Tester, David J.

Introduction: Sudden Infant Death Syndrome (SIDS) is defined as the sudden death of an infant which remains unexplained after comprehensive clinical and pathological assessment. Prior studies have implicated inherited cardiac conditions (ICC). Hypothesis: Whole exome sequencing will identify a significant burden of putative pathogenic (PP) and rare functionally significant (RFS) cardiac genetic variants in SIDS. Methods: 96 SIDS cases with complete phenotypic data were selected from 434 cases. Prone sleeping position was excluded to enrich for genetic risk. Genomic DNA underwent target enrichment using Agilent SureSelect Human All Exon v5 kit and sequencing on Illumina HiSeq. Reads were aligned to GRCh37 reference genome with Novoalign and variants called with SAMtools. Principal component analysis identified 82 cases of North European (NE) ancestry. Variants were annotated with Annovar and compared to control data from a local exome database, 1000 Genomes and ESP. Analysis focused on 83 known ICC susceptibility genes (27 channelopathy-associated [CH]; 56 cardiomyopathy-associated [CM]). PP was defined as ultra rare nonsense or frameshift mutations or other variants previously reported as disease-associated and absent in controls. RFS was defined as variants with abnormal functional characterization and <0.5% prevalence in controls. Results: 20/82 cases (24%) carried at least 1 PP/RFS variant: 12 CH variants in 9 genes (AKAP9, ANK2, CACNB2, CAV3, KCNE2, RYR2, SCN5A, SNTA1, TRDN) in 13 cases; 7 CM variants in 5 genes (ABCC9, ACTN2, CSRP3, TCAP, TTN) in 8 cases. Variants were most commonly identified in the 3-6 months age group (p=0.009). There were no other significant phenotypic differences between variant-positive and negative cases (Table 1). Conclusions: There was an 11% yield of PP variants in non-prone sleeping SIDS cases of NE ancestry, with an additional 13% hosting RFS variants. These may provide a potential substrate in the pathophysiology of SIDS.

Introduction: Approximately 10 – 15% of sudden infant death syndrome (SIDS) may be caused by cardiac channelopathies including Brugada syndrome (BrS). Type 1 BrS (BrS1), due to mutations in the SCN5A-encoded sodium channel, accounts for approximately 20% of BrS. Recently, a novel mutation in glycerol-3-phosphate dehydrogenase 1-like (GPD1-L) disrupted trafficking of SCN5A in a multi-generational family with BrS. We hypothesized that mutations in GPD1-L may be responsible for some cases of SIDS. Methods: Using DHPLC and direct DNA sequencing, we performed comprehensive open reading frame/splice site mutational analysis of GPD1-L on genomic DNA extracted from necropsy tissue of 228 anonymous cases of SIDS (86 females, 142 males, average age = 3 + 2 months, range 6 hours to 12 months). Results: Three putative, SIDS-associated GPD1-L missense mutations, E83K, I124V, and R273C, were discovered in a three-month-old white male, a five-week-old white female, and a one-month-old white male, respectively. All mutations occurred in highly conserved residues and were absent in 600 reference alleles. Compared to wild type GPD1-L, GPD1-L mutations co-expressed with SCN5A in heterologous HEK cells produced a 60% reduction in peak sodium current density (p<0.004). Adenovirus-mediated gene transfer of the E83K-GPD1-L mutation into neonatal mouse myocytes markedly attenuated the sodium current as well (p<0.01). These decreases in current density are consistent with sodium channel loss-of-function diseases like BrS. Conclusion: This study is the first to report mutations in GPD1-L as a pathogenic cause for a small subset (~1%) of SIDS via a secondary loss-of-function mechanism whereby perturbations in GPD1-L precipitate a marked decrease in the peak sodium current and a potentially lethal, BrS-like pro-arrhythmic substrate.

Background: Long QT Syndrome (LQTS) genetic testing has been available clinically since August 2004. In cases of bona fide LQTS, the yield of genetic testing is approximately 75%. Recently (HRS 2007), Priori and colleagues have suggested that referral for LQTS genetic testing should be based on the screening QTc citing a yield of < 2% for cases referred to their research program with a QTc < 440 ms. Here, we analyze the effect of QTc on the yield in two of the largest assembled cohorts of unrelated patients (pts) referred for LQTS genetic testing. Methods: From August 1997 to July 2006, 1125 unrelated pts (717 females, average age at diagnosis 23.5 yrs, average QTc 473 ms) were referred to either the Mayo Clinic Windland Smith Rice Sudden Death Genomics Laboratory (N = 541) for research-based LQTS genetic testing or to PGxHealth (N = 574) for the FAMILION ® LQTS genetic test. Comprehensive open reading frame and splice site analysis (60 exons) for LQTS-causing mutations involving KCNQ1 (LQT1), KCNH2 (LQT2), SCN5A (LQT3), KCNE1 (LQT5), and KCNE2 (LQT6) was conducted by either dHPLC/DNA sequencing (Mayo), or by direct DNA sequencing (FAMILION). Results: The probability of a positive genetic test increased dramatically with increasing QTc, ranging from 5% for the 38 pts referred for genetic testing despite a QTc < 400 ms to 60% for the 168 pts with a QTc > 500 ms (p < 0.00001). Akin to the Priori cohort, the yield of the genetic test was similar among the subset of pts with a QTc < 470 ms (64% Priori, 61% Mayo cohort, and 53% FAMILION, p = NS). However, the yield for those pts with a QTc between 440 and 470 ms was much greater in the Mayo (49%) and FAMILION (36%) cohorts compared to the Priori cohort (14%, p < 0.0005). In contrast to the 2% yield observed in Italy, 22% of pts with QTc < 440 ms referred to Mayo Clinic and 14.5% of the FAMILION referrals had a positive genetic test (p < 0.001). Conclusions: Whereas only 2% of patients with a QTc < 440 ms had a positive genetic test in Italy, the yield was ten times greater among those referred to Mayo Clinic despite the same QTc cut-off. The cumulative index of suspicion for LQTS rather than a pre-specified QTc cut-off should guide clinical decision making as to the appropriate utilization of the genetic test.

A-Kinase Anchoring Protein 9 (AKAP9), or Yotiao, binds to the carboxy-terminus (C-T) of the IKs channel alpha subunit KCNQ1 and recruits cAMP-dependent protein kinase A (PKA) and protein phosphatase 1 (PP1) to the channel to control its phosphorylation state. Previously, an inherited LQT1-causing mutation in the KCNQ1 C-T/Yotiao binding domain was shown to disrupt the complex and render the channel functionally insensitive to cAMP-dependent regulation. Despite the critical role of Yotiao in mediating this fundamental physiological response, the large size of the Yotiao protein (over 200 kD) has hindered patient screens and, to date, no disease-linked mutations have been found in this adaptor protein. The objectives of the present study were first to locate reciprocal binding site(s) for KCNQ1 on Yotiao and to test whether mutation(s) in the binding module(s) might cause LQTS. Using GST fusion protein pull down and immunoprecipitation (IP) techniques, we identified two regions in Yotiao necessary for binding to KCNQ1: one within the Yotiao N-terminus (residues 29 – 46), and a second within its C-terminus (1574 –1643). Targeted sequence analysis was then carried out in fifty clinically diagnosed but genotype negative LQTS patients in the coding regions for these two KCNQ1 binding domains in AKAP9-encoded Yotiao. This analysis identified a novel missense mutation, S1570L-Yotiao, present in two siblings diagnosed with LQTS but absent in the normal population. We found that the S1570L mutation, located close to the Yotiao C-T binding motif for KCNQ1, disrupted the interaction between the two proteins and significantly reduced both basal and exogenous cAMP-induced KCNQ1 phosphorylation. Perforated patch experiments in CHO cells revealed that expression of S1570L-Yotiao significantly reduced basal IKS channel activity and completely ablated the response of the IKs channel to cAMP plus the phosphatase inhibitor okadaic acid. These reconstituted cellular consequences of the inherited S1570L-Yotiao mutation are consistent with delayed repolarization of the ventricular action potential observed in the affected siblings. Thus, we have demonstrated for the first time a link between an inherited AKAP mutation and human disease, LQTS.

Introduction : Type 1 catecholaminergic polymorphic ventricular tachycardia (CPVT1) is pursuant to mutations in the RyR2- encoded cardiac ryanodine receptor. Recently, we discovered nearly as many CPVT1-causing mutations as LQT3-causing mutations in a large cohort of unrelated LQTS referral patients. Here, we examine potential predictors of RyR2 genotype (CPVT1) among LQTS referral patients. Methods : Previously, comprehensive mutational analysis of the LQT1–10 susceptibility genes was completed for 541 unrelated patients (358 females, average age at diagnosis = 24 years, average QTc = 482 ms) referred for LQTS genetic testing. Using PCR, DHPLC, and DNA sequencing, we performed targeted analysis of RyR2 involving 44 exons (8–15, 37–49, 83–105) for the 260 unrelated referrals (172 females, average age at diagnosis = 25 years, average QTc = 470 ms) that remained “genotype-negative” following LQTS genetic testing. Results : Overall, 22 distinct RyR2 mutations (19 novel) were seen in 23 (8.8%, 12 females, average age at diagnosis = 22 ±14 years) unrelated “genotype-negative” LQTS individuals. Compared to patients with either genetically proven LQTS or the remaining genotype negative cohort, the RyR2 -positive cohort had significantly lower average QTc (419 ms vs 494 ms, p =2.4 ×10 −6 ; vs 472 ms, p = 0.005) and was more likely to experience either syncope (88% vs 66%, p = 0.06 vs 53%, p = 0.008) or cardiac arrest (56% vs 23%, p = 0.006 vs 17%, p = 0.001). Overall, 123 of the 541 LQTS referrals (23%) exhibited exertion-induced cardiac events (68 LQTS gene positive; 42 gene negative, and 13 RyR2-positive). Every RyR2 -positive case with an exertionally-triggered event had a QTc ≤ 460 ms, compared to only 17% of LQTS gene positive cases (p = 0.1 × 10 −8 ) and 24% of LQT1 cases. None of the exertion-induced LQTS gene positive cases had a QTc ≤ 420 ms. Conclusions: Failure to distinguish CPVT from LQTS can be a fatal mistake, as individuals with CPVT1 have a more severe phenotype and are much less protected by beta blocker therapy than those with LQT1. CPVT1, rather than concealed LQT1, should lead the differential diagnosis in the setting of syncope/cardiac arrest during exercise and a normal resting QTc.

Background: Bicuspid aortic valve disease (BAV), the most common congenital cardiovascular malformation, has an incidence of 0.5–1.0% of live births. While most cases of BAV appear to be sporadic, familial inheritance patterns have been observed consistent with autosomal dominant inheritance with variable penetrance. However, little is known about specific genetic loci responsible for familial BAV. Here, we performed linkage analysis on a large multi-generational pedigree affected with BAV. Methods: We identified a large, five-generation pedigree (136 family members) with 10 individuals having BAV. Two-dimensional echocardiography was used to assign aortic valve phenotype. Genome-wide linkage analysis using 430 microsatellite markers (Marshfield Clinic) and fine mapping using 100 single nucleotide polymorphisms (Affymetrix) on chromosome 9 was performed on genomic DNA from all available family members. Logarithm of odds (LOD) scores of >2.0 were considered suggestive of linkage. Comprehensive splice site/open reading frame mutational analysis of candidate genes residing in the putative locus is underway using PCR, DHPLC, and DNA sequencing. A candidate gene, KLF9, Krüppel-like factor 9 was analyzed for mutations because of its role in cardiogenesis. Results: Multi-point genome-wide linkage analysis demonstrated a 7 cM region on chromosome 9q21 that was suggestive of linkage for familial BAV with a maximum multipoint LOD score of 2.8 flanked by the microsatellite markers GATA7D12 and D9S1834. This region contains several candidate genes with biological plausibility for BAV phenotype. KLF9- encoded Krüppel-like factor 9, localized to chromosome 9q21, was targeted as a prime candidate gene for familial BAV. However, no mutations involving the translated exons of KLF9 were detected. Further fine mapping studies and candidate gene analysis are currently underway. Conclusions: We report a novel susceptibility locus on chromosome 9q21 for BAV in a large multi-generational family. Although coding region mutations in KLF9 are not responsible for BAV in this pedigree, several candidate genes with biological plausibility for the development of congenital BAV lie within this region and warrant further scrutiny.

Pathogenic mutations in the RYR2 -encoded cardiac ryanodine receptor cause type 1 catecholaminergic polymorphic ventricular tachycardia (CPVT1), a cardiac channelopathy with increased propensity for lethal ventricular dysrhythmias. Most RYR2 mutational analyses target three canonical domains encoded by < 40% of the translated exons. The extent of mutations localizing outside of these domains remains unknown as RYR2 has not been examined comprehensively in most patient cohorts. Mutational analysis of all 105 RYR2 exons was performed using PCR, DHPLC, and DNA sequencing on 108 unrelated patients (57% females, 96% white, age at diagnosis 18 ± 13 years, mean QTc 424 ± 26 ms) with either an explicit referral diagnosis of CPVT (N = 60) or an initial diagnosis of exercise-induced long QT syndrome (LQTS) but with QTc < 480 ms and a subsequent negative LQTS genetic test (N = 48). Two hundred healthy individuals from the Human Genetic Cell Repository were examined to assess allelic frequency for all non-synonymous variants detected. Thirty-eight (15 novel) possible CPVT1-associated mutations absent in 400 reference alleles, were detected in 44 unrelated patients (41%). Three cases (7%) had >1 RYR2 mutation. Besides the 25 exons known previously to contain causative mutations, eight new mutation-containing exons were identified: 10, 12, 13, 21, 26, 40, 42, and 48. Nearly two-thirds of the CPVT1-positive patients had mutations that localized to one of only 7 exons: 8, 14, 47, 90, 93, 100, and 101. In addition, 5 (2 novel) common non-synonymous single nucleotide polymorphisms were identified. This study represents one of the largest cohorts of patients for which RYR2 was examined in its entirety. Possible CPVT1 mutations in RYR2 were identified in approximately 41% of both CPVT referrals and LQTS gene-negative patients with exercise-induced syncope and QTc <480 ms. Including the eight new exons hosting mutations in this study, 33 of the 105 translated exons are now known to host possible mutations. Considering that two-thirds of CPVT1-positive cases would be discovered by selective analysis of <10 exons, a tiered targeting strategy for mutation discovery may afford a more cost-effective approach to CPVT genetic testing.

Long QT syndrome (LQTS) is a heterogeneous group of channelopathies characterized by increased risk of potentially lethal ventricular arrhythmias. LQT1, LQT2, and LQT3 comprise 95% of genetically proven cases and exhibit a number of established genotype-phenotype correlations. The study aimed at examining the phenotypes of genotype-negative LQTS, accounting for ~25% of LQTS cases. An IRB-approved retrospective analysis was conducted on 56 patients (39 female, 25 ± 17 years) who, after genetic testing either in our sudden death genomics laboratory or with the commercially available Familion test, were negative for mutations in the 3 principal LQTS-susceptibility genes ( KCNQ1, KCNH2, and SCN5A), and the minor genes underlying LQT5 and LQT6. All had been diagnosed with LQTS, with a clinical diagnostic score of ≥ 3.5 or QTc ≥ 480 ms. The mean diagnostic score was 4.4 (95% CI 4.2 – 4.7); mean QTc was 525 ms (95% CI 508 – 543 ms). Two-thirds were symptomatic (syncope, cardiac arrest, and/or seizures) with exercise-triggered events in 10 (26%). Twenty-one (38%) had a family history of sudden cardiac arrest. ECG showed a T wave pattern suggestive of LQT1 in 32%, LQT2 in 43%, and LQT3 in 18%. In those with exercise-induced symptoms, the ECG was LQT2-like in 50% and LQT1-like in 30%. One patient had post-partum syncope with an LQT2-like ECG. None had an auditory trigger, but 3 patients, all with an LQT2-like ECG, had a family history of auditory-triggered events. One-third of the patients had received an ICD, 58% as secondary prevention. Over 2/3 were on beta-blockers. Among the 45 patients so far tested for mutations in minor LQTS-susceptibility genes, 2 had LQTS-causing mutations in ANKB (LQT4), 1 in SCN4B (LQT10), 1 in AKAP9 (LQT11) and 2 in SNTA1 (LQT12). Genotype-negative patients with a firm LQTS diagnosis show marked phenotypic heterogeneity, suggesting multiple underlying pathogenic pathways. Only a few patients have LQTS-causing mutations in minor genes after complete LQT1–12 genetic testing. Classifying genotype negative patients into LQT1-, LQT2-, or LQT3-like profiles may guide the discovery of novel genes encoding channel interacting proteins corresponding to those specific signaling pathways.

An estimated 10–15% of sudden infant death syndrome (SIDS) may be channelopathic with approximately half of the SIDS-associated channel mutations involving the sodium channel (NaV1.5) macromolecular complex. Recently, SNTA1, which encodes the sodium channel interacting protein (ChIP), syntrophin alpha, was discovered to be a novel long QT syndrome (LQTS) susceptibility gene (LQT12) with mutations causing increased late sodium current secondary to increased s-nitrosylation of the sodium channel’s alpha subunit. Therefore, SNTA1 is a plausible candidate gene for SIDS. In this IRB-approved study, mutational analysis of SNTA1’s eight translated exons was performed using PCR, DHPLC, and direct DNA sequencing of DNA derived from 286 SIDS cases (112 females, 197 white, average age 2.9 ± 1.9 months, range 3 days -12 months). The anonymized study design precluded a determination of mutation status as familial or sporadic. Overall, 8/286 cases of SIDS harbored missense mutations in SNTA1 that were absent in 800 reference alleles. Six distinct, rare SNTA1 missense mutations were detected in 7/112 (6.25%) female infants but only 1/174 (0.6%) male infants (p < 0.01). Comprehensive open reading frame analysis of all 800 reference alleles also revealed rare genetic variants in 1% of the control subjects. A common pair of non-synonymous single nucleotide polymorphisms (P74L/A257G), with a heterozygous frequency of 1%, was identified similarly in both cases and controls. This study provides the first molecular evidence implicating mutations in SNTA1 as a novel pathogenic substrate for channelopathic SIDS. However, given the ~1% background rate of rare variants in controls, mechanistic studies of each putative SIDS-associated SNTA1 mutation will be necessary to confirm that the mutant ChIP yields a functionally perturbed NaV1.5 macromolecular complex akin to the investigations that established SNTA1 as a novel LQTS-susceptibility gene.

Cardiac I Kr is a critical repolarizing current in the heart and a target for inherited and acquired long QT syndrome. Biochemical studies show that native I Kr channels are heteromers composed of both hERG 1a and 1b subunits, yet our current understanding of I Kr functional properties derives primarily from studies of homo-oligomers of the original hERG 1a isolate. The hERG 1a and 1b subunits are identical except at the amino (NH2) terminus, which in hERG 1b is much shorter and has a unique primary sequence. We compared the biophysical properties of currents produced by hERG 1a and 1a/1b channels expressed in HEK-293 cells at near-physiological temperatures. We found that heteromeric hERG 1a/1b currents are much larger than hERG 1a currents and conduct 80% more charge during an action potential. This surprising difference corresponds to a two-fold increase in the apparent rates of activation and recovery from inactivation, which reduces rectification and facilitates current rebound during repolarization. Kinetic modeling shows these gating differences account quantitatively for the differences in current amplitude between the two channel types. Depending on the action potential model used, loss of 1b predicts an increase in action potential duration of 27 ms (7%) or 41 ms (17%), respectively. Drug sensitivity was also different. Compared to homomeric 1a channels, heteromeric 1a/1b channels were inhibited by E-4031 with a slower time course and a corresponding four-fold positive shift in the IC 50 . Differences in current kinetics and drug sensitivity were modeled by “NH2 mode” gating with conformational states bound by the amino terminus in hERG 1a homomers but not 1a/1b heteromers. The importance of hERG 1b in vivo is supported by the identification of a 1b-specific A8V missense mutation in 1/269 unrelated genotype-negative LQTS patients and absent in 400 control alleles. Mutant 1bA8V expressed alone or with hERG 1a in HEK-293 cells nearly eliminated 1b protein. Thus, mutations specifically disrupting hERG 1b function are expected to reduce cardiac I Kr , prolong QT interval and enhance drug sensitivity, thus representing a potential mechanism underlying inherited or acquired LQTS.