Autoantibodies Associated with Congenital Heart Block
S. Salomonsson, Rheumatology Unit, Department of Medicine, Karolinska Institutet, 171 76 Stockholm, Sweden. E-mail: email@example.com
Congenital heart block is the most severe manifestation of neonatal lupus syndrome. It is a passively acquired disease where transplacental passage of maternal autoantibodies is associated with irreversible damage of the foetal cardiac conduction system. It is well established that the condition, in the absence of structural abnormalities, is strongly associated with maternal autoantibodies to the Ro/La antigens. More specifically the disease has been closely linked to antibodies to the Ro52 component of the antigen complex. Congenital heart block constitutes a unique model where specific autoantibodies target and mediate organ-specific disease. A wide panel of maternal antibodies has been discussed in literature in association with the disease and are described in this review.
Neonatal lupus erythematosus is a passively acquired autoimmune condition closely associated with maternal autoantibodies (reviewed in ). The syndrome includes several clinical manifestations, congenital heart block and cutaneous lupus being the most common, while complications such as hepatitis and cytopenias may occur . The non-cardiac manifestations of neonatal lupus are transient and resolve as maternal antibodies are cleared from the neonatal circulation , while a complete atrioventricular block in a structurally normal heart, is considered permanent.
Congenital heart block develops during gestational week 18–25 and presents with a low ventricular rate, usually ranging from 40 to 60 beats per minute . A complete heart block is a potentially lethal condition and morbidity in surviving foetuses is substantial, with more than two-thirds of affected children requiring permanent pacemaker implantation [2, 4, 5].
Congenital heart block is thought to result from an inflammation of the foetal heart tissue mainly affecting the atrioventricular node, as documented in several histological studies of heart tissue from foetuses dying from the condition. Histological studies in affected diseased foetuses have confirmed signs of inflammation in the foetal heart including lymphocytic infiltrates, deposition of antibodies and complement components, as well as calcification and fibrosis [6–9].
Maternal Ro/La autoantibodies
Development of congenital heart block is closely related to the presence of maternal autoantibodies associated with the rheumatic diseases Sjögren’s syndrome or SLE, but the mother of an affected child may also be asymptomatic. A relationship between maternal antibodies against the Ro/La antigen and congenital heart block is well established, and the antigen contains several components to which rheumatic patients develop autoantibodies. Cumulative data is somewhat heterogeneous and the linkage between disease and the specific antigen components Ro52, Ro60 and La proteins varies. However, a majority of the attempts to screen for a specific maternal antibody profile have demonstrated an almost universal presence of antibodies targeting the Ro52 protein [10–20]. Interestingly, the prevalence of having a child with congenital heart block is 2% in women with anti-Ro antibodies [17, 21] and 10–20% in mothers with a previously affected infant [2, 4, 22, 23] clearly indicating involvement of other factors besides anti-Ro52 antibodies in establishment of the disease.
Antibodies to Ro60 and La have been suggested to have a minor role in predicting the foetal clinical outcome in anti-Ro and anti-La antibody–positive mothers [14, 16, 24], although an association also between these autoantibodies and the incidences of congenital heart block has been demonstrated [14, 25]. The level of antibodies to the La protein has been found to be higher in mothers of children developing cutaneous lupus rather than heart block . In summary, although congenital heart block may develop independently of maternal antibodies against Ro60 and La these autoantibodies might, if present, be able to amplify the immunological response after onset in affected foetuses .
In addition, antibodies against an alternatively spliced transcript of Ro52, Ro52β was implicated in congenital heart block after finding higher levels of Ro52β mRNA compared to full-length Ro52 mRNA in foetal heart during the susceptible gestational weeks . However, Ro52β protein expression has not been demonstrated in animals or humans, although in vitro-translated 52β was shown to be antigenic using sera from Ro52-positive patients and from healthy donors .
Anti-Ro52 antibodies associated with congenital heart block
A specific maternal antibody profile correlating with congenital heart block would enable identification of mothers at high risk for complications with the condition and might help to determine the pathogenic mechanism that induces this autoimmune condition. Anti-Ro52 antibodies are highly associated with congenital heart block and systematic analyses to identify a subpopulation and specificity of the maternal Ro52 antibodies that cause disease have been undertaken. Attempts to define a specific antibody profile demonstrate a major antigenic region present in the central part of Ro52 [16, 29–33]. An extensive epitope mapping using overlapping synthetic peptides covering this immunodominant region revealed specific antibodies against amino acid sequence 200–239 (p200) of the Ro52 protein, to be associated with a higher risk of developing congenital heart block [16, 18, 20]. The denoted immunodominant region encompasses a functional domain, a leucine-zipper structure. Association with autoantibodies specific for a functional domain is not a unique feature for congenital heart block. In the case of heart block binding of anti-p200 autoantibodies might possibly mediate inhibition or decrease the functional activity of the antigen, as is evident for autoantibodies against RNA polymerase I inhibiting RNA synthesis, or against DNA topoisomerase I causing relaxation of super-coiled DNA (reviewed in ).
Proposed mechanism of cross-reactivity
Another possible scenario, besides interaction with Ro52, is that the maternal anti-Ro52 autoantibodies cross-react with another protein expressed in foetal cardiac tissue. There are several proteins that have been suggested as cross-reactive targets of Ro52 antibodies including the 5-HT4 serotoninergic receptor , the α1C and the α1D subunits of the L-type calcium channel , as well as the T-type calcium channel .
Eftekhari and colleagues  demonstrated that antibodies reactive with the second extracellular loop of the 5-HT4 serotoninergic receptor, cloned from human adult atrium, can bind to Ro52 and that sera from mothers with affected children recognize the 5-HT4 receptor. However, others have not been able to confirm the 5-HT receptor as a target of the immune response in mothers with affected children .
Several publications have shown arrythmogenic effects of anti-Ro52 antibodies and evidence is emerging to support a direct effect of the antibodies on cardiocyte function, possibly because of cross-reactivity. This hypothesis has been supported by the demonstration that human affinity purified anti-Ro52-positive sera induce AV block in whole young rabbit hearts , and human foetal hearts  and inhibit inward calcium fluxes across cell membranes [39, 40]. More specifically, maternal antibodies have been proposed to interact with the pore-forming α1C subunit of calcium channels, possibly leading to internalization with subsequent cell death and exposure of intracellular Ro and La proteins, ultimately resulting in an inflammatory reaction . Ro/La-positive IgG has been demonstrated to inhibit currents through both subunits of the L-type calcium channel as well as the T-type calcium channel [36, 41, 42].
The Ca channel α1D subunit has been shown to be expressed in human foetal hearts . In a recent study, it has been demonstrated that a fraction of sera from mothers of children with congenital heart block react to the extracellular loop of the calcium channel α1D subunit and that these maternal antibodies can inhibit α1D calcium currents in vitro .
The potential role of the specific anti-Ro52 antibodies targeting p200 in the mechanism underlying congenital heart block remains to be embellished; however, experimental findings suggest that anti-p200 antibodies may interact with cardiomyocytes and disturb calcium homeostasis  supporting a mechanism involving a direct interaction with the calcium ion channel complex.
In addition to antibodies directed to the Ro and La proteins, several other targets have been suggested to be associated with development of congenital heart block. Antibodies against neonatal heart muscarinic acetylcholine receptor (mAChR) have been implicated to have a role in the disease supported by binding, and biological effects in in vitro studies [44, 45]. Calreticulin, a calcium-binding, multifunctional protein, involved in calcium storage has been suggested a target in the disease initiation, and antibodies to calreticulin have been demonstrated in sera from mothers with affected children . Another suggested antibody specificity associated with congenital heart block is antibodies to p57, which was identified in a child with the disease . Antibodies to a cleavage product of α-fodrin has, in addition to being identified as an organ-specific antibody in Sjögren’s syndrome, been suggested as an additional serologic marker in congenital heart block . Lately, Llanos and colleagues  investigated the role of antibodies to α-enolase associated with the condition, but a relationship could not be confirmed.
Congenital heart block is considered a passively acquired disease where maternal antibodies against Ro/La antigens are potentially affecting the developing foetal heart resulting in a complete atrioventricular block. No antibody specificity has been closer associated with congenital heart block than anti-Ro52 antibodies, which are detectable in the vast majority of autoantibody-positive mothers of affected children. However, considering clinical observations of only 10–20% reoccurrence rate in Ro/La-positive mothers with a previously affected infant, this indicates that maternal autoantibodies are necessary but not sufficient for induction of disease. Clinical factors such as maternal disease and infection [50–52] as well as antibody levels [13, 19, 51] and subclasses [19, 53] have been studied without reaching a common conclusion.
A two-stage model for the development of congenital heart block has been suggested, including transferred anti-Ro52 antibodies as initiators of cell death and the cardiac insult. This reaction may in later phases be exaggerated by other autoantibodies targeting intracellular autoantigens now exposed on the cell surface resulting in permanent damage in genetically susceptible foetuses . Remaining questions are whether the binding of maternal autoantibodies is direct or indirect, as well as an explanation to why the foetal heart is selectively vulnerable compared to the adult maternal heart, and why congenital heart block only affects a small portion of foetuses in Ro52 seropositive women.
Financial support for this work was obtained from KIRCNET (Karolinska Institutet Circulation and Respiratory Research Network), the Magn Bergvalls Foundation, the Jerring Foundation, Stiftelsen Samariten, the Karolinska Institute and The Royal Swedish Academy of Sciences.