Screening for Copy Number Variation in Genes Associated with the Long QT Syndrome Clinical Relevance

Journal of the American College of Cardiology Vol. 57, No. 1, 2011 © 2011 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00 Published by Elsevier Inc. doi:10.1016/j.jacc.2010.08.621

Heart Rhythm Disorders

Screening for Copy Number Variation in Genes Associated With the Long QT Syndrome Clinical Relevance

Julien Barc, PHD,*‡§ François Briec, MD,*†‡§ Sébastien Schmitt, MD,ʈ Florence Kyndt, PharmD, PHD,*‡§ʈ Martine Le Cunff, BS,*‡§ Estelle Baron, BS,*‡§ Claude Vieyres, MD,¶ Frédéric Sacher, MD,# Richard Redon, PHD,*‡§ Cédric Le Caignec, MD, PHD,*‡§ʈ Hervé Le Marec, MD, PHD,*†‡§ Vincent Probst, MD, PHD,*†‡§ Jean-Jacques Schott, PHD*†‡§ Nantes, Angoulême, and Bordeaux, France

Objectives The aim of this study was to investigate, in a set of 93 mutation-negative long QT syndrome (LQTS) probands, the frequency of copy number variants (CNVs) in LQTS genes.

Background LQTS is an inherited cardiac arrhythmia characterized by a prolonged heart rate–corrected QT (QTc) interval associated with sudden cardiac death. Recent studies suggested the involvement of duplications or deletions in the occurrence of LQTS. However, their frequency remains unknown in LQTS patients.

Methods Point mutations in KCNQ1, KCNH2, and SCN5A genes were excluded by denaturing high-performance liquid chromatography or direct sequencing. We applied Multiplex Ligation-dependent Probe Amplification (MLPA) to detect CNVs in exons of these 3 genes. Abnormal exon copy numbers were confirmed by quantitative multiplex PCR of short fluorescent fragment (QMPSF). Array-based comparative genomic hybridization (array CGH) analysis was performed using Agilent Human Genome 244K Microarrays to further map the genomic rearrangements.

Results We identiﬁed 3 different deletions in 3 unrelated families: 1 in KCNQ1 and 2 involving KCNH2. We showed in the largest family that the deletion involving KCNH2 is fully penetrant and segregates with the long QT phenotype in 7 affected members.

Conclusions Our study demonstrates that CNVs in KCNQ1 and KCNH2 explain around 3% of LQTS in patients with no point mutation in these genes. This percentage is likely higher than the frequency of point mutations in ANKB, KCNE1, KCNE2, KCNJ2, CACNA1C, CAV3, SCN4B, AKAP9, and SNTA1 together. Thus, we propose that CNV screening in KCNQ1 and KCNH2 may be performed routinely in LQTS patients. (J Am Coll Cardiol 2011;57:40–7) © 2011 by the American College of Cardiology Foundation

Long QT syndrome (LQTS) is an inherited cardiac ar- sudden death caused by torsades de pointes or polymorphic rhythmia characterized by a prolonged heart rate–corrected ventricular tachycardia. LQTS can be an autosomal reces- QT (QTc) interval, which is associated with syncope and sive disorder (1), but the most common form is an autosomal dominant disorder called Romano-Ward syndrome

From *INSERM, UMR915, l’institut du thorax, Service de cardiologie Nantes, See page 48 Nantes, France; †CHU Nantes, l’institut du thorax, Service de cardiologie Nantes, Nantes, France; ‡CNRS, ERL3147, Nantes, France; §Université de Nantes, Nantes, France; ʈCHU Nantes, Service de génétique médicale, Nantes, France; ¶Cabinet de (2,3). LQTS affects between 1 in 5,000 and 1 in 2,000 cardiologie, Angoulême, France; and the #Service de rythmologie, Hôpital cardi- ologique du Haut Leveque, Bordeaux, France. This work was supported by Pro- individuals (4,5). Molecular diagnosis is an important tool gramme Hospitalier de Recherche Clinique 2001 R20/03 and 2004 R20/07 from to guide diagnosis, treatment, and prevention strategies in Centre Hospitalier Universitaire de Nantes, France; Société française de cardiologie, LQTS patients. To date, more than 600 mutations (6) have Paris, France; the fondation Leducq Trans-Atlantic Network of Excellence grant (05 CVD 01, Preventing Sudden death), Paris, France; and Groupe de Réﬂexion sur la been identiﬁed among 12 different genes: 5 genes encoding Recherche Cardiovasculaire (PhD grant to Dr. Barc), Paris, France. The authors have ion channel alpha subunits (KCNQ1 [7], KCNH2 [8], reported that they have no relationships to disclose. Drs. Barc, Briec, and Schmitt SCN5A [9], KCNJ2 [10], and CACNA1C [11]) and 7 genes contributed equally to this work. Manuscript received February 24, 2010; revised manuscript received July 16, 2010, encoding ion channel regulatory proteins (ANKB [12], accepted August 10, 2010. KCNE1 [13], KCNE2 [14], CAV3 [15], SCN4B [16], JACC Vol. 57, No. 1, 2011 Barc et al. 41 December 28, 2010/January 4, 2011:40–7 Copy Number Variation in LQTS Genes

AKAP9 [17], and SNTA1 [18]). In total, molecular diagno- MRC-Holland protocol (25). The Abbreviations sis can resolve up to 70% of cases. More than 90% of those SALSA P114 MLPA kit contains and Acronyms cases are due to mutations in KCNQ1, KCNH2, and SCN5A, 20 probes interrogating the comparative ؍ CGH corresponding to LQT1, LQT2, and LQT3, respectively KCNQ1 gene, 9 for the KCNH2 genomic hybridization copy number variant ؍ The lack of mutation detection in the remaining cases gene, and 3 for the SCN5A gene. CNV .(19,20) has been attributed to phenotyping errors, incomplete sensi- Abnormal profiles in MLPA anal- -denaturing high ؍ dHPLC tivity of screening methods (denaturing high-performance ysis were completed with a locus- performance liquid liquid chromatography [dHPLC]) and direct sequencing), specific Quantitative Multiplex chromatography /electrocardiogram ؍ mutations in noncoding regions, or mutations in as of yet PCR of Short Fluorescent Frag- ECG unknown genes. ment (QMPSF) (26) (see the eclectrocardiographic implantable ؍ Another source of negative molecular screening could be supplemental Online Methods). ICD the presence of copy number variants (CNVs) affecting the Oligonucleotide complementary cardioverter-defibrillator logarithm of the ؍ major genes for LQTS, that would not be detectable using sequences to exons 5 and 15 of LOD capillary sequencing. Interestingly, Bisgaard et al. (21) KCNH2 or exon 7, intron 7-8, and odds long QT syndrome ؍ described in 2006 a large deletion (217 genes) including exon 8 of KCNQ1 were coampli- LQTS -Multiplex Ligation ؍ KCNH2 in a patient with mental retardation and a LQTS. fied by PCR with an additional MLPA In addition, Koopmann et al. (22) detected a 3.7-kb fragment, corresponding to exon dependent Probe intragenic KCNH2 duplication in a Dutch family affected by 14 of MLH1, a gene located on Amplification quantitative ؍ LQTS. More recently, Eddy et al. (23) identified 2 dele- chromosome 11 used as a control. QMPSF multiplex PCR of short tions—1 in KCNQ1, a second in KCNH2—and a duplica- The QMPSF conditions and fluorescent fragment tion in KCNH2 in LQTS patients. These different studies primer sequences are available heart ؍ QTc suggest that gene duplications or deletions can explain upon request. Quantitative PCR rate–corrected QT interval LQTS. However, the precise frequency of CNVs involving (qPCR) experiments were per- the main LQTS genes in patients with LQTS remains formed using the LightCycler unknown. In this study, we investigated the involvement of 480 (Roche Molecular Systems, Mannheim, Germany) to rare deletions and duplications affecting KCNH2 and validate genes with variable copy number. PCR reactions KCNQ1 in 93 probands with LQTS and in particular in 1 were prepared using the Power SYBR-Green PCR reagent large family for which no putatively causative point muta- kit (Applied Biosystems, Foster City, California) according tions had been identified previously. to the manufacturer’s protocol. Array comparative genomic hybridization. Array comparative genomic hybridization (CGH) analysis was per- Methods formed using the Agilent Human Genome Microarray Kit LQTS patients. This study was in agreement with the 244A (Agilent Technologies, Santa Clara, California). La- local guidelines for genetic research and has been approved beling, hybridization, washes, and data analysis were per- by the local ethical committee. Two experts for rare arrhyth- formed according to the protocol provided by Agilent mic diseases at the University Hospital of Nantes defined (Protocol version 4.0, June 2006). Graphical overviews were the LQTS phenotype by independent electrocardiogram obtained using the CGH Analytics software (version 3.4, (ECG) readings. Diagnosis of LQT syndrome was based on Agilent Technologies). DNA sequence information refers the QTc duration, the morphology of the T-wave, and the to the public UCSC Genome Browser database (Human patient’s clinical and family history. The Schwartz score has Genome Browser, March 2006 Assembly). also been calculated for the 93 patients, and all of them have Linkage analysis. Two-point linkage analysis was performed a Schwartz score of 3 or greater. QTc duration was with easy LINKAGE Plus software (version 5.02), by using an calculated according to Bazett’s formula. A prolongation of autosomal-dominant model of inheritance with complete pen- the QTc duration was defined as Ն440 ms for men etrance and a disease allele frequency of 0.001 (T. Lindner, (borderline between 430 and 439 ms) and as Ն460 ms for University of Würzburg, Würzburg, Germany). women (borderline between 450 and 459 ms) (24). Each patient underwent full medical examination to rule out Results syndromic forms of QT prolongation. Blood samples were collected after written informed consent. Mutations in Ninety-three patients with LQTS were included in this coding regions and exon–intron boundaries for the 3 main study. No potentially causative point mutations in the LQTS-causing genes—KCNQ1, KCNH2, and SCN5A— KCNQ1, KCNH2, and SCN5A genes were identified by were excluded by dHPLC or direct sequencing. dHPLC or direct sequencing in any of these individuals. MLPA, QMPSF, and qPCR analyses. Multiplex Ligation- The patients (63% women) showed an average QTc dura- dependent Probe Amplification (MLPA) was performed tion of 556 Ϯ 60 ms and an average age at diagnosis of 35 using the SALSA P114 MLPA kit (MRC-Holland, Ϯ 22 years. Among the 93 patients, 33 (36%) have a Amsterdam, the Netherlands) and according to the Schwartz score of 3 and 60 (64%) show a Schwartz score 42 Barc et al. JACC Vol. 57, No. 1, 2011 Copy Number Variation in LQTS Genes December 28, 2010/January 4, 2011:40–7

Figure 1 Family 1: Clinical Findings

(A) The disease phenotype is transmitted as an autosomal-dominant trait. Open symbols depict unaffected members; solid symbols, long QT phenotypes; gray symbols, undetermined members; and question marks, unknown phenotypes. Circles indicate females; and squares, males. The proband is indicated by an arrow. Genotypes are marked with ϩ/Ϫ for heterozygous mutation and ϩ/ϩ for wild type. Heart rate-corrected QT interval (QTc) values (Bazett formula) are indicated below each individual. Proband’s electrocardiogram (ECG) of arrhythmias is presented. (B) ECGs are presented for affected members. ICD ϭ implantable cardioverter-defibrillator. of 4 or greater. Twenty-three patients (25%) had been ventricular fibrillation successfully resuscitated at age 23 years. resuscitated from sudden cardiac death, 58 (62%) had Her ECG showed a prolonged QT interval (QTc ϭ 554 ms) presented with a syncope, 30 (32%) with an arrhythmia and a bifid T-wave strongly suggestive of LQT2 syndrome event, and 15 (16%) had been implanted with an im- (Fig. 1). Because of the aborted cardiac arrest, she was plantable cardioverter-defibrillator (ICD). The T-wave implanted with an ICD. Her clinical examination found no patterns for all patients were classified into 5 categories: other abnormalities. Familial recruitment led to the identifica- normal T-wave morphology (n ϭ 3, 3%), LQT1 (n ϭ 11, tion of 16 relatives through 3 generations (Fig. 1A). In 12%), LQT2 (n ϭ 47, 51%), LQT3 (n ϭ 13, 14%), and addition to patient III:4, 6 family members had ECG abnor- nonspecific (n ϭ 19, 20%). malities strongly suggestive of LQTS (Fig. 1B): patient I:2 We screened for CNV involving the KCNQ1, KCNH2, (QTc ϭ 465 ms), patient II:1 (QTc ϭ 590 ms), patient II:7 and SCN5A genes in these 93 patients, by MLPA. We (QTc ϭ 615 ms), patient III:1 (QTc ϭ 670 ms), patient III:5 identified 1 heterozygote deletion in KCNQ1 and 2 hetero- (QTc ϭ 518 ms), and patient III:6 (QTc ϭ 610 ms). Each of zygote deletions involving KCNH2 in 3 unrelated patients. these relatives was asymptomatic. ECGs performed in the No CNVs were detected in SCN5A. Familial investigations other family members were normal (data not shown). were carried out for those 3 cases. Using MLPA and QMPSF, a heterozygote genomic Family 1. Patient III:4 was diagnosed for LQTS after the deletion was identified in patient III:4 involving exons 4 to occurrence of several episodes of torsades de pointes and 14 of KCNH2 (Fig. 2A). JACC Vol. 57, No. 1, 2011 Barc et al. 43 December 28, 2010/January 4, 2011:40–7 Copy Number Variation in LQTS Genes

Figure 2 Detection of CNVs Involving the KCNH2 and KCNQ1 Genes in 3 Long QT Patients by MLPA

Each proﬁle corresponds to relative peak ratios for control probes, KCNH2, KCNQ1, and SCN5A. The asterisks mark the detection of exonic deletions. (A) KCNH2/exons 4 to 14 are deleted in Family 1; (B) KCNH2/exons 1 to 14 are deleted in Family 2; (C) KCNQ1/exons 7 to 8 are deleted in Family 3. Y-axis represents the relative ratio of copy number. CNV ϭ copy number variant; Ex ϭ exon; Int ϭ intron; MLPA ϭ Multiplex Ligation-dependent Probe Ampliﬁcation.

QMPSF was performed using 2 probes: the first in exon was performed to evaluate segregation between the deletion 5 confirmed the deletion, and the second in exon 15 and the LQT phenotype. Under the dominant model of demonstrated that the 3= coding region of KCNH2 was also inheritance, we obtained a maximum logarithm of the odds deleted (data not shown). (LOD) score of 3.13 (␪ ϭ 0%). High-resolution array CGH analysis refined the size of the Family 2. The proband is a woman diagnosed at age 28 years genomic rearrangement: the deletion, delimited by probes with LQTS after the occurrence of syncope triggered by A_16_P18164054 (intron 3-4, centromeric breakpoint) to acoustic stimulus (phone ringing). Her ECG showed a pro- A_16_P01835269 (telomeric breakpoint), is 650 kb in length. longed QT interval (QTc ϭ 563 ms) and a bifid T-wave. Nineteen other genes are included in this rearrangement ECGs were performed in the first-degree relatives: the mother, (including ABP1)(Fig. 3A), which maps to band 7q36.1. the brother, and the 2 sons (Fig. 4A). Prolongation of the QT Interestingly, Redon et al. (27) identified a CNV in ABP1 interval was found only in the mother (I:2; QTc ϭ 467 ms). downstream of the KCNH2 gene in individual NA12762 MLPA analysis showed a heterozygous deletion of (chr7: 149, 976, 469–150, 101, 345) from the HapMap KCNH2 (Fig. 2B). QMPSF confirmed the deletion of collection. We checked whether KCNH2 is encompassed by KCNH2 (probes in exons 5 and 15) and showed the deletion this CNV, using QMPSF on KCNH2 exons 5 and 15 and of ABP1 (probe in exon 2; data not shown). Array CGH ABP1 exon 2. We found that the deletion identified by analysis revealed that the CNV is 145 kb in length (probe Redon et al. is limited to ABP1 and does not affect KCNH2 A_16_P18165049 to probe A_16_P1835307) and includes (data not shown). entire copies of the KCNH2 and ABP1 genes (7q36.1) The KCNH2 CNV was present in the 6 other affected (Fig. 3B). family members (I:2, II:1, II:7, III:1, III:5, and III:6) but The same CNV was also found in 2 relatives: her affected absent from 9 healthy members (Table 1). Linkage analysis mother (I:2) and her healthy brother (II:1) (Table 1). 44 Barc et al. JACC Vol. 57, No. 1, 2011 Copy Number Variation in LQTS Genes December 28, 2010/January 4, 2011:40–7

Figure 3 Array-Based CGH Allows Better Delineation of Deletion Breakpoints

(A) The proband from Family 1 carries a 650-kb-long deletion including KCNH2 and 19 other genes. (B) The proband in Family 3 carries a 145-kb-long deletion including KCNH2 and APB1. Black arrows indicate the clones ﬂanking the deletion breakpoints. CGH ϭ comparative genomic hybridization.

Family 3. Patient II.1 was a girl diagnosed at age 14 years and 8 in proband II:1 (Fig. 2C). The CNV was conﬁrmed with the LQTS after the occurrence of an episode of by 2 independent techniques: QMPSF with probes in exons syncope resulting from a stress with a temporary loss of 7 and 8, and qPCR with primers in exons 7 and 8 as well as hearing and tachycardia. Her ECG showed a prolonged QT intron 7-8 (data not shown). The CNV was inherited from interval (QTc ϭ 490 ms) as well as a broad-base T-wave the proband’s father (I:1), whereas the proband’s sister (II:2) morphology on derivations V4 to V6. Her father presented did not carry the deletion (Fig. 4B, Table 1). with a T-wave pattern suggestive of LQT1 and a borderline QTc duration of 438 ms (Fig. 4B). Her mother and sister Discussion have normal T-wave morphology and QT interval duration (data not shown). CNV detection in LQTS genes. In this study, we evalu- We identiﬁed by MLPA 1 heterozygote deletion that ated the involvement of CNVs in KCNQ1, KCNH2, and maps to band 11p15.5 and is restricted to KCNQ1 exons 7 SCN5A genes in patients affected by a typical form of JACC Vol. 57, No. 1, 2011 Barc et al. 45 December 28, 2010/January 4, 2011:40–7 Copy Number Variation in LQTS Genes

ClinicalTable 1 andClinical Genetic and Data Genetic for Families Data 1,for 2, Families and 3 1, 2, and 3

Individual Sex Age (yrs) QTc (ms) Symptoms Phenotype CNV Carrier/Noncarrier Family 1 I:1 M 72 422 — Unaffected Noncarrier I:2 F 70 465 — Affected Carrier II:1 M 50 590 — Affected Carrier II:2 F 50 430 — Unaffected Noncarrier II:3 M 45 392 — Unaffected Noncarrier II:4 M 41 388 — Unaffected Noncarrier II:6 M 46 430 — Unaffected Noncarrier II:7 F 49 505 — Affected Carrier III:1 M 16 670 — Affected Carrier III:2 F 22 421 — Unaffected Noncarrier III:3 F 29 442 — Unaffected Noncarrier III:4 (proband) F 25 554 TdP/VF/SD resuscitated Affected Carrier III:5 F 18 518 — Affected Carrier III:6 F 18 610 — Affected Carrier III:7 F 24 438 — Unaffected Noncarrier III:8 F 22 423 — Unaffected Noncarrier Family 2 I:2 F 53 467 — Affected Carrier II:1 M 30 402 — Unaffected Carrier II:3 (proband) F 31 563 Syncope (after phone ringing) Affected Carrier III:1 M 7 389 — Unaffected Noncarrier III:2 M 3 437 — Undetermined Noncarrier Family 3 I:1 M 49 438 — Undetermined Carrier I:2 F 53 438 — Unaffected Noncarrier II:1 (proband) F 18 490 Syncope (after stress) Affected Carrier II:2 F 20 431 — Unaffected Noncarrier

CNV ϭ copy number variation; SD ϭ sudden death; TdP ϭ torsades de pointes; VF ϭ ventricular ﬁbrillation.

LQTS, for whom previous molecular analysis had failed to components of the delayed rectifier potassium current) identify point mutations in the known major LQTS genes. (29,30). We detected 3 CNVs in 3 unrelated patients: 1 deletion in Family segregation analysis. We found perfect cosegrega- KCNQ1 and 2 deletions involving KCNH2. tion between long QT phenotype and CNV inheritance in Functional effect of the CNVs. The first deletion includes Family 1 (odds [LOD] score Ͼ3), demonstrating that the the exons 4 to 15 of KCNH2. This deletion could lead to a KCNH2 deletion is responsible for this familial form of truncated protein. More probably, its effect may be similar LQTS. One nonpenetrant patient (II:1) was identified in to that of a premature termination codon mutation with Family 2, in line with the observations of a previous study subsequent nonsense-mediated decay degradation of the from others (31). In Family 3, the KCNQ1 deletion was KCNH2 mRNA (28). Although this deletion includes 19 found in 2 members presenting a T-wave pattern suggestive other genes, the patient presented with no abnormal phe- of LQT1. One of them presents with a prolonged QTc notype other than prolonged QT duration. duration, the other with a borderline value (Fig. 4B). The second deletion includes the whole KCNH2 Clinical implications. The 3 deletions identified are ex- gene, probably leading to haploinsufficiency and decreas- pected to lead to haploinsufficiency. Nonsense mutations or Յ ing levels of the potassium current IKr (rapid components frameshift mutations brought about 50% reduction in of the delayed rectifier potassium current) in ventricular cardiac repolarizing IKs or IKr potassium channel current. cardiomyocytes. They are associated with a less severe phenotype than The third deletion spans exons 7 to 8 of KCNQ1. dominant-negative mutations but are known to cause sud- Assuming that the deletion includes full copies of both den cardiac death (32). This less severe phenotype can exons, it leads to the lack of the second part of the p-loop explain the nonpenetrant case in Family 2 and the number (including GYGD motif), as well as the S6 transmembrane of asymptomatic individuals. Furthermore, it is known that segment and 23 amino acids from the C-terminal part of the the nonpenetrant mutation carriers found in familial studies KvLQT1 channel subunit. Interestingly, previous studies could experience syncope or cardiac arrest (33). However, have described splicing variants, involving deletion of either more CNV studies are required to establish a genotype– exon 7, exon 8, or both, leading to decreased IKs (slow phenotype correlation. 46 Barc et al. JACC Vol. 57, No. 1, 2011 Copy Number Variation in LQTS Genes December 28, 2010/January 4, 2011:40–7

Figure 4 Families 2 and 3: Clinical Findings

(A) Family 2 and (B) Family 3. The disease phenotype is transmitted as an autosomal-dominant trait. Open symbols depict unaffected members; solid symbols, long QT phenotypes; gray symbols, undetermined members; and question marks, unknown phenotypes. Circles indicate females; and squares, males. The probands are indicated by arrows. Genotypes are marked with ϩ/Ϫ for heterozygous mutation and ϩ/ϩ for wild type. QTc values (Bazett formula) are indicated below each individual. ECG are presented for affected members with ECG of arrhythmia when documented. Abbreviations as in Figure 1.

Indeed, our approach is focused on the exonic portions of morbid gene such as BRCA1 (35). In our and others’ the 3 most relevant genes involved in LQTS. We can expect experience, the frequency of CNV detection in KCNH2 and that other functional and also nonfunctional CNVs may KCNQ1 is higher than the frequency of point mutations in exist within the portions of the genes that have not been ANKB, KCNE1, KCNE2, KCNJ2, CACNA1C, CAV3, investigated (introns, regulatory elements, coding region not SCN4B, AKAP9 and SNTA1 together (4,20). Thus, we covered by the MLPA kit). conclude that CNV detection in the 2 major LQTS genes We note that similarly to that observed for the previously may be performed for patients diagnosed with LQTS when described point mutations, the T-wave morphologies ob- no point mutation has been detected by sequencing for served in probands’ ECGs are predictive for which LQTS KCNQ1, KCNH2, and SCN5A. In our experience (0 of 93 gene is altered by CNV and thus remain a relevant indicator patients, 95% CI: 0 to 0.039) and from results of another of the morbid gene (34). study, CNVs involving SCN5A seems to be extremely rare in LQTS (36). Conclusions Our study demonstrates that genomic rearrangements in In summary, our study demonstrates that CNVs involving KCNQ1 and KCNH2 genes explain 3% of the LQTS cases major LQTS genes explain around 3% of additional LQTS in which no point mutation was found in the genes cases (3 out of 93 patients, 95% conﬁdence interval [CI]: commonly involved. In consequence, we propose that 0.007 to 0.09) after failure of the classical molecular diag- screening for genomic rearrangement may be considered in nostic screen (dHPLC and direct sequencing). This fre- the routine workup of LQTS in the absence of point quency corresponds to the percentage of CNVs versus point mutations in the 3 major LQTS genes and before screening mutations already described in the published reports for a for point mutations in other LQTS genes. JACC Vol. 57, No. 1, 2011 Barc et al. 47 December 28, 2010/January 4, 2011:40–7 Copy Number Variation in LQTS Genes

Acknowledgments 16. Medeiros-Domingo A, Kaku T, Tester DJ, et al. SCN4B-encoded The authors are greatly indebted to the patients and their sodium channel beta4 subunit in congenital long-QT syndrome. Circulation 2007;116:134–42. families. They thank Christine Fruchet, Marion Chaventré, 17. Chen L, Marquardt ML, Tester DJ, Sampson KJ, Ackerman MJ, Kass Christine Poulain, Patricia Bouillet, and Maïder Bessouet RS. Mutation of an A-kinase-anchoring protein causes long-QT for assistance in the family screening; Anne Ponchaux, syndrome. Proc Natl Acad SciUSA2007;104:20990–5. 18. Ueda K, Valdivia C, Medeiros-Domingo A, et al. Syntrophin muta- Jerome Buscail, Thierry Marsaud, Olivier Pichon, and tion associated with long QT syndrome through activation of the Simon Lecointe for technical assistance; Fabrice Airaud for nNOS-SCN5A macromolecular complex. Proc Natl Acad Sci U S A graphical representation matrix of MLPA data; and Dr. 2008;105:9355–60. 19. Kapplinger JD, Tester DJ, Salisbury BA, et al. Spectrum and preva- Stéphane Bézieau (Service de génétique médicale, Nantes) lence of mutations from the first 2,500 consecutive unrelated patients for his excellent work. The authors are also grateful to the referred for the FAMILION long QT syndrome genetic test. Heart Biogenouest platforms, in particular to Françoise Gros Rhythm 2009;6:1297–303. 20. Napolitano C, Priori SG, Schwartz PJ, et al. Genetic testing in the (plateforme de séquençage et de Génotypage), Catherine long QT syndrome: development and validation of an efficient ap- Chevalier, and Remi Houlgatte (plateforme transcriptome proach to genotyping in clinical practice. JAMA 2005;294:2975–80. et puce à ADN) for their technical support. 21. Bisgaard AM, Rackauskaite G, Thelle T, Kirchhoff M, Bryndorf T. Twins with mental retardation and an interstitial deletion 7q34q36.2 leading to the diagnosis of long QT syndrome. Am J Med Genet A Reprint requests and correspondence: Prof. Vincent Probst, 2006;140:644–8. Service de cardiologie du CHU de Nantes, CHU de Nantes, 22. Koopmann TT, Alders M, Jongbloed RJ, et al. Long QT syndrome Hôpital Nord, Bd Jacques Monod, 44093 Nantes Cedex, France. caused by a large duplication in the KCNH2 (HERG) gene undetect- able by current polymerase chain reaction-based exon-scanning meth- E-mail: [email protected]. odologies. Heart Rhythm 2006;3:52–5. 23. Eddy CA, MacCormick JM, Chung SK, et al. Identification of large gene deletions and duplications in KCNQ1 and KCNH2 in patients REFERENCES with long QT syndrome. Heart Rhythm 2008;5:1275–81. 24. Roden DM. Clinical practice. Long-QT syndrome. N Engl J Med 1. Jervell A, Lange-Nielsen F. Congenital deaf-mutism, functional heart 2008;358:169–76. disease with prolongation of the Q-T interval and sudden death. Am 25. MRC-Holland MLPA. Available at: http://www.mrc-holland.com. Heart J 1957;54:59–68. Accessed November 12, 2010. 2. Romano C, Gemme G, Pongiglione R. Rare cardiac arrhythmias of 26. Casilli F, Di Rocco ZC, Gad S, et al. Rapid detection of novel the pediatric age. II. Syncopal attacks due to paroxysmal ventricular BRCA1 rearrangements in high-risk breast-ovarian cancer families fibrillation. Clin Pediatr (Bologna) 1963;45:656–83. using multiplex PCR of short fluorescent fragments. Hum Mutat 3. Ward OC. A new familial cardiac syndrome in children. J Ir Med 2002;20:218–26. Assoc 1964;54:103–6. 27. Redon R, Ishikawa S, Fitch KR, et al. Global variation in copy number 4. Tester DJ, Will ML, Haglund CM, Ackerman MJ. Compendium of in the human genome. Nature 2006;444:444–54. cardiac channel mutations in 541 consecutive unrelated patients 28. Hentze MW, Kulozik AE. A perfect message: RNA surveillance and referred for long QT syndrome genetic testing. Heart Rhythm nonsense-mediated decay. Cell 1999;96:307–10. 2005;2:507–17. 29. Lande G, Kyndt F, Baro I, et al. Dynamic analysis of the QT interval 5. Schwartz PJ, Priori SG, Napolitano C. How really rare are rare in long QT1 syndrome patients with a normal phenotype. Eur Heart J diseases?: the intriguing case of independent compound mutations in 2001;22:410–22. the long QT syndrome. J Cardiovasc Electrophysiol 2003;14:1120–1. 30. Tsuji K, Akao M, Ishii TM, et al. Mechanistic basis for the 6. Fondazione Salvatore Maugeri. Gene Connection for the Heart. June pathogenesis of long QT syndrome associated with a common splicing 20, 2007. Available at: http://www.fsm.it/cardmoc/. Accessed Febru- mutation in KCNQ1 gene. J Mol Cell Cardiol 2007;42:662–9. ary 8, 2010. 31. Priori SG, Napolitano C, Schwartz PJ. Low penetrance in the 7. Wang Q, Curran ME, Splawski I, et al. Positional cloning of a novel long-QT syndrome: clinical impact. Circulation 1999;99:529–33. potassium channel gene: KVLQT1 mutations cause cardiac arrhyth- 32. Moss AJ, Shimizu W, Wilde AA, et al. Clinical aspects of type-1 mias. Nat Genet 1996;12:17–23. long-QT syndrome by location, coding type, and biophysical function 8. Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED, of mutations involving the KCNQ1 gene. Circulation 2007;115: Keating MT. A molecular basis for cardiac arrhythmia: HERG 2481–9. mutations cause long QT syndrome. Cell 1995;80:795–803. 33. Vincent GM, Timothy KW, Leppert M, Keating M. The spectrum of 9. Wang Q, Shen J, Li Z, et al. Cardiac sodium channel mutations in symptoms and QT intervals in carriers of the gene for the long-QT patients with long QT syndrome, an inherited cardiac arrhythmia. syndrome. N Engl J Med 1992;327:846–52. Hum Mol Genet 1995;4:1603–7. 34. Zhang L, Timothy KW, Vincent GM, et al. Spectrum of ST-T-wave 10. Plaster NM, Tawil R, Tristani-Firouzi M, et al. Mutations in Kir2.1 patterns and repolarization parameters in congenital long-QT syn- cause the developmental and episodic electrical phenotypes of Ander- drome: ECG findings identify genotypes. Circulation 2000;102: sen’s syndrome. Cell 2001;105:511–9. 2849–55. 11. Splawski I, Timothy KW, Sharpe LM, et al. Ca(V)1.2 calcium 35. Engert S, Wappenschmidt B, Betz B, et al. MLPA screening in the channel dysfunction causes a multisystem disorder including arrhyth- BRCA1 gene from 1,506 German hereditary breast cancer cases: novel mia and autism. Cell 2004;119:19–31. deletions, frequent involvement of exon 17, and occurrence in single 12. Mohler PJ, Schott JJ, Gramolini AO, et al. Ankyrin-B mutation early-onset cases. Hum Mutat 2008;29:948–58. causes type 4 long-QT cardiac arrhythmia and sudden cardiac death. 36. Koopmann TT, Beekman L, Alders M, et al. Exclusion of multiple Nature 2003;421:634–9. candidate genes and large genomic rearrangements in SCN5A in a 13. Splawski I, Tristani-Firouzi M, Lehmann MH, Sanguinetti MC, Dutch Brugada syndrome cohort. Heart Rhythm 2007;4:752–5. Keating MT. Mutations in the hminK gene cause long QT syndrome and suppress IKs function. Nat Genet 1997;17:338–40. Key Words: CNV y deletion y long QT syndrome y MLPA y 14. Abbott GW, Sesti F, Splawski I, et al. MiRP1 forms IKr potassium sudden death. channels with HERG and is associated with cardiac arrhythmia. Cell 1999;97:175–87. 15. Vatta M, Ackerman MJ, Ye B, et al. Mutant caveolin-3 induces APPENDIX persistent late sodium current and is associated with long-QT syndrome. Circulation 2006;114:2104–12. For supplemental methods, please see the online version of this article.