FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |   Search Term  Advanced Search

Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe

Previous Volume 354:2677-2688 June 22, 2006 Number 25 Next

Abstract PDF PDA Full Text PowerPoint Slide Set Supplementary Material Editorial by Saffitz, J. E. Add to Personal Archive Add to Citation Manager Notify a Friend E-mail When Cited E-mail When Letters Appear PubMed Citation

ABSTRACT

Background Atrial fibrillation is the most common type of cardiac arrhythmia and a leading cause of cardiovascular morbidity, particularly stroke. The cardiac gap-junction protein connexin 40 is expressed selectively in atrial myocytes and mediates the coordinated electrical activation of the atria. We hypothesized that idiopathic atrial fibrillation has a genetic basis and that tissue-specific mutations in GJA5, the gene encoding connexin 40, may predispose the atria to fibrillation.

Methods We sequenced GJA5 from genomic DNA isolated from resected cardiac tissue and peripheral lymphocytes from 15 patients with idiopathic atrial fibrillation. Identified GJA5 mutations were transfected into a gap-junction–deficient cell line to assess their functional effects on protein transport and intercellular electrical coupling.

Results Four novel heterozygous missense mutations were identified in 4 of the 15 patients. In three patients, the mutations were found in the cardiac-tissue specimens but not in the lymphocytes, indicating a somatic source of the genetic defects. In the fourth patient, the sequence variant was detected in both cardiac tissue and lymphocytes, suggesting a germ-line origin. Analysis of the expression of mutant proteins revealed impaired intracellular transport or reduced intercellular electrical coupling.

Conclusions Mutations in GJA5 may predispose patients to idiopathic atrial fibrillation by impairing gap-junction assembly or electrical coupling. Our data suggest that common diseases traditionally considered to be idiopathic may have a genetic basis, with mutations confined to the diseased tissue.

Studies investigating a molecular pathogenesis of atrial fibrillation have focused on familial forms of the disease. Recently, a mutation in KCNQ1, a potassium-channel gene also implicated in a form of the long-QT syndrome, was identified as the molecular basis of autosomal dominant atrial fibrillation in a single family from China.⁵ Two additional families with autosomal dominant atrial fibrillation have been identified and the loci mapped to 10q22–q24 and 6p14–p16.^6,7 However, a causal gene has not been identified. Although the study of rare, familial forms of atrial fibrillation may provide insight into the molecular pathways involved in these selective cases of the disease, the genetic defects identified may not be representative of the pathogenesis in the more common, nonfamilial form.

Idiopathic forms of disease are traditionally not considered genetic in origin. However, somatic mutations are increasingly being identified as the cause of sporadic cases of disease in humans.^8,9 Though somatic mutations are mostly recognized as a factor in the genetics of cancer, recent studies have confirmed the presence of somatic mutations that give rise to nonmalignant disease.^10,11 We hypothesized that idiopathic atrial fibrillation has a genetic basis and that mutations predisposing the atria to fibrillation are somatic in nature. We focused on GJA5, the gene encoding connexin 40, a gap-junction protein with gene expression restricted principally to atrial tissue in humans. Connexin 40 gap junctions play a critical role in mediating atrial conduction through electrical coupling between cells, and GJA5-knockout mice have increased vulnerability to atrial reentrant arrhythmias.^12,13 Furthermore, recent data suggest that sequence variations in potential regulatory regions of GJA5 may increase the risk of atrial fibrillation.¹⁴

Methods

Study Subjects

We obtained archival tissue specimens from 15 patients with idiopathic atrial fibrillation who had undergone surgical isolation of the pulmonary veins between 1998 and 2002 at the London Health Science Center, London, Ontario. These patients had an early age at onset of atrial fibrillation (average, 45 years), atrial fibrillation that was refractory to multiple medications, and no evidence of coronary artery, valvular, or hypertensive heart disease. All 15 patients provided blood samples for further DNA analysis. Lymphocyte DNA from 120 healthy persons and cardiac tissue obtained at the time of cardiac surgery from 50 patients without a history of atrial fibrillation were analyzed as controls. All study subjects were of Western European descent. All samples used in this study were collected after written informed consent had been obtained from the study subjects and were approved for study by the institutional review boards of the University of Western Ontario and the University of Ottawa Heart Institute.

Detection and Confirmation of Mutations

Genomic DNA was isolated from cardiac-tissue specimens according to the phenol–chloroform method. The coding region of GJA5 was amplified with the use of the polymerase chain reaction (PCR). Mutation screening was performed with the use of direct sequencing. To confirm somatic mutations, genomic DNA and total RNA from the cardiac-tissue specimens were extracted independently according to a silica-gel–column method (DNAeasy Tissue Kit and RNAeasy Micro Kit, Qiagen). An additional step of probe sonication was performed to enhance RNA extraction.

For the genomic DNA, PCR products were subcloned into the pGEM-T vector (Promega) and mutations were confirmed by restriction-digest analysis, direct sequencing, or both. For RNA, after reverse-transcriptase PCR was performed, mutations were confirmed by direct sequencing of the complementary DNA (cDNA). All sequencing was performed on an ABI 3100 DNA sequencer.

Protein-Transport Experiments

GJA5 mutations were introduced into a wild-type GJA5 clone by site-directed mutagenesis with the use of a mutagenesis kit (QuickChange, Stratagene). Clones were sequenced to confirm the mutation and exclude any other sequence variants. For localization experiments, wild-type and mutant GJA5 were tagged with green fluorescent protein (GFP). GFP-tagged wild-type or mutant GJA5 cDNA was transfected into neuroblastoma (N2A) cells, a gap-junction–deficient cell line. The connexin 40 expressed in the transfected N2A cells was immunolabeled with a 1:100 dilution of anti–connexin 40 antibody (Chemicon) followed by anti-IgG conjugated with Texas Red (Jackson ImmunoResearch Laboratories). Cells were viewed by confocal microscopy.

Immunohistochemical Analysis

Immunohistochemical analysis was performed on sections (7 µm thick) of atrial cardiac tissue fixed in formalin and embedded in paraffin. Tissue sections were deparaffinized, and antigens were retrieved by placing them in a microwave oven for 12.5 minutes in a citrate buffer (pH 5.6). The specimens were then blocked in 1:20 normal horse serum and incubated with 1:100 anti–connexin 40 antibody. The primary antibody was then detected with the diaminobenzidine chromagen system (Dako).

Electrophysiological Experiments

To assess protein function, we selected pairs of N2A cells with visible plaques at the cell–cell junction. Dual whole-cell voltage-clamp recordings were carried out at room temperature. To simulate the effect of heterozygosity and to evaluate the effect of the mutant connexin 40 protein on other cardiac connexins, we injected oocytes of the genus xenopus with complementary RNA (cRNA) for wild-type connexin 40, wild-type connexin 43, mutant connexin 40, or a mixture of wild-type and mutant connexin 40 cRNA in a 1:1 ratio, as previously described.¹⁵

Statistical Analysis

We used Fisher's exact test to analyze differences in the frequencies of coding-sequence variations between samples from patients with atrial fibrillation and samples from controls.

Results

Mutations

We identified four novel heterozygous mutations in GJA5 in 4 of the 15 patients with idiopathic atrial fibrillation (Table 1). The mutations were in the highly conserved transmembrane-spanning domains of the connexin 40 protein (Figure 1A). A 262CT mutation, predicting the substitution of serine for proline at codon 88 (Pro88Ser), was identified in two patients (Figure 1B). An additional patient had two nonallelic mutations, as determined by subcloning, resulting in Met163Val (487AG) and Gly38Asp (113GA) substitutions (Figure 1C and 1D, respectively). These mutations were detected in genomic DNA and RNA isolated from cardiac-tissue specimens. However, sequenced DNA from lymphocytes from these three patients showed no evidence of mutation, indicating a somatic source of the genetic defects (Table 2). The Pro88 residue, located in the second transmembrane domain of connexin 40, is conserved evolutionarily in mammalian and zebrafish connexins (Figure 1F). An identical Pro88Ser substitution in connexin 32 and connexin 50, resulting in dominantly inherited Charcot–Marie–Tooth disease and cataract disease, respectively, has been reported in five unrelated families.^{16,17,18,19,20} Similarly, Met163 substitutions in connexin 26 have been associated with hereditary deafness.^21,22

View this table: [in this window] [in a new window]   Table 1. Clinical Characteristics of Patients with Idiopathic Atrial Fibrillation and Status of Connexin 40 Mutation.

 

View larger version (53K): [in this window] [in a new window]   Figure 1. Detected Substitutions in Connexin 40 in Patients with Idiopathic Atrial Fibrillation. Panel A illustrates the locations of the detected substitutions in connexin 40, all predicted to occur within the transmembrane-spanning regions. Gly38 (represented with a G) is located within the first transmembrane domain (TM1), Pro88 (P) and Ala96 (A) within TM2, and Met163 (M) within TM3. The results of sequence analysis (Panel B, left-hand side) show the detection of the Pro88Ser substitution in cardiac tissue but not lymphocyte tissue, indicating a somatic source of the mutation in GJA5. Subcloning of the left atrial alleles and restriction-enzyme digestion with HypCH4 IV liberated a predicted 331-bp fragment created by the mutation (Panel B, center and right-hand side). Panels C and D show the results of sequencing and allele subcloning for the somatic substitutions Met163Val and Gly38Asp, respectively, which were also present in cardiac tissue but not lymphocyte tissue. Panel E shows the Ala96Ser sequence variation, which was the only variant detected in lymphocyte tissue from a patient. A high degree of cross-species sequence homology of the identified mutations is present at the level of both DNA and protein (Panel F). The highly conserved transmembrane regions are shaded.

 

View this table: [in this window] [in a new window]   Table 2. Connexin 40 Clones from Patients with Somatic Mutations.

 
Cardiac-tissue specimens from a fourth patient had a 286GT mutation, predicting an Ala96Ser substitution (Figure 1E). This sequence variant was also identified in lymphocyte DNA, indicating a germ-line origin. The GJA5 genes from the patient's immediate family members were sequenced. The variant was absent in the patient's three siblings and wife but was present in his two sons, who did not have a history of atrial fibrillation. However, on surface electrocardiography, the carrier sons had an abnormally prolonged P-wave duration (>120 msec), a known predictor of atrial fibrillation²³ (Figure 1 in the Supplementary Appendix, available with the full text of this article at www.nejm.org). The variant was also identified in the lymphocyte DNA from one 48-year-old control subject (population frequency, 0.6 percent). Long-term follow-up of asymptomatic persons carrying the allele encoding Ser96 will be needed to confirm its clinical significance.

We also sequenced GJA1 (which encodes connexin 43) from cardiac-tissue specimens. No coding-region mutations were identified. To assess the prevalence of GJA5 mutations in randomly selected patients with nonvalvular atrial fibrillation, we sequenced DNA isolated from lymphocytes from 20 such patients. No coding-region mutations were identified, suggesting either that the prevalence of GJA5 defects is lower in a more random population of patients with atrial fibrillation than in a more specific subgroup of patients with idiopathic atrial fibrillation or that tissue-specific genetic analyses are needed to measure the prevalence more accurately.

A cross-species comparison of connexin 40 shows a homology of more than 80 percent between the DNA and amino acid sequences, reflecting the evolutionary conservation of the structures of the GJA5 gene and the connexin 40 protein (Figure 1F). The direct sequencing of 100 alleles from cardiac-tissue DNA and 240 alleles from lymphocyte DNA from controls showed no evidence of the Pro88Ser, Met163Val, or Gly38Asp substitution, providing further support for a potentially pathogenetic role of these substitutions. Two other sequence variations were identified in lymphocyte DNA from the controls: 369CT, which causes a Tyr123 silent mutation, and 377CT, which causes a Pro126Leu substitution. Pro126 is deleted in the connexin 40 orthologue in rats and occurs within the cytoplasmic loop of connexin 40, a region of known variability among connexin proteins.¹² The frequency of sequence variations in GJA5 was significantly higher in patients than in controls (P<0.001).

Transport of Connexin 40 Mutants

Wild-type and mutant connexin 40 were expressed in N2A cells, a gap-junction–deficient cell line. As expected, cells expressing wild-type connexin 40 were localized subcellularly to sites of cell–cell contact (Figure 2A). Cells expressing the Pro88Ser mutant formed no visible gap-junction plaques and had intracellular retention of connexin 40 (Figure 2C). Similarly, a greater degree of intracellular localization and only sparse gap-junction plaques were evident in cells expressing the Gly38Asp mutant protein (Figure 2I), as compared with the wild type. Expression of the Ala96Ser variant was also localized intracellularly to a greater degree than that of wild-type connexin 40, but gap-junction plaques were present in similar amounts (Figure 2E). In contrast, the Met163Val mutant protein exhibited a pattern of subcellular distribution (Figure 2G) similar to that of wild-type connexin 40.

View larger version (27K): [in this window] [in a new window]   Figure 2. Differential Localization of Connexin 40 Mutants in N2A Cells. N2A cells were transfected with cDNA — encoding either wild-type connexin 40 or the connexin 40 mutant Pro88Ser, Ala96Ser, Met163Val, or Gly38Asp, tagged with green fluorescent protein (GFP) at the carboxyl terminus (left-hand images). In cells expressing nontagged connexin 40 (Panel A, inset; anti–connexin 40 antibody staining followed by anti-IgG conjugated with Texas Red), GFP-tagged connexin 40 (Panel A), or the mutants Ala96Ser (Panel E) or Met163Val (Panel G), gap-junction–like structures were readily seen at cell–cell interfaces (arrows). However, cells expressing GFP-tagged (Panel C) or nontagged (Panel C, inset) Pro88Ser mutants failed to assemble gap junctions. The expression of the Gly38Asp mutant (Panel I) was primarily within intracellular compartments, with occasional formation of a gap-junction–like structure. Transmitted-light images of the same cells (right-hand images) reveal that all cells analyzed had contact with neighboring cells.

 
Immunohistochemical Staining of Atrial Tissue

Atrial-tissue specimens from controls and patients with atrial fibrillation who carried wild-type or mutant connexin 40 were stained with anti–connexin 40 antibody. Immunostaining showed that connexin 40 localization and gap-junction formation were indistinguishable in the atrial tissue of patients without GJA5 mutations (Figure 3B) and in controls (Figure 3A). In contrast, atrial tissue from the two patients with the Pro88Ser substitution consistently showed a mosaic pattern of abnormal gap-junction formation and intracellular retention of connexin 40 (Figure 3C). A similar pattern, although weaker, was evident in the atrial tissue of Patient 9, who had the nonallelic Gly38Asp and Met163Val substitutions (Figure 3D). The abnormal localization of connexin 40 in cardiac cells confirms an origin of the mutations in myocytes, as opposed to fibroblasts. Atrial tissue harboring the Ala96Ser variant did not show any obvious abnormalities in connexin 40 distribution (Figure 3E).

View larger version (27K): [in this window] [in a new window]   Figure 3. Localization of Connexin 40 and Formation of Gap Junctions in Atrial Tissue from Patients with Detected Mutations. Panel A shows normal gap-junction formation at intercalated disks of adjacent cardiac myocytes in atrial tissue from a control subject. Panel B shows an atrial-tissue specimen from Patient 12, who had atrial fibrillation but no GJA5 mutation. The localization of connexin 40 and gap-junction formation is indistinguishable from that in the control tissue. In contrast, atrial tissue from Patient 11, who had the Pro88Ser substitution, showed a mosaic pattern of staining, with frequent patchy areas of intracellularly retained connexin 40 and abnormal gap-junction formation (Panel C). Similarly, but to a lesser extent, intracellular retention of connexin 40 and abnormal gap-junction formation were observed in atrial tissue from Patient 9, who had the nonallelic Gly38Asp and Met163Val substitutions (Panel D). Atrial tissue from Patient 6 (Panel E), who had the Ala96Ser substitution, showed no evidence of abnormal connexin 40 localization or gap-junction formation. Anti–connexin 40 antibody was detected with the use of the diaminobenzidine chromagen system.

 
Electrophysiological Function of Mutant Connexin 40

The electrical-coupling properties of wild-type and mutant connexin 40 were recorded for the paired N2A cells that exhibited visible gap-junction plaques, when possible. No visible plaques were observed in cells expressing the GFP-tagged Pro88Ser mutant, and as expected, there was no cell–cell coupling between cells (Figure 4A and Table 3). In cell pairs expressing the Gly38Asp mutant, cell–cell coupling occurred at a significantly lower junctional conductance than in wild-type connexin 40 cells (Table 3). However, in two cell pairs that had gap-junction plaques, the coupling conductances were similar to those of wild-type connexin 40 (data not shown), suggesting that the functional consequence of this mutation is related predominantly to impaired transport of the protein to the cell surface and reduced gap-junction formation, rather than to impaired coupling properties. In contrast, paired cells that formed visible gap-junction plaques and expressed the Ala96Ser variant consistently demonstrated no or weak cell–cell coupling, as compared with wild-type connexin 40 (Table 3). The finding suggests that this variant significantly impairs functional cell–cell coupling. Paired cells expressing the Met163Val mutant had cell-coupling properties similar to those of wild-type connexin 40 (Table 3), suggesting that this mutation may represent a benign polymorphism.

View larger version (30K): [in this window] [in a new window]   Figure 4. Effect of Pro88Ser and Ala96Ser Connexin 40 Mutations on Cell Coupling and the Activity of Wild-Type Connexins. Panel A illustrates the voltage protocol (–100 to +100 mV) used to assess coupling and voltage-dependent gating of pairs of N2A cells transfected with connexin 40 cDNA. The traces at the bottom show the superimposed current recordings from one cell of the pair in response to the voltage protocol applied to the other cell. Coupling and voltage-dependent gating were observed in cells transfected with wild-type connexin 40 (left-hand side) but not in cells transfected with the Pro88Ser mutant (right-hand side). Panel B compares junctional conductances of pairs of xenopus oocytes injected with xenopus connexin 38 antisense oligonucleotide alone (AS), wild-type connexin 40 (wtCx40), or mutant connexin 40 or coinjected with equal amounts of wild-type and mutant cRNA. Both the Pro88Ser substitution (P88S) (left-hand side) and the Ala96Ser variant (A96S) (center) showed absent or minimal functional cell–cell coupling, similar to that in N2A cells transfected with the mutant protein, and significantly impaired the activity of wild-type connexin 40 when the two were coinjected. The data are the means (±SE). The number of cell pairs studied are indicated in parentheses; the amount of wild-type connexin 40 cRNA injected into each oocyte was held constant to facilitate comparison of the data for the homotypic and heteromeric oocyte pairs in the left-hand and center plots. The right-hand plot shows that connexin 40 mutants also inhibit the activity of connexin 43 (Cx43). When Pro88Ser was coexpressed with the mixture of wild-type connexins at a cRNA ratio of 2:1:1, it almost completely blocked the channel activity of both wild types. Coexpression of Ala96Ser with the wild-type connexins also caused a considerable reduction in wild-type activity.

 

View this table: [in this window] [in a new window]   Table 3. Dual Whole-Cell Voltage-Clamp Recordings from N2A Cells Transfected with Human Wild-Type and Mutant Connexin 40.

 
To test the hypothesis that GJA5 mutants act as dominant negative inhibitors of wild-type connexin 40 activity, we compared the ability of wild-type connexin 40 and mutant connexin 40 to induce gap-junctional coupling in pairs of oocytes of the genus xenopus. Oocyte pairs expressing wild-type connexin 40 had high levels of coupling. In contrast, oocyte pairs expressing the Pro88Ser mutant had no detectable coupling. Furthermore, the Pro88Ser mutant completely inhibited the activity of coexpressed wild-type connexin 40 (Figure 4B, left-hand plot). The Ala96Ser variant also had a marked reduction in gap-junctional coupling, as compared with wild-type connexin 40. The coexpression of Ala96Ser and wild-type connexin 40 significantly inhibited the activity of wild-type connexin 40 (Figure 4B, center plot).

To determine whether mutant connexin 40 could interfere with the function of other cardiac connexins, oocytes were injected with a 1:1 mixture of cRNA for wild-type connexin 40 and wild-type connexin 43, either alone or in combination with an equal amount of cRNA for a connexin 40 mutant. Oocyte pairs injected with the mixture of wild-type connexin 43 and wild-type connexin 40 had high levels of gap-junctional coupling. When the Pro88Ser or Ala96Ser mutant was coexpressed with the mixture of wild-type connexins at a cRNA ratio of 2:1:1, the channel activity of both wild-type connexins was considerably reduced (Figure 4B, right-hand plot).

Discussion

We identified tissue-specific or somatic mutations in the GJA5 gene, which encodes connexin 40, in 3 of 15 patients with idiopathic atrial fibrillation. A fourth patient had a germ-line sequence variant. These patients had an early age at onset of atrial fibrillation, and clinical investigations had ruled out coexisting valvular or hypertensive heart disease.

Molecular mechanisms leading to tissue mosaicism include chimerism or spontaneous mutation in a progenitor-cell lineage. The possibility of chimerism was excluded in our study, since cardiac-tissue specimens (containing mutant cells) and lymphocytes from the same patient had identical polymorphic DNA markers (data not shown). Therefore, since myocardial cells do not divide, somatic mosaicism in cardiac tissue must have resulted from a somatic mutation in an early myocardial progenitor cell during embryogenesis.

Our results are supported by the finding of Lerman et al. of a somatic mutation of the G-protein subunit G_ai2 in cardiac tissue that caused arrhythmia.²⁴ Similar to our findings, the mutation was identified only in cardiac tissue and was absent in peripheral lymphocytes. Connexin-tissue mosaicism as a substrate for cardiac arrhythmia has also been demonstrated in chimeric connexin-43 mice.²⁵ Chimeric mice with patchy myocardial expression of connexin 43 had discrete areas of conduction delay, which presumably were responsible for the observed reentrant cardiac arrhythmias in these animals.²⁵ These data support the concept that tissue mosaicism of connexin 40 may be responsible for atrial arrhythmias in humans.

The functional relevance of the identified mutations was confirmed in protein localization experiments, by immunostaining of atrial tissue from patients, and in electrophysiological recordings of cell–cell coupling mediated by gap junctions. The immunostaining of atrial tissue from patients with the Pro88Ser and Gly38Asp substitutions confirmed the retention of intracellular connexin 40 and abnormal gap-junction formation present in a mosaic pattern, a finding consistent with somatic mutation in a subpopulation of cells. The Ala96Ser variant appeared to result in the assembly of gap junctions between adjacent cells; however, these paired cells consistently demonstrated absent or weak cell–cell electrical coupling. The coexpression of mutant and wild-type connexin 40 in xenopus oocytes significantly impaired functional cell–cell coupling, confirming that mutant connexin 40 acts as a dominant negative inhibitor of wild-type connexin 40 activity. Since connexin 43 is expressed in atrial tissue in addition to connexin 40 and evidence suggests that connexin 40 and connexin 43 may form heteromeric channels,²⁶ we confirmed that mutant connexin 40 also inhibits the functional coupling of wild-type connexin 43. Thus, genetic defects in GJA5 may induce a loss of function by means of two mechanisms: either impaired connexin transport and gap-junction assembly at the cell surface or the inhibition of cell–cell electrical coupling.

The Pro88Ser substitution was identified in two unrelated patients. It is interesting that this substitution has been identified in connexin 32 and connexin 50, in which it has been shown to cause dominantly inherited Charcot–Marie–Tooth disease and cataract disease, respectively.^{16,17,18,19,20} Even more interesting is the fact that the same Pro88Ser substitution has been identified in a naturally occurring strain of Caenorhabditis elegans that has abnormal electrical coupling in myocytes and a phenotype of uncoordinated locomotion.^27,28 This is an example of a genetic mutation in a gene family associated with three different diseases in humans, in addition to a disease phenotype in an invertebrate species. Our data on the function of the Pro88Ser mutant are consistent with those on the parallel Pro88Ser substitution in connexin 50, which causes familial cataracts.¹⁵ Pro88 has been implicated previously as a key residue in the transduction of voltage gating in the connexin-protein family, further highlighting the biologic significance of this amino acid in connexin proteins.^29,30

Studies involving high-resolution electrical mapping during atrial fibrillation have documented that multiple reentry circuits are a fundamental component of the perpetuation of the arrhythmia.^31,32 The myocardial substrate for reentry initiation is dependent on regional variability in conduction velocity, which creates localized asymmetry in the depolarizing wavefront through the myocardium.³³ A substrate with localized asymmetry in conduction promotes reentry, whereby depolarizing wavefronts continually reenter themselves, resulting in erratic and uncoordinated myocardial electrical activity. Since the cell–cell coupling of depolarizing current is mediated only at gap junctions, the biophysical properties of gap junctions are critical in ensuring normal myocardial conduction.¹² Atrial myocardium harboring a population of cells containing mutant connexin 40 protein that adversely affect electrical coupling would give rise to regions of heterogeneous conduction in tissue, providing a substrate that predisposes the atria to microreentry arrhythmias.

As is the case for all known cardiac arrhythmias with a genetic basis, periods of arrhythmia are episodic, despite the presence of mutant protein within the myocardial tissue of affected patients from birth. The onset of genetically based arrhythmias requires physiological stressors to precipitate the arrhythmia, presumably by exacerbating the already abnormal myocardial electrophysiology in the presence of mutant protein. In idiopathic atrial fibrillation, a common precipitant is increased vagal tone, mediated by muscarinic receptors.³⁴ The activation of muscarinic receptors has been shown to impair the cell–cell coupling mediated by gap junctions.³⁵ Together with our data, this observation provides a possible mechanistic link between enhanced vagal tone, impaired cell–cell electrical coupling, and the development of paroxysmal atrial fibrillation.

Although we identified GJA5 mutations in 4 of 15 patients, our patients were highly selected and this prevalence may not be representative of the prevalence of GJA5 sequence variations in a more random cohort of patients with atrial fibrillation. Larger-scale studies of patients with atrial fibrillation who have an age-related onset and coexisting coronary artery, valvular, or hypertensive heart disease will provide important additional data that should be considered before our findings are generalized. However, our findings provide evidence of the critical role of connexin 40 in atrial electrophysiology and evidence of a new paradigm — that common diseases traditionally considered idiopathic may have a genetic basis, with mutations confined to the diseased tissue.

Supported in part by grants from the J.P. Bickell Foundation (to Dr. Gollob), the Heart and Stroke Foundation of Ontario (to Drs. Gollob and Krahn), Canada Research Chair Program (to Dr. Bai), the Canadian Institutes for Health Research (to Drs. Jones and Bai), and the National Institutes of Health (to Dr. Ebihara).

Drs. Jones, Krahn, and Gollob report having submitted a patent application for the Pro88Ser connexin 40 substitution. No other potential conflict of interest relevant to this article was reported.

We are indebted to our patients for their dedicated support; to Ms. Megan Fortier, Mrs. Bonnie Spindler, Mr. Robbie Davies, Ms. Morette Wong, Ms. Wendy Pritchett, Dr. Subrata Charkrabarti, Mr. David Mallott, and Ms. Pierrette Bolongo for technical assistance; and to Dr. Hongling Wang for both technical assistance and constructing the connexin 40 model.

Source Information

From the Department of Medicine, University of Ottawa Heart Institute, Ottawa (M.H.G., L.D., J.P.V., A.S.L.T., A.F.R.S., F.T.), and the Departments of Medicine (M.H.G., D.L.J., A.D.K., G.J.K., R.Y., A.C.S.), Physiology and Pharmacology (D.L.J., X.-Q.G., D.B.), Anatomy and Cell Biology (Q.S.), and Surgery (G.M.G.), University of Western Ontario, London, Ont. — both in Canada; and the Department of Physiology and Biophysics, Rosalind Franklin University of Medicine and Science, North Chicago, Ill. (X.L., L.E.).

Address reprint requests to Dr. Gollob at the Arrhythmia Research Laboratory and Division of Cardiology, University of Ottawa Heart Institute, Rm. H350, 40 Ruskin St., Ottawa, ON K1Y 4W7, Canada, or at mgollob{at}ottawaheart.ca.

References

1. Wolf PA, Abott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983-988. [Free Full Text]
2. Benjamin EJ, Wolf PA, D'Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation 1998;98:946-952. [Free Full Text]
3. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution, and gender of patients with atrial fibrillation analysis and implications. Arch Intern Med 1995;155:469-473. [Free Full Text]
4. Wattigney WA, Mensah GA, Croft JB. Increasing trends in hospitalization for atrial fibrillation in the United States, 1985 through 1999: implications for primary prevention. Circulation 2003;108:711-716. [Free Full Text]
5. Chen YH, Xu SJ, Bendahhou S, et al. KCNQ1 gain-of-function mutation in familial atrial fibrillation. Science 2003;299:251-254. [Free Full Text]
6. Brugada R, Tapscott T, Czernuszewicz GZ, et al. Identification of a genetic locus for familial atrial fibrillation. N Engl J Med 1997;336:905-911. [Free Full Text]
7. Ellinor PT, Shin JT, Moore RK, Yoerger DM, MacRae CA. Locus for atrial fibrillation maps to chromosome 6q14-16. Circulation 2003;107:2880-2883. [Free Full Text]
8. Erickson RP. Somatic gene mutation and human disease other than cancer. Mutat Res 2003;543:125-136. [CrossRef][Web of Science][Medline]
9. Puck JM, Straus SE. Somatic mutations -- not just for cancer anymore. N Engl J Med 2004;351:1388-1390. [Free Full Text]
10. Yamaguchi H, Calado RT, Ly H, et al. Mutations in TERT, the gene for telomerase reverse transcriptase, in aplastic anemia. N Engl J Med 2005;352:1413-1424. [Free Full Text]
11. Beck JA, Poulter M, Campbell TA, et al. Somatic and germline mosaicism in sporadic early-onset Alzheimer's disease. Hum Mol Genet 2004;13:1219-1224. [Free Full Text]
12. Kanno S, Saffitz JE. The role of myocardial gap junctions in electrical conduction and arrhythmogenesis. Cardiovasc Pathol 2001;10:169-177. [CrossRef][Web of Science][Medline]
13. Hagendorff A, Schumacher B, Kirchhoff S, Luderitz B, Willecke K. Conduction disturbances and increased atrial vulnerability in connexin40-deficient mice analyzed by transesophageal stimulation. Circulation 1999;99:1508-1515. [Free Full Text]
14. Firouzi M, Ramanna H, Kok B, et al. Association of human connexin40 gene polymorphisms with atrial vulnerability as a risk factor for idiopathic atrial fibrillation. Circ Res 2004;95:e29-e33. [Free Full Text]
15. Pal JD, Berthoud VM, Beyer EC, Mackay D, Shiels A, Ebihara L. Molecular mechanism underlying a Cx50-linked congenital cataract. Am J Physiol 1999;276:C1443-C1446. [Free Full Text]
16. Nelis E, Simokovic S, Timmerman V, et al. Mutation analysis of the connexin32 (Cx32) gene in Charcot-Marie-Tooth neuropathy type 1: identification of five new mutations. Hum Mutat 1997;9:47-52. [CrossRef][Web of Science][Medline]
17. Janssen EA, Kemp S, Hensels GW, et al. Connexin32 gene mutations in X-linked dominant Charcot-Marie-Tooth disease (CMTX1). Hum Genet 1997;99:501-505. [CrossRef][Web of Science][Medline]
18. Bort S, Nelis E, Timmerman V, et al. Mutational analysis of the MPZ, PMP22, and Cx32 genes in patients of Spanish ancestry with Charcot-Marie-Tooth disease and hereditary neuropathy with liability to pressure palsies. Hum Genet 1997;99:746-754. [CrossRef][Web of Science][Medline]
19. Shiels A, Mackay D, Ionides A, Berry V, Moore A, Bhattacharya SA. A missense mutation in the human connexin50 gene (GJA8) underlies autosomal dominant "zonular pulverulent" cataract, on chromosome 1q. Am J Hum Genet 1998;62:526-532. [CrossRef][Web of Science][Medline]
20. Kuntzer T, Dunand M, Schorderet DF, Vallat JM, Hahn AF, Bogousslavsky J. Phenotypic expression of a Pro 87 to Leu mutation in the connexin 32 gene in a large Swiss family with Charcot-Marie-Tooth neuropathy. J Neurol Sci 2003;207:77-86. [CrossRef][Web of Science][Medline]
21. Bayazit YA, Cable BB, Cataloluk O, et al. GJB2 gene mutations causing familial hereditary deafness in Turkey. Int J Pediatr Otorhinolaryngol 2003;67:1331-1335. [CrossRef][Web of Science][Medline]
22. Krutovskikh V, Yamasaki H. Connexin gene mutations in human genetic diseases. Mutat Res 2000;462:197-207. [CrossRef][Web of Science][Medline]
23. Dilaveris PE, Gialafos EJ, Sideris SK, et al. Simple electrocardiographic markers for the prediction of paroxysmal idiopathic atrial fibrillation. Am Heart J 1998;135:733-738. [CrossRef][Web of Science][Medline]
24. Lerman BB, Dong B, Stein KM, Markowitz SM, Linden J, Catanzaro DF. Right ventricular outflow tract tachycardia due to a somatic cell mutation in G protein subunit ai2. J Clin Invest 1998;101:2862-2868. [Web of Science][Medline]
25. Gutstein DE, Morely GE, Vaidya D, et al. Heterogeneous expression of gap junction channels in the heart leads to conduction defects and ventricular dysfunction. Circulation 2001;104:1194-1199. [Free Full Text]
26. Valiunas V, Gemel J, Brink PR, Beyer EC. Gap junction channels formed by coexpressed connexin40 and connexin43. Am J Physiol Heart Circ Physiol 2001;281:H1675-H1689. [Free Full Text]
27. Phelan P, Starich TA. Innexins get into the gap. Bioessays 2001;23:388-396. [CrossRef][Web of Science][Medline]
28. Barnes TM, Hekimi S. The Caenorhabditis elegans avermectin resistance and anesthetic response gene unc-9 encodes a member of a protein family implicated in electrical coupling of excitable cells. J Neurochem 1997;69:2251-2260. [Web of Science][Medline]
29. Suchyna TM, Xu LX, Gao F, Fourtner CR, Nicholson BJ. Identification of a proline residue as a transduction element involved in voltage gating of gap junctions. Nature 1993;365:847-849. [CrossRef][Medline]
30. Ri Y, Ballesteros JA, Abrams CK, et al. The role of a conserved proline in mediating conformational changes associated with voltage gating of Cx32 gap junctions. Biophys J 1999;76:2887-2898. [Web of Science][Medline]
31. Cox JL, Canavan TE, Schuessler RB, et al. The surgical treatment of atrial fibrillation. II. Intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and fibrillation. J Thorac Cardiovasc Surg 1991;101:406-426. [Abstract]
32. Konings KT, Kirchhof CJ, Smeets JR, Wellens HJ, Penn OC, Allessie MA. High-density mapping of electrically induced atrial fibrillation in humans. Circulation 1994;89:1665-1680. [Free Full Text]
33. Spach MS, Josephson ME. Initiating reentry: the role of nonuniform anisotropy in small circuits. J Cardiovasc Electrophysiol 1994;5:182-209. [Web of Science][Medline]
34. Coumel P. Role of the autonomic nervous system in paroxysmal atrial fibrillation. In: Waldo AL, Touboul T, eds. Atrial flutter: advances in mechanisms and management. Armonk, N.Y.: Futura Publishing, 1996:248-61.
35. Fritz S, Kunz L, Dimitrijevic N, Grunert R, Heiss C, Mayerhofer A. Muscarinic receptors in human luteinized granulosa cells: activation blocks gap junctions and induces the transcription factor early growth response factor-1. J Clin Endocrinol Metab 2002;87:1362-1367. [Free Full Text]

Abstract PDF PDA Full Text PowerPoint Slide Set Supplementary Material Editorial by Saffitz, J. E. Add to Personal Archive Add to Citation Manager Notify a Friend E-mail When Cited E-mail When Letters Appear PubMed Citation

This article has been cited by other articles:

• Kirchhof, P., Bax, J., Blomstrom-Lundquist, C., Calkins, H., Camm, A. J., Cappato, R., Cosio, F., Crijns, H., Diener, H.-C., Goette, A., Israel, C. W., Kuck, K.-H., Lip, G. Y.H., Nattel, S., Page, R. L., Ravens, U., Schotten, U., Steinbeck, G., Vardas, P., Waldo, A., Wegscheider, K., Willems, S., Breithardt, G. (2009). Early and comprehensive management of atrial fibrillation: Proceedings from the 2nd AFNET/EHRA consensus conference on atrial fibrillation entitled 'research perspectives in atrial fibrillation'. Europace 11: 860-885 [Full Text]  
• Kaab, S., Darbar, D., van Noord, C., Dupuis, J., Pfeufer, A., Newton-Cheh, C., Schnabel, R., Makino, S., Sinner, M. F., Kannankeril, P. J., Beckmann, B. M., Choudry, S., Donahue, B. S., Heeringa, J., Perz, S., Lunetta, K. L., Larson, M. G., Levy, D., MacRae, C. A., Ruskin, J. N., Wacker, A., Schomig, A., Wichmann, H.-E., Steinbeck, G., Meitinger, T., Uitterlinden, A. G., Witteman, J. C.M., Roden, D. M., Benjamin, E. J., Ellinor, P. T. (2009). Large scale replication and meta-analysis of variants on chromosome 4q25 associated with atrial fibrillation. Eur Heart J 30: 813-819 [Abstract] [Full Text]  
• Benjamin, E. J., Chen, P.-S., Bild, D. E., Mascette, A. M., Albert, C. M., Alonso, A., Calkins, H., Connolly, S. J., Curtis, A. B., Darbar, D., Ellinor, P. T., Go, A. S., Goldschlager, N. F., Heckbert, S. R., Jalife, J., Kerr, C. R., Levy, D., Lloyd-Jones, D. M., Massie, B. M., Nattel, S., Olgin, J. E., Packer, D. L., Po, S. S., Tsang, T. S.M., Van Wagoner, D. R., Waldo, A. L., Wyse, D. G. (2009). Prevention of Atrial Fibrillation: Report From a National Heart, Lung, and Blood Institute Workshop. Circulation 119: 606-618 [Abstract] [Full Text]  
• Severs, N. J., Bruce, A. F., Dupont, E., Rothery, S. (2008). Remodelling of gap junctions and connexin expression in diseased myocardium. Cardiovasc Res 80: 9-19 [Abstract] [Full Text]  
• Tsai, C.-T., Lai, L.-P., Hwang, J.-J., Lin, J.-L., Chiang, F.-T. (2008). Molecular Genetics of Atrial Fibrillation. J Am Coll Cardiol 52: 241-250 [Abstract] [Full Text]  
• Hodgson-Zingman, D. M., Karst, M. L., Zingman, L. V., Heublein, D. M., Darbar, D., Herron, K. J., Ballew, J. D., de Andrade, M., Burnett, J. C. Jr., Olson, T. M. (2008). Atrial Natriuretic Peptide Frameshift Mutation in Familial Atrial Fibrillation. NEJM 359: 158-165 [Abstract] [Full Text]  
• Schoonderwoerd, B. A., Smit, M. D., Pen, L., Van Gelder, I. C. (2008). New risk factors for atrial fibrillation: causes of 'not-so-lone atrial fibrillation'. Europace 10: 668-673 [Abstract] [Full Text]  
• Nattel, S., Burstein, B., Dobrev, D. (2008). Atrial Remodeling and Atrial Fibrillation: Mechanisms and Implications. Circ Arrhythmia Electrophysiol 1: 62-73 [Full Text]  
• Sinner, M. F., Pfeufer, A., Akyol, M., Beckmann, B.-M., Hinterseer, M., Wacker, A., Perz, S., Sauter, W., Illig, T., Nabauer, M., Schmitt, C., Wichmann, H.-E., Schomig, A., Steinbeck, G., Meitinger, T., Kaab, S. (2008). The non-synonymous coding IKr-channel variant KCNH2-K897T is associated with atrial fibrillation: results from a systematic candidate gene-based analysis of KCNH2 (HERG). Eur Heart J 29: 907-914 [Abstract] [Full Text]  
• Kanaporis, G., Mese, G., Valiuniene, L., White, T. W., Brink, P. R., Valiunas, V. (2008). Gap Junction Channels Exhibit Connexin-specific Permeability to Cyclic Nucleotides. JGP 131: 293-305 [Abstract] [Full Text]  
• Hatcher, C. J., Basson, C. T. (2008). Disrupted Intercalated Discs: Is Kindlin-2 Required?. Circ. Res. 102: 392-394 [Full Text]  
• Ehrlich, J. R., Biliczki, P., Hohnloser, S. H., Nattel, S. (2008). Atrial-Selective Approaches for the Treatment of Atrial Fibrillation. J Am Coll Cardiol 51: 787-792 [Abstract] [Full Text]  
• Laird, D. W. (2008). Closing the Gap on Autosomal Dominant Connexin-26 and Connexin-43 Mutants Linked to Human Disease. J. Biol. Chem. 283: 2997-3001 [Abstract] [Full Text]  
• Lehnart, S. E., Ackerman, M. J., Benson, D. W. Jr, Brugada, R., Clancy, C. E., Donahue, J. K., George, A. L. Jr, Grant, A. O., Groft, S. C., January, C. T., Lathrop, D. A., Lederer, W. J., Makielski, J. C., Mohler, P. J., Moss, A., Nerbonne, J. M., Olson, T. M., Przywara, D. A., Towbin, J. A., Wang, L.-H., Marks, A. R. (2007). Inherited Arrhythmias: A National Heart, Lung, and Blood Institute and Office of Rare Diseases Workshop Consensus Report About the Diagnosis, Phenotyping, Molecular Mechanisms, and Therapeutic Approaches for Primary Cardiomyopathies of Gene Mutations Affecting Ion Channel Function. Circulation 116: 2325-2345 [Abstract] [Full Text]  
• Fatkin, D., Otway, R., Vandenberg, J. I. (2007). Genes and Atrial Fibrillation: A New Look at an Old Problem. Circulation 116: 782-792 [Full Text]  
• Gong, X.-Q., Shao, Q., Langlois, S., Bai, D., Laird, D. W. (2007). Differential Potency of Dominant Negative Connexin43 Mutants in Oculodentodigital Dysplasia. J. Biol. Chem. 282: 19190-19202 [Abstract] [Full Text]  
• Scheinman, M. M., Keung, E. (2007). The Year in Clinical Cardiac Electrophysiology. J Am Coll Cardiol 49: 2061-2069 [Full Text]  
• Sawaya, S. E., Rajawat, Y. S., Rami, T. G., Szalai, G., Price, R. L., Sivasubramanian, N., Mann, D. L., Khoury, D. S. (2007). Downregulation of connexin40 and increased prevalence of atrial arrhythmias in transgenic mice with cardiac-restricted overexpression of tumor necrosis factor. Am. J. Physiol. Heart Circ. Physiol. 292: H1561-H1567 [Abstract] [Full Text]  
• Peters, N. S. (2006). Gap Junctions: Clarifying the Complexities of Connexins and Conduction. Circ. Res. 99: 1156-1158 [Full Text]  
• Saffitz, J. E. (2006). Connexins, conduction, and atrial fibrillation.. NEJM 354: 2712-2714 [Full Text]