Author + information
- Received December 22, 2015
- Revision received November 18, 2016
- Accepted November 22, 2016
- Published online July 17, 2017.
- Dan Hu, MD, PhDa,b,∗ (, )
- Yang Li, MD, PhDc,
- Jiancheng Zhang, MD, PhDd,
- Ryan Pfeiffer, BSb,
- Michael H. Gollob, MDe,
- Jeff Healey, MDf,
- Daniel Toshio Harrell, MDg,
- Naomasa Makita, MD, PhDg,
- Haruhiko Abe, MD, PhDh,
- Yaxun Sun, MDi,
- Jihong Guo, MDj,
- Li Zhang, MDk,
- Ganxin Yan, MDk,
- Douglas Mah, MDl,
- Edward P. Walsh, MDl,
- Harris B. Leopold, MDm,
- Carla Giustetto, MDn,
- Fiorenzo Gaita, MDn,
- Agnieszka Zienciuk-Krajka, MDo,
- Andrea Mazzanti, MDp,
- Silvia G. Priori, MD, PhDp,q,
- Charles Antzelevitch, PhDk and
- Hector Barajas-Martinez, PhDb,† ()
- aDepartment of Cardiology and Cardiovascular Research Institute, Renmin Hospital of Wuhan University, Wuhan, China
- bMolecular Genetics Department, Masonic Medical Research Laboratory, Utica, New York
- cInstitute of Geriatric Cardiology, Chinese PLA General Hospital, Beijing, China
- dDepartment of Cardiology, Provincial Clinical Medicine College of Fujian Medical University, Fujian, China
- eDepartment of Cardiology, Toronto General Hospital, University of Toronto, Toronto, ON, Canada
- fPopulation Health Research Institute, McMaster University, Hamilton, ON, Canada
- gDepartment of Molecular Physiology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan
- hDepartment of Heart Rhythm Management, University of Occupational and Environmental Health, Fukuoka, Japan
- iSir Run Shaw Hospital, Zhejiang University, Hangzhou, Zhejiang, China
- jDepartment of Cardiac Electrophysiology, Division of Cardiology, People’s Hospital, Peking University, Beijing, China
- kLankenau Institute for Medical Research and Jefferson Medical College, Philadelphia, Pennsylvania
- lCardiac Electrophysiology Division, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts
- mConnecticut Children’s Medical Center, University of Connecticut School of Medicine, Hartford, Connecticut
- nDivision of Cardiology, University of Torino, Department of Medical Sciences, “Città della Salute e della Scienza” Hospital, Torino, Italy
- oDepartment of Cardiology and Electrotherapy, Medical University of Gdansk, Gdansk, Poland
- pMolecular Cardiology, IRCCS Salvatore Maugeri Foundation, Pavia, Italy
- qDepartment of Molecular Medicine, University of Pavia, Pavia, Italy
Objectives This study sought to evaluate the phenotypic and functional expression of an apparent hotspot mutation associated with short QT syndrome (SQTS).
Background SQTS is a rare channelopathy associated with a high risk of life-threatening arrhythmias and sudden cardiac death (SCD).
Methods Probands diagnosed with SQTS and their family members were evaluated clinically and genetically. KCNH2 wild-type (WT) and mutant genes were transiently expressed in HEK293 cells, and currents were recorded using whole-cell patch clamp and action potential (AP) clamp techniques.
Results KCNH2-T618I was identified in 18 members of 7 unrelated families (10 men; median age: 24.0 years). All carriers showed 100% penetrance with variable expressivity. Eighteen members in 7 families had SCD. The average QTc intervals of probands and all carriers was 294.1 ± 23.8 ms and 313.2 ± 23.8 ms, respectively. Seven carriers received an implantable cardioverter-defibrillator. Quinidine with adequate plasma levels was effective in prolonging QTc intervals among 5 cases, but 3 cases still had premature ventricular contraction or nonsustained ventricular tachycardia. Bepridil successfully prevented drug-refractory ventricular fibrillation in 1 case with 19-ms prolongation of the QTc interval. Functional studies with KCNE2 revealed a significant increase of IKr (rapidly activating delayed rectifier potassium channel) tail-current density in homozygous (119.0%) and heterozygous (74.6%) expression compared with WT. AP clamp recordings showed IKr was larger, and peak repolarizing current occurred earlier in mutant versus WT channels.
Conclusions We reported the clinical characteristics and biophysical properties of the highly frequent mutation that contributes to genetically identified SQTS probands. These findings extend our understanding of the spectrum of KCNH2 channel defects in SQTS.
Short QT syndrome (SQTS) is a rare genetic disease characterized by an abnormally short QT interval in subjects with structurally normal hearts. It is a recognized cause of cardiac rhythm disorders, including both atrial and ventricular arrhythmias, and sudden cardiac death (SCD) (1–4). As an inherited channelopathy, the molecular basis for SQTS has been associated with mutations in 6 genes: KCNH2 (rapidly activating delayed rectifier potassium channel [IKr], SQTS1), KCNQ1 (slowly activating delayed rectifier potassium channel [IKs], SQTS2), and KCNJ2 (inward rectifier potassium channel [IK1], SQTS3), which encode different potassium channels; and CACNA1C, CACNB2b, and CACNA2D1 (SQTS4-6), which encode the L-type calcium channel (ICa) (5). Loss-of-function of ICa is reported to induce a combined Brugada-SQTS phenotype (6–11). When expressed in heterologous expression systems, SQTS mutations display an increase in the potassium currents involved or a decrease in the calcium current, which results in acceleration of the repolarization process and an abnormal abbreviation of the QT interval. The precise QTc cutoff that defines SQTS remains controversial, similar to long QT syndrome. To improve a diagnostic approach, Gollob et al. (12) proposed a diagnostic scorecard for SQTS that incorporates both clinical and genetic criteria. Later, the modified Gollob score was applied in some studies as a guide to risk stratification (4). To date, few mutations have been identified to cause SQTS. Attempts to perform genotype−phenotype correlation are hampered by the small numbers of mutation carriers. Herein, we report the data from the largest cohort of SQTS patients carrying the hotspot mutation T618I in KCNH2, and provide genotype—phenotype correlation on 18 mutation carriers from 7 unrelated families. The clinical and functional characteristics and electrophysiological characteristics of this hotspot mutation are described.
Clinical investigation and follow-up
SQTS index cases were defined as those with a QTc interval ≤340 ms, even if they were asymptomatic. Those with QTc intervals between 340 and 360 ms, associated with ≥1 of the following, were also considered to be cases of SQTS: 1) a confirmed pathogenic mutation; 2) a family history of SQTS; 3) a family history of sudden death at age younger than 40 years; and 4) survival after a ventricular tachycardia/ventricular fibrillation (VT/VF) episode in the absence of heart disease (13). Structural heart disease was excluded by echocardiography and/or magnetic resonance imaging. The final study group from the North America, Europe, and Asia, included 7 probands (4 men, 57.1%; median age: 30.0 years; interquartile range [IQR]: 27.0 years; 9.0 to 46.0 years; 25th and 75th quartiles: 16.0 and 43.0 years) and their families. There were 18 cases of SCD and/or sudden death and 2 aborted SCDs in the families. Limited aspects of 4 families (families 4 to 7) were previously reported (14–17).
Demographic data of probands and their family members were collected. In all patients, at least 1 12-lead electrocardiogram (ECG) was obtained at a stable heart rate. ECG parameters, including RR, PR, QRS, QT, QTc, J-Tpeak data, and early repolarization (ER), were manually measured. The QT interval was measured using the tangent method, and the QTc interval was calculated using Bazett’s formula. The J point was defined as the end of the QRS interval and the beginning of the ST-segment. ER was defined as an elevation of >0.1 mV of the J point from baseline. Tpeak was measured at the highest point of the T-wave. The U-wave was best visualized in the precordial lead that had the highest amplitude (leads V2 to V4). The Gollob Diagnostic Score (12) was calculated in those cases in which all parameters needed were available. Long-term follow-up (72.7 ± 24.2 months) was available for all 7 families.
Molecular genetic analysis
Informed consent was obtained, and blood samples were collected with institutional review board approvals. Genomic DNA was extracted from peripheral blood leukocytes using a commercial kit (Gentra System, Puregene, Valencia, California). All exons and intron borders of cardiac ion channel genes, including all isoforms of 6 known short QT candidate genes (KCNH2, KCNQ1, KCNJ2, CACNA1C, CACNB2b, CACNA2D1), were amplified and analyzed by direct sequencing from both directions using an ABI PRISM 3100-Avant Automatic DNA sequencer (Applied Biosystems, Foster City, California). Genomic DNA from 430 ethnically-matched healthy reference alleles was used as the control group. The KCNH2 primers of exon 7 used for screening are 5′-CTTGCCCCATCAACGGAATG-3′ (sense) and 5′-CTAGCAGCCTCAGTTTCCTC-3′ (antisense).
Mutagenesis and transfection in HEK 293 cells
The KCNH2-T618I mutation was engineered into wild-type (KCNH2-WT) cDNA cloned in pcDNA3.1 (Invitrogen, Carlsbad, California) by overlap extension using mutation-specific primers and a Quick Change Site-Directed Mutagenesis Kit (Stratagene, La Jolla, California). The presence of mutations was confirmed by sequence analysis. HEK293 cells were transfected with KCNH2 cDNA (WT or T618I, respectively) using Lipofectamine (Life Technologies, Frederick, Maryland). In another set of experiments, KCNE2-WT was co-expressed with WT and/or T618I. CD8 cDNA was co-transfected as a reporter gene. CD8-positive cells identified by Dynabeads (M-450 CD8, Dynal, Grand Island, NY) were studied 48 to 72 h after transfection.
Electrophysiological study and analysis
Membrane currents were measured using whole-cell, patch-clamp procedures with Axopatch 700B amplifiers (Axon Instruments Foster City, California). Patch electrodes were pulled from borosilicate glass (Hilgenberg, Malsfeld, Hesse, Germany) on a Sutter P-97 horizontal puller (Sutter Instruments, Novato, California). All signals were acquired at 0.33 kHz (Digidata 1440A, Axon Instruments) and analyzed with pCLAMP version 10.2 (Axon Instruments). Series resistance and capacitive transients were compensated using standard techniques. Membrane currents were low-pass filtered at 10 kHz and digitized at 100 kHz. Recordings were made at room temperature using an internal solution containing (in millimoles per liter) 140 Potassium Aspartate, 10 ethylene glycol tetraaceticcid, 4 magnesium adenosine triphosphate, 1 magnesium chloride, 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, (pH: 7.2 with KOH). External solution contained (in millimoles per liter) 140 sodium chloride, 1 calcium chloride, 1 magnesium chloride, 4 potassium chloride, 10 HEPES, and 5 glucose (pH: 7.4 with NaOH). To ensure reproducibility, after whole-cell conditions were established by rupturing the cell membrane, we allowed a dialysis period of 4 minutes before beginning measurements. During the dialysis period, we monitored current−voltage relationships to ensure stability and consistency of recordings.
Data are presented as percentages, mean ± SD, median with IQR for clinical values, and mean ± SEM for experimental values. A p value <0.05 was considered statistically significant according to Student’s t-test, analysis of variance analysis, or the univariate generalized linear model for continuous variables; chi-square, Fisher’s exact, or Mantel-Haenszel tests were used for categorical data, as appropriate. pClamp 10.2 (Axon Instruments), Excel (Microsoft, Redmond, Washington), and Origin (Microcal Software, Wellesley Hills, Massachusetts) software were used for data acquisition and analysis.
Clinical and demographic profiles of probands and their families
The KCNH2-T618I mutation was identified in 18 cases of 7 unrelated families (10 men, 55.6%; median age: 24.0 years; IQR: 23.0 years; ranging from 7.0 to 46.0 years). All carriers were clinically diagnosed with SQTS in the absence of an identifiable etiology and showed 100% penetrance with variable expressivity (Figure 1, Table 1). A flat QT-Heart Rate relationship was observed in all 8 mutation carriers (families 3, 5, 6, and 7) and absent in healthy relatives who underwent exercise tests. Combined information in our database with reported results suggested that the proportions of KCNH2-T618I among genetically identified SQTS patients and probands were 29.0% and 25.9%, respectively. The other apparently highly frequent mutation of SQTS is KCNH2-N588K (Table 2, Figure 9A, Online Table 1).
In all, there was a high incidence of SCD (18 members in 7 families; average age of death: 29.1 ± 19.3 years, with median age of death of 34.0 years) and aborted SCD (probands in family 2 [II-2] and family 7 [III-3]). The average QTc intervals of probands and all mutant carriers were 294.1 ± 23.8 ms and 313.2 ± 23.8 ms, respectively (range: 260 to 344 ms), which were significantly shorter than the QTc intervals of 12 mutation-negative, first-degree relatives (426.2 ± 26.4 ms; p < 0.05). ECGs typically showed tall, peaked, and symmetrical T waves, preceded by a short or absent ST-segment in most affected subjects. There was no gender preference among mutant carriers. Despite the tendency of longer QTc intervals in female carriers (323.7 ± 16.3 ms vs. 304.8 ± 26.2 ms in females and males; p = 0.095), females were more prone to be symptomatic (62.5% [n = 5] vs. 10.0% [n = 1]; p < 0.05], which was not observed in N588K carriers (symptomatic, 66.7% [n = 6] vs. 85.7% [n = 6] in females and males; p > 0.05). Similar to previous observations (4), there was no statistical difference in QTc intervals between symptomatic and asymptomatic cases (310.9 ± 27.0 ms vs. 314.3 ± 23.3 ms; p > 0.05). There was also no difference in age at diagnosis between probands and affected family members (29.4 ± 14.1 years vs. 25.3 ± 11.3 years; p > 0.05), but differences were observed with respect to QTc intervals (294.1 ± 23.8 ms vs. 325.3 ± 13.9 ms; p < 0.01) (Table 3). Arrhythmias were common in symptomatic carriers with KCNH2-T618I, including ventricular ectopic beats, VT, and VF. In contrast to a higher incidence of atrial fibrillation (AF) in KCNH2-N588K (60.0% for probands and 45.5% for affected relatives), none of the KCNH2-T618I carriers experienced AF (0.0% for both groups). Among all 18 KCNH2-T618I carriers, 50.0% presented with a minor ER pattern in leads V2 to V4. A distinct U-wave was evident in the precordial leads (most prominently in leads V2 to V4) in 72.2% of cases. Demographic and ECG characteristics of the probands are shown in Table 1 and Figure 2. The Gollob Diagnostic Score for SQTS ranged from 7 to 11 for all mutation carriers, indicating a high sensitivity and clinical accuracy of the score for cases independently diagnosed as SQTS.
Treatment and follow-up
During 72.7 ± 24.2 months of follow-up, an ICD was implanted in 7 patients: 3 probands and 1 family member were implanted after presenting with symptoms (such as unexplained syncope, documented VT/VF); and 2 asymptomatic probands and 1 asymptomatic family member for primary prophylaxis. The proband of family 2 and the other 2 family carriers only experienced occasional ventricular premature beats during follow-up. Proband 5 experienced 5 inappropriate shocks 10 days after ICD implantation (Epic 296, St. Jude Inc., St. Paul, Minnesota) due to T-wave oversensing (sinus rhythm at 106 beats/min, which the ICD recognized as VF of 212 beats/min). After programming to decrease R-wave sensing threshold, no further T-wave oversensing or ICD shocks were recorded for 8.5 years. Recently, the patient developed frequent short-coupled pre-mature ventricular contractions from the right ventricular outflow tract (16,000/day) with symptoms. Proband 6 received 1 inappropriate shock during fast sinus rhythm (200 beats/min), which triggered ICD shock (VR V-193, Atlars, St. Jude Inc., St. Paul, Minnesota). One member of family 7 with an ICD and 2 other carriers with loop recorders only exhibited nonsustained VT during follow-up.
The proband of family 3 and her son were treated with quinidine (900 mg/day). The increments of the QTc intervals after medication were 65 and 90 ms (Figures 3A and 3B), and the effective refractory period (ERP) was also normalized during the electrophysiological study. Quinidine prolonged, but failed to completely normalize the QTc interval in proband 7 (333 ms after 1,000 mg). However the ventricular ectopy resolved (12% of the total heartbeats before therapy). It also showed reduced efficacy in the other 2 carriers of family 7, because nonsustained VT was observed during follow-up in both, although the QTc interval was normalized by the drug (Table 4). Quinidine will be administered to the proband of family 5 soon.
Sotalol showed poor clinical efficacy in all 3 carriers who received it, with unchanged or even reduced QTc intervals. The proband of family 4 experienced spontaneous VF, and the implanted ICD appropriately terminated the arrhythmia 3 times. Bisoprolol (maximum 5 mg/day) failed to prevent VF recurrence. Oral bepridil (150 mg/day) was then administered to the patient after the last VF episode. The QTc interval was slightly prolonged (341 ms), and no VF was observed for 32 months (Figure 3C, Table 4).
Molecular genetic analysis
Polymerase chain reaction−based sequencing analysis revealed a double peak in the sequence of the KCNH2 gene among all SQTS subjects, but not in relatives with a normal ECG. The change consisted of a C-to-T transition at nucleotide 1853 (c.1853 C>T), which predicts a substitution of isoleucine for threonine at residue 618 (p.Thr618Ile, T618I, rs199472947) of Kv11.1 (Figure 4A). This nucleotide change was observed in 0 of 200 healthy control subjects (400 alleles; frequency, 0.0%) in our database; a frequency of 0.0% was reported in the 1,000 genome project database; and 0.0% in Exome Sequencing Project. It was predicted to be damaging and probably damaging using the Sift (18) and PolyPhen (19) tools.
Alignment of the amino acid sequence of Kv11.1 proteins showed that threonine at position 618 is highly conserved among different species (Figure 4B). Residue T618 is located at the pore helix (intramembrane) region of the Kv11.1 channel (Figure 4C). Figure 4D presents a plot of QTc values for all known KCNH2 mutations in SQT1 carriers. N588K and T618I, both in the pore region, are associated with briefer QTc intervals and a higher frequency of carriers, whereas mutations in other regions (E50D in N-terminal and R1135H in C-terminal) have been reported as much longer QTc intervals in only single individuals to date, which suggests that the KCNH2 P-loop may be the critical region for a more extreme SQTS phenotype.
All T618I mutation carriers in family 2 also carried a KCNH2-R1047L rare polymorphism. Both the proband and positive carriers in family 3 and 7 also had KCNH2-K897T polymorphism (Table 1, Online Table 2). These single nucleotide polymorphisms have been reported to exert a modifying effect to prolong QTc intervals (20,21).
Figures 5A and 5B present current traces for WT (top panel) and T618I mutant (lower panel) KCNH2 channels expressed in HEK293 cells. The current−voltage relationship revealed that the step current density of T618I significantly increased at more positive potentials than 0 mV of the WT (Figure 5C). Tail current densities of at −40 and −110 mV were larger for T618I versus WT (Figures 5D and 5E). Current densities of T618I and WT at test potentials of −40 mV after depolarizing from +50 mV were 207.5 ± 25.8 and 122.7 ± 12.2 pA/pF (n = 10; p < 0.01). Current densities of T618I and WT at −110 mV after depolarizing from +50 mV were −267.4 ± 33.2 pA/pF versus −164.5 ± 27.6 pA/pF (n = 10; p < 0.05) (Figures 5F and 5G).
The envelope test of KCNH2 tail current was applied to measure the rates of the channel activation. The protocol used and original WT and T618I current traces are shown in Figure 6A. Figures 6B and 6C showed the time dependence of fractional tails fitted with a single exponential association. Activation was considerably faster for T618I than for WT (Tau of T618I and WT: 121.58 ± 10.5 vs. 240.98 ± 7.16; n = 11, respectively; p = 0.0156). The voltage dependence of current activation was assessed using standard tail current analysis. The steady-state V1/2 of activation amounted to −29.59 ± 3.12 mV and −13.13 ± 2.18 mV of T618I and WT channels, respectively (n = 10; p < 0.05), with similar slope factors (k: 10.4 ± 1.16 vs. 10.92 ± 2.18; p > 0.05) (Figure 6D). The T618I mutation caused a negative shift in the voltage dependence of channel activation and increased activation availability of the KCNH2 channel, which accounted for the increase in current densities.
To determine how the mutation altered the kinetics of the current during an action potential (AP), we elicited T618I and WT currents by a stimulus generated by a previously recorded AP. WT current displayed a waveform with slow activation kinetics and a rapid increase at the phase 3 repolarization due to the rapid recovery of inactivated channels. In sharp contrast, T618I current displayed fast activation kinetics and a rapid increase during the early phase of AP repolarization, thus indicating a more important contribution to repolarization during the early phases (Figure 6E).
The inhibitory effects of 10 nM dofetilide on the KCNH2 channel with a repolarizing pulse of −40 mV are shown in Figure 7. At a test potential of +50 mV, the WT tail current was reduced by 46.2% after exposure to 10 nM dofetilide, whereas T618I tail currents were only reduced by 27.9% after the same treatment (p < 0.05).
Figure 8A shows current traces for WT, T618I, and WT+T618I co-expressed with KCNE2-WT in HEK293 cells. Similar to Figure 5C, the current−voltage relationship indicated that step current densities of T618I and WT+T618I were significantly greater than WT at a potential of 0 mV (Figure 8B). Tail current density recorded at −40 mV after depolarizing from +10 mV was larger for T618I and WT+T618I versus WT (Figure 8C). Current densities of T618I, WT+T618I, and WT recorded at a test potential of −40 mV following a step to +10 mV were 177.8 ± 27.8, 141.8 ± 25.8 and 81.2 ± 12.6 pA/pF (n = 6 to 8 for each condition, respectively; p < 0.01 vs. WT).
In 1993, a 2-year follow-up study reported that subjects with an abbreviated mean QTc interval over 24 h (<400 ms) had as high a risk of SCD (∼2.4-fold increased risk), similar to that of subject with long QT intervals (>440 ms, ∼2.4-fold increased risk) (22). However, it was not until 2000, that Gussak et al. (1) suggested a new inherited arrhythmia syndrome characterized by abbreviated QT intervals, AF and SCD, which they termed SQTS. In the present study, we reported on the largest cohort of SQTS patients carrying the hotspot mutation T618I in KCNH2. We presented data collected over an average of 5 years, which provided new insights into the natural history, genetics, and response to therapy of these SQT1 patients.
Demographics and ECG characteristics
Hotspots are not simply the highest frequency of mutations. Nucleotide positions with an exceptionally high mutation frequency, which also has ≥95% probability, are called “hotspots” (23). Table 2 lists the genetic basis for all 27 SQTS probands with genetic mutation. Five probands carried KCNH2-N588K (18.5%), and 7 had KCNH2-T618I (25.9%). Familial clustering of the SQTS phenotype was present in all of the SQTS probands with KCNH2-T618I, similar to probands identified with KCNH2-N588K, which indicated 100% inherited probability. Within the familial cases, all family members with a high probability of the Gollob diagnostic score were subsequently determined to be KCNH2-T618I gene-positive, which is consistent with an autosomal dominant pattern and 100% phenotype penetrance. This percentage was higher than a previous report (∼50%) of SQTS families (4). KCNH2-T618I was discovered in unrelated SQTS families from North America, Europe, and Asia (Table 1, Figure 9B), which suggests the mutation did not arise from a recent “founder effect.” Our data indicated that KCNH2-T618I is a hotspot mutation in SQTS.
The QTc intervals of healthy males is known to be shorter than that of females (24,25), and similar sex distinctions were observed in the present study. Contrary to previous reports of overwhelming predominance of males in the phenotype of 73 SQTS (91.0%) (4), KCNH2-T618I carriers were fairly evenly distributed (55.6% in males). Our data were consistent with previous findings, that no clear sex difference was observed in KCNH2-N588K carriers (43.8% in males), or in all 37 carriers with SQTS1 (54.1% in males). Interestingly, the symptoms appear to be more severe in KCNH2-T618I females (62.5% [n = 5] vs. 10.0% [n = 1]; p < 0.05). In general, the risk of cardiac events and lethal arrhythmias in affected SQTS1 females was not less than males. Similar to other SQTS cohorts, symptoms in the KCNH2-T618I carriers included palpitations, syncope, and SCD across the family.
A family history of SCD was commonly reported in our KCNH2-T618I families; however, only a small portion of living carriers screened were symptomatic. This might suggest that lethal cardiac events and SCD are common in KCNH2-T618I families, and could be the first clinical symptom. In comparison, 75.0% of family carriers with the KCNH2-N588K mutation were reported to be symptomatic, which could be explained by the slightly but significantly longer QTc intervals in KCNH2-T618I family carriers than those with KCNH2-N588K mutations (Table 3). The longer QTc interval could be due to different genetic backgrounds and environmental modifiers in those cases. Some genetic variants (K897T and R1047L) were previously reported to be associated with drug-induced or acquired long QT syndrome, and some were shown to reduce IKr (26,27). These QT interval modifiers might also contribute to atypical ECG morphology in several cases (23).
SQTS1 is generally characterized by the appearance of an abbreviated QTc interval, which is often associated with high amplitude and symmetrical T waves (7). The Tpeak to Tend interval (Tp-e) reflects spatial, including transmural, dispersion of repolarization. In our study, the Tp-e interval was relatively prolonged compared with other SQTS cases. Poor rate adaption of the QT interval was also common in our KCNH2-T618I cases (Figure 1). These factors were suggested to produce a substrate for re-entry that leads to VF (Online Figure 1) (28).
As expected, a distinct U-wave was commonly observed in our cohort. Schimpf et al. (29) provided evidence for dissociation between ventricular repolarization and the end of mechanical systole in SQTS patients. Coincidence of the U-wave with termination of mechanical systole provides support for the mechano-electrical hypothesis for the origin of the U-wave. This concept was also supported by a recent report by Araki et al. (30).
Watanabe et al. (31) reported a high prevalence of ER in patients with SQTS associated with arrhythmic events (65%) compared with a short QT control cohort (30%) or a normal QT control cohort (10%). In the EuroShort Registry, this percentage was 33% (32). In the present study, 50.0% of KCNH2-T618I SQTS carriers presented with an ER pattern. AF incidence was higher in subjects with a short QT interval or SQTS than that in the general population (1,3,9,33). Electrophysiological studies in these cases were characterized by significant shortening of atrial and ventricular refractory periods, and inducibility of both atrial and ventricular arrhythmias (34). Although AF occurred in 50% of SQTS patients with KCNH2-N588K, and might occur in up to 20% of total SQTS patients (12,35), this arrhythmia has not yet been observed in KCNH2-T618I mutation carriers. The significant diversity among these studies might stem from different modifier genes or simply be due to the unique effects of the respective mutations.
Therapy of SQTS patients with KCNH2-T618I
Therapy of SQTS has met with several challenges. Patients with the KCNH2-N588K mutation are known to be resistant to traditional IKr blockers (e.g., such as sotalol, dofetilide, flecainide, and ibutilide) that have a higher affinity for the inactivated state of the IKr channel, which has been lost due to the effect of the mutation (36,37). Quinidine, presumably due to its interaction with the activated state of the channel, was shown to be effective in prolonging the QT interval and preventing inducibility of VT/VF in patients with SQTS, including KCNH2-N588K carriers (38,39). Disopyramide might be a suitable alternative (40,41). Intravenous administration of nifekalant is useful for rapid restoration of extremely short QT intervals to normal ranges, such as during frequent discharge of ICDs due to incessant VF (41).
In the present study, we provided further information for the effectiveness of quinidine in SQTS cases. Quinidine administration prolonged QTc intervals in all 5 KCNH2-T618I cases who received this drug, but only suppressed arrhythmia in 3 of them during follow-up (Figure 9C). Although previous in vitro research indicated that sotalol (in high dosages) might be effective for T618I-linked SQTS1 patients (15,42), our results showed the ineffectiveness of sotalol and bisoprolol in KCNH2-T618I carriers. Our present experimental study also indicated the poor effect of another class III drug, dofetilide. Bepridil (a class IV antiarrhythmic drug with potassium- and sodium-blocking properties) effectively terminated drug-refractory VF in 1 carrier in our study, providing the new potential alternative in SQTS. This drug was reported as effective in preventing VF, including electric storms, in several long-term reports with patients with Brugada syndrome (43,44). However, the clinical efficacy of quinidine and bepridil are still under investigation.
Implantation of an ICD is first-line therapy for high-risk patients. However, long-term follow-up of patients with SQTS indicated that 58% of patients who received ICDs had device-related complications, especially inappropriate shocks secondary to T-wave oversensing in SQTS1 (45). We encountered 1 case of this in our cohort that was corrected by re-programming ventricular sensitivity for VF detection. In such patients, careful assessment of the QT interval and T-wave amplitude is indispensable to avoid the problem.
Genetic analysis and biophysical features of KCNH2-T618I mutation
Since our group identified the first SQTS mutation in 2004 (KCNH2-N588K) (7), mutations in 6 genes have now been associated with SQTS. The genes responsible for SQT1-4 are the same genes responsible for the congenital Long QT syndrome (LQTS) 2, 1, 7, and 8. In SQTS1, KCNH2-N588K results in reduced inactivation and greater current flow during the plateau potentials of the cardiac AP (5). Hence, the ventricular and atrial APs have a shorter duration, and therefore, a shorter QT on the ECG. The other 3 KCNH2 mutations associated with SQTS are I560T, R1135H, and E50D. The patient with R1135H also demonstrated a type 1 Brugada ECG pattern. Functional studies of the mutant channels consistently showed prolonged deactivation time constants compared with WT channels, without significant changes in any other gating parameters (46). The I560T mutant showed an increase in peak current density, and a positive shift of the inactivation curve (14). Our unpublished data on E50D showed a gain of function of IKr tail density, slower deactivation, and a modest positive shift in the voltage dependence of inactivation. Our result with KCNH2-T618I showed a significant gain-of-function in IKr, particularly in the tail density, which plays an important role in phases 2 and 3 of ventricular APs. Then, we confirmed KCNH2-T618I had fast activation kinetics, and a rapid increase using AP waveforms. Similar effects of KCNH2-T618I were reported at room or body temperature (15,42). Compared with a previously reported 6.0-fold increase of IKr of KCNH2-T618I mutation on the peak steady density (15), the present study displayed a 6.25-fold increase of IKr under similar in vitro conditions. In addition, our study showed that the gain-of-function was also caused by a negative shift in the voltage dependence of activation, which allowed more potassium channels to open. Moreover, the present study first showed that the heterozygously expressed KCNH2-T618I and KCNH2-WT co-expressed with KCNE2 also resulted in a significant gain-of-function of IKr.
Study limitation and future scope
Although it was the largest collection of SQTS carriers with the same mutation thus far reported, the sample size of our study was necessarily small because of the rarity of the syndrome. Further long-term follow-up of patients are needed to clarify the prognosis of KCNH2-T618I mutation carriers and the most efficacious drug approaches to therapy.
Summary and Conclusions
KCNH2-T618I is the highly frequent mutation associated with SQTS worldwide, with a high incidence of VT/VF or SCD, and complete penetrance. Female carriers appear more prone to symptoms and cardiac events. Together with the other highly frequent hotspot (KCNH2-N588K), it accounts for 85.0% of SQT1 and 54.8% of all genetically identified SQTS cases. ICD implantation is still the first-line therapy for KCNH2-T618I patients, although the high rates of inappropriate shock and complication have also been encountered. Our study demonstrated that quinidine is effective in prolonging QTc, but whether this translated into decreased SCD is still unknown. Bepridil exerted an ameliorative effect in 1 case, suggesting it might be the new alternative to prevent VT/VF, which has not been reported previously. The mutation causes a major gain of function in IKr, leading to acceleration of repolarization and abbreviation of APs. Abbreviation of the ventricular AP underlies the abbreviation of the QT interval. This, in turn, is believed to be responsible for the heterogeneous abbreviation of refractoriness, leading to the development of a re-entrant substrate that gives rise to arrhythmias.
COMPETENCY IN MEDICIAL KNOWLEDGE: Carriers with the highly frequent SQTS mutation (KCNH2-T618I) have complete penetrance, and a high incidence and family history of VT/VF and/or SCD after long-term follow-up. This accounts for 25.9% of genetically identified SQTS probands without a clear sex preference. Our data describes the clinical characteristics of patients with this hotspot mutation and provides guidance on treatment options.
TRANSLATIONAL OUTLOOK: Although quinidine is highly effective in prolonging the QTc interval among KCNH2-T618I carriers, it does not eliminate severe ventricular arrhythmias, which may results from the kinetic malfunction of the HERG channel.
The authors are grateful to Judy Hefferon for preparing figures and Susan Bartkowiak for maintaining the MMRL genetic database.
This study was supported by grants from CONACYT #FM201866; NIH-NHLBI Grant #HL47678; the Masons of New York, Florida, Massachusetts, Connecticut, Maryland, Wisconsin, and Rhode Island; and National Natural Science Foundation of China #81100067 and #81670304. Drs. Dan Hu, Yang Li, and Jiancheng Zhang contributed equally to this work.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- action potential
- early repolarization
- genome-wide association studies
- implantable cardioverter-defibrillator
- L-type calcium channel
- inward rectifier potassium channel
- rapidly activating delayed rectifier potassium channel
- slowly activating delayed rectifier potassium channel
- long QT syndrome
- minor allele frequency
- PR interval
- sudden cardiac death
- short QT syndrome
- ventricular tachycardia/ventricular fibrillation
- wild type
- Received December 22, 2015.
- Revision received November 18, 2016.
- Accepted November 22, 2016.
- 2017 American College of Cardiology Foundation
- Mazzanti A.,
- Kanthan A.,
- Monteforte N.,
- et al.
- Bellocq C.,
- Van Ginneken A.C.,
- Bezzina C.R.,
- et al.
- Brugada R.,
- Hong K.,
- Dumaine R.,
- et al.
- Priori S.G.,
- Pandit S.V.,
- Rivolta I.,
- et al.
- Antzelevitch C.,
- Pollevick G.D.,
- Cordeiro J.M.,
- et al.
- Gollob M.H.,
- Redpath C.J.,
- Roberts J.D.
- Priori S.G.,
- Blomström-Lundqvist C.,
- Mazzanti A.,
- et al.
- Harrell D.T.,
- Ashihara T.,
- Ishikawa T.,
- et al.
- Giustetto C.,
- Scrocco C.,
- Giachino D.,
- Rapezzi C.,
- Mognetti B.,
- Gaita F.
- ↵SIFT. J. Craig Venter Institute. Available at: http://sift.jcvi.org/. Accessed January 12, 2017.
- ↵Polyphen SMP Data Collection. Available at: http://genetics.bwh.harvard.edu/pph/data/. Accessed January 12, 2017.
- Algra A.,
- Tijssen J.G.,
- Roelandt J.R.,
- Pool J.,
- Lubsen J.
- Rogozin I.B.,
- Pavlov Y.I.
- Newton-Cheh C.,
- Guo C.Y.,
- Larson M.G.,
- et al.
- Extramiana F.,
- Antzelevitch C.
- Giustetto C.,
- Scrocco C.,
- Giachino D.,
- Antzelevitch C.,
- Gaita F.
- Gaita F.,
- Giustetto C.,
- Bianchi F.,
- et al.
- Villafane J.,
- Atallah J.,
- Gollob M.H.,
- et al.
- Gaita F.,
- Giustetto C.,
- Bianchi F.,
- et al.
- Frea S.,
- Giustetto C.,
- Capriolo M.,
- et al.
- Kaneko Y.,
- Horie M.,
- Niwano S.,
- et al.
- Giustetto C.,
- Schimpf R.,
- Mazzanti A.,
- et al.