF, TaylorGP, Mullen JB, Benson LN, Hornberger LK. En-docardial fibroelastosis associated with mater-nal anti-Ro and anti-La antibodies in the ab-sence of atrioventricular block. J Am Coll Car-diol 2002; 40 (4): 796-802

34. Nield LE, Silverman ED, Taylor GP, SmallhornJF, Mullen JB, Silverman NH, Finley JP, Law YM,Human DG, Seaward PG, Hamilton RM, Horn-berger LK. Maternal anti-Ro and anti-La anti-body-associated endocardial fibroelastosis. Cir-culation 2002; 105 (7): 843-848

35. Eronen M, Siren MK, Ekblad H, Tikanoja T,Julkunen H, Paavilainen T. Short- and long-termoutcome of children with congenital completeheart block diagnosed in utero or as a new-born. Pediatrics 2000; 106 (1 Pt 1): 86-91

36. Guettrot-Imbert G, Cohen L, Fermont L, VillainE, Frances C, Thiebaugeorges O, Foliguet B,Leroux G, Cacoub P, Amoura Z, Piette JC,Costedoat-Chalumeau N. A new presentationof neonatal lupus: 5 cases of isolated mild en-docardial fibroelastosis associated with mater-nal Anti-SSA/Ro and Anti-SSB/La antibodies. JRheumatol 38 (2): 378-386

37. Raboisson MJ, Fouron JC, Sonesson SE, Ny-man M, Proulx F, Gamache S. Fetal Dopplerechocardiographic diagnosis and successfulsteroid therapy of Luciani-Wenckebach phe-nomenon and endocardial fibroelastosis relat-ed to maternal anti-Ro and anti-La antibodies. JAm Soc Echocardiogr 2005; 18 (4): 375-380

38. Pises N, Acherman RJ, Iriye BK, Rollins RC,Castillo W, Herceg E, Evans WN. Positive ma-ternal anti-SSA/SSB antibody-related fetal rightventricular endocardial fibroelastosis withoutatrioventricular block, reversal of endocardialfibroelastosis. Prenat Diagn 2009; 29 (2): 177-178

39. Xiao GQ, Qu Y, Hu K, Boutjdir M. Down-regu-lation of L-type calcium channel in pups bornto 52 kDa SSA/Ro immunized rabbits. FASEB J2001; 15 (9): 1539-1545

40. Miranda-Carus ME, Boutjdir M, Tseng CE, Di-Donato F, Chan EK, Buyon JP. Induction of an-tibodies reactive with SSA/Ro-SSB/La and de-velopment of congenital heart block in a

murine model. J Immunol 1998; 161 (11):5886-5892

41. Suzuki H, Silverman ED, Wu X, Borges C, ZhaoS, Isacovics B, Hamilton RM. Effect of maternalautoantibodies on fetal cardiac conduction: anexperimental murine model. Pediatr Res 2005;57 (4): 557-562

42. Salomonsson S, Sonesson SE, Ottosson L,Muhallab S, Olsson T, Sunnerhagen M,Kuchroo VK, Thoren P, Herlenius E, Wahren-Herlenius M. Ro/SSA autoantibodies directlybind cardiomyocytes, disturb calcium homeo -stasis, and mediate congenital heart block. JExp Med 2005; 201 (1): 11-17

43. Garcia S, Nascimento JH, Bonfa E, Levy R,Oliveira SF, Tavares AV, de Carvalho AC. Cellu-lar mechanism of the conduction abnormali-ties induced by serum from anti-Ro/SSA-posi-tive patients in rabbit hearts. J Clin Invest 1994;93 (2): 718-724

44. Viana VS, Garcia S, Nascimento JH, Elkon KB,Brot N, Campos de Carvalho AC, Bonfa E. In-duction of in vitro heart block is not restrictedto affinity purified anti-52 kDa Ro/SSA anti-body from mothers of children with neonatallupus. Lupus 1998; 7 (3): 141-147

45. Horsfall AC, Venables PJ, Taylor PV, Maini RN.Ro and La antigens and maternal anti-La idio-type on the surface of myocardial fibres incongenital heart block. J Autoimmun 1991; 4(1): 165-176

46. Reichlin M, Brucato A, Frank MB, MaddisonPJ, McCubbin VR, Wolfson-Reichlin M, Lee LA.Concentration of autoantibodies to native 60-kd Ro/SS-A and denatured 52-kd Ro/SS-A ineluates from the heart of a child who died withcongenital complete heart block. ArthritisRheum 1994; 37 (11): 1698-1703

47. Casciola-Rosen LA, Anhalt G, Rosen A. Au-toantigens targeted in systemic lupus erythe-matosus are clustered in two populations ofsurface structures on apoptotic keratinocytes. JExp Med 1994; 179 (4): 1317-1330

48. Baboonian C, Venables PJ, Booth J, WilliamsDG, Roffe LM, Maini RN. Virus infection in-duces redistribution and membrane localiza-tion of the nuclear antigen La (SS-B): a possiblemechanism for autoimmunity. Clin Exp Im-munol 1989; 78 (3): 454-459

49. Miranda ME, Tseng CE, Rashbaum W, OchsRL, Casiano CA, Di Donato F, Chan EK, BuyonJP. Accessibility of SSA/Ro and SSB/La anti-gens to maternal autoantibodies in apoptotic


human fetal cardiac myocytes. J Immunol1998; 161 (9): 5061-5069

50. Clancy RM, Kapur RP, Molad Y, Askanase AD,Buyon JP. Immunohistologic evidence sup-ports apoptosis, IgG deposition, and novelmacrophage/fibroblast crosstalk in the patho-logic cascade leading to congenital heartblock. Arthritis Rheum 2004; 50 (1): 173-182

51. Clancy RM, Backer CB, Yin X, Kapur RP, Mo-lad Y, Buyon JP. Cytokine polymorphisms andhistologic expression in autopsy studies: contri-bution of TNF-alpha and TGF-beta 1 to thepathogenesis of autoimmune-associated con-genital heart block. J Immunol 2003; 171 (6):3253-3261

52. Ucker DS. Death by suicide: one way to go inmammalian cellular development? New Biol1991; 3 (2): 103-109

53. Hale AJ, Smith CA, Sutherland LC, StonemanVE, Longthorne VL, Culhane AC, Williams GT.Apoptosis: molecular regulation of cell death.Eur J Biochem 1996; 236 (1): 1-26

54. Caricchio R, Reap EA, Cohen PL. Fas/Fas lig-and interactions are involved in ultraviolet-B-in-duced human lymphocyte apoptosis. J Im-munol 1998; 161 (1): 241-251

55. Fadok VA, Bratton DL, Konowal A, Freed PW,Westcott JY, Henson PM. Macro phages thathave ingested apoptotic cells in vitro inhibitproinflammatory cytokine production throughautocrine/paracrine mechanisms involvingTGF-beta, PGE2, and PAF. J Clin Invest 1998;101 (4): 890-898

56. Clancy RM, Neufing PJ, Zheng P, O’MahonyM, Nimmerjahn F, Gordon TP, Buyon JP. Im-paired clearance of apoptotic cardiocytes islinked to anti-SSA/Ro and -SSB/La antibodiesin the pathogenesis of congenital heart block.J Clin Invest 2006; 116 (9): 2413-2422

57. Clancy RM, Askanase AD, Kapur RP,Chiopelas E, Azar N, Miranda-Carus ME, Buy-on JP. Transdifferentiation of cardiac fibro -blasts, a fetal factor in anti-SSA/Ro-SSB/La anti-body-mediated congenital heart block. J Im-munol 2002; 169 (4): 2156-2163

58. Clancy RM, Buyon JP. Clearance of apoptoticcells: TGF-beta in the balance between inflam-mation and fibrosis. J Leukoc Biol 2003; 74 (6):959-960

59. Briassouli P, Komissarova EV, Clancy RM, Buy-on JP. Role of the urokinase plasminogen acti-vator receptor in mediating impaired efferocy-tosis of anti-SSA/Ro-bound apoptotic cardio-

cytes: Implications in the pathogenesis of con-genital heart block. Circ Res 107 (3): 374-387

60. Buyon JP, Slade SG, Reveille JD, Hamel JC,Chan EK. Autoantibody responses to the „ na-tive“ 52-kDa SS-A/Ro protein in neonatal lupussyndromes, systemic lupus erythematosus, andSjogren’s syndrome. J Immunol 1994; 152 (7):3675-3684

61. Clancy RM, Buyon JP, Ikeda K, Nozawa K, Ar-gyle DA, Friedman DM, Chan EK. Maternal an-tibody responses to the 52-kd SSA/RO p200peptide and the development of fetal conduc-tion defects. Arthritis Rheum 2005; 52 (10):3079-3086

62. Eftekhari P, Salle L, Lezoualc’h F, Mialet J,Gastineau M, Briand JP, Isenberg DA, FournieGJ, Argibay J, Fischmeister R, Muller S, Hoe-beke J. Anti-SSA/Ro52 autoantibodies block-ing the cardiac 5-HT4 serotoninergic receptorcould explain neonatal lupus congenital heartblock. Eur J Immunol 2000; 30 (10): 2782-2790

63. Alvarez D, Briassouli P, Clancy RM, Zavadil J,Reed JH, Abellar RG, Halushka M, Fox-TalbotK, Barrat FJ, Buyon JP. A novel role of endothe-lin-1 in linking TLR7-mediated inflammation tofibrosis in congenital heart block. J Biol Chem

64. Horsfall AC, Rose LM. Cross-reactive maternalautoantibodies and congenital heart block. JAutoimmun 1992; 5: 479-493

65. Orth T, Dorner T, Meyer Zum BuschenfeldeKH, Mayet WJ. Complete congenital heartblock is associated with increased autoanti-body titers against calreticulin. Eur J Clin Invest1996; 26 (3): 205-215

66. Llanos C, Chan EK, Li S, Abadal GX, Izmirly P,Byrne C, Clancy RM, Buyon JP. Antibody reac-tivity to alpha-enolase in mothers of childrenwith congenital heart block. J Rheumatol2009; 36 (3): 565-569

67. Qu Y, Baroudi G, Yue Y, Boutjdir M. Novel mo-lecular mechanism involving alpha1D (Cav1.3)L-type calcium channel in autoimmune-associ-ated sinus bradycardia. Circulation 2005; 111(23): 3034-3041

68. Qu Y, Xiao GQ, Chen L, Boutjdir M. Autoanti-bodies from mothers of children with congen-ital heart block downregulate cardiac L-typeCa channels. J Mol Cell Cardiol 2001; 33 (6):1153-1163

69. Xiao GQ, Hu K, Boutjdir M. Direct inhibition ofexpressed cardiac L- and T-type calcium chan-nels by igg from mothers whose children have


congenital heart block. Circulation 2001; 103(11): 1599-1604

70. Julkunen H, Kaaja R, Wallgren E, Teramo K. Iso-lated congenital heart block: fetal and infantoutcome and familial incidence of heart block.Obstet Gynecol 1993; 82 (1): 11-16

71. Guo J, Ono K, Noma A. A sustained inwardcurrent activated at the diastolic potentialrange in rabbit sino-atrial node cells. J Physiol1995; 483 ( Pt 1): 1-13

72. Noble D. The surprising heart: a review of re-cent progress in cardiac electrophysiology. JPhysiol 1984; 353: 1-50

73. Mitsuiye T, Shinagawa Y, Noma A. Sustainedinward current during pacemaker depolariza-tion in mammalian sinoatrial node cells. CircRes 2000; 87 (2): 88-91

74. Sanguinetti MC, Jurkiewicz NK. Two compo-nents of cardiac delayed rectifier K+ current.Differential sensitivity to block by class III an-tiarrhythmic agents. J Gen Physiol 1990; 96(1): 195-215

75. Platzer J, Engel J, Schrott-Fischer A, Stephan K,Bova S, Chen H, Zheng H, Striessnig J. Con-genital deafness and sinoatrial node dysfunc-tion in mice lacking class D L-type Ca2+ chan-nels. Cell 2000; 102 (1): 89-97

76. Zhang Z, Xu Y, Song H, Rodriguez J, Tuteja D,Namkung Y, Shin HS, Chiamvimonvat N. Func-tional Roles of Ca(v)1.3 (alpha(1D)) calciumchannel in sinoatrial nodes: insight gained us-ing gene-targeted null mutant mice. Circ Res2002; 90 (9): 981-987

77. Mangoni ME, Couette B, Bourinet E, Platzer J,Reimer D, Striessnig J, Nargeot J. Functionalrole of L-type Cav1.3 Ca2+ channels in cardiacpacemaker activity. Proc Natl Acad Sci U S A2003; 100 (9): 5543-5548

78. Mancarella S, Yue Y, Karnabi E, Qu Y, El-SherifN, Boutjdir M. Impaired Ca2+ homeostasis isassociated with atrial fibrillation in the al-pha1D L-type Ca2+ channel KO mouse. Am JPhysiol Heart Circ Physiol 2008; 295 (5):H2017-2024

79. Qu Y, Karnabi E, Ramadan O, Yue Y, ChahineM, Boutjdir M. Perinatal and postnatal expres-sion of Cav1.3 alpha1D Ca(2) channel in therat heart. Pediatr Res 69 (6): 479-484

80. Karnabi E, Qu Y, Wadgaonkar R, Mancarella S,Yue Y, Chahine M, Clancy RM, Buyon JP,Boutjdir M. Congenital heart block: identifica-tion of autoantibody binding site on the extra-cellular loop (domain I, S5-S6) of alpha(1D) L-type Ca channel. J Autoimmun 34 (2): 80-86

81. Muth JN, Yamaguchi H, Mikala G, Grupp IL,Lewis W, Cheng H, Song LS, Lakatta EG, Vara-di G, Schwartz A. Cardiac-specific overexpres-sion of the alpha(1) subunit of the L-type volt-age-dependent Ca(2+) channel in transgenicmice. Loss of isoproterenol-induced contrac-tion. J Biol Chem 1999; 274 (31): 21503-21506

82. Karnabi E, Qu Y, Mancarella S, Boutjdir M. Res-cue and Worsening of Congenital Heart Block-Associated Electrocardiographic Abnormalitiesin Two Transgenic Mice. J Cardiovasc Electro-physiol

83. Lennon VA, Kryzer TJ, Griesmann GE, O’Suil-leabhain PE, Windebank AJ, Woppmann A,Miljanich GP, Lambert EH. Calcium-channelantibodies in the Lambert-Eaton syndrome andother paraneoplastic syndromes. N Engl J Med1995; 332 (22): 1467-1474

Correspondence addressMohamed Boutjdir, PhD, FAHAResearch and Development (151)VA New York Harbor Healthcare System800 Poly PlaceBrooklyn, NY [email protected]

pathophysiology of autoimmune-associated congenital heart ... · applied cardiopulmonary...

Documents