Molecular Genetic And Cellular Electrophysiological Study Of Short QT Syndrome

Background and Objective:Short QT syndrome (SQTS), a new clinical entity, is characterized by theabbreviated QT interval on electrocardiogram(ECG), associated with a high incidenceof atrial fibrillation, syncope, and increased risk of sudden cardiac death (SCD) due toventricular tachyarrhythmias(VT). Molecular genetic study has approved that SQTSis a genetically determined ion-channel disorder. Thus far, mutations in six differentgenes encoding potassium and calcium channel subunits have been identified that areassociated with SQTS, indicating a inheritance heterogeneous disease.Long QT syndrome (LQTS), another genetically determined channelopathy, ischaracterized by prolonged QT interval on ECG and increased predisposition to SCDdue to VT. LQTS and SQTS are the two opposite disease phenotypes with cardiacrepolarization abnormalities. To data, four identified pathogenic genes (KCNQ1,KCNH2, KCNJ2and CACNA1C) in SQTS were also linked to LQTS with acompletely opposite function. The SCN5A gene, encoding for the α subunit of cardiacvoltage-gated sodium channel-Nav1.5, is a well known gene for LQTS withgain-of-function. However, to date, there is no report about SCN5A mutations inSQTS-the opposite disease phenotype of LQTS.To our knowledge, There is no systematic study about SQTS in China, exceptone report of KCNH2mutation. So, screening the SQTS patients in Chinese people tofind the diseas-causing gene and mutations, and reveal the molecular geneticmechanism and cellular electrophysiological pathogenesis, are the main target of thisstudy. On the basis of the genotype of SQTS, to screen the effective anti-arrhythmicdrugs for SQTS patients in Chinese, is another purpose of our study.Materials and Methods:Clinical investigationPatients with QTc≤370ms and the family members were recruited according tothe diagnostic guideline of SQTS(the international ECG guideline of2010). Detailed clinical evaluation was performed and acquired causes of QT interval abbreviationwere excluded.Genetic ScreeningAll the exons and exon-intron boundaries of SCN5A gene and the known genesfor SQTS including KCNH2, KCNQ1, KCNJ2, CACNA1C, CACNB2B, andCACNA2D1were screened by DNA direct sequencing.To distinguish betweencommon genetic polymorphisms and rare disease-causing mutations,100healthycontrol individuals of the same ethnic background were used. According the2011SQTS diagnostic criteria, the scores of the patients were assessed.Mutagenesis and TransfectionMutated SCN5A channel cDNA was generated by site-directed mutagenesissystem.The primers for mutagenesis were the following: E428G sense primer5'gctgagaccgGggagaaggaaaagcgcttccag3' and anti-sense primer5'ttccttctccCcggtctcagcgatggtggcttg3'; V1951L sense primer5'atcgcctacTtgatgagtgagaacttctc3' and anti-sense primer5'tcactcatcaAgtaggcgatgaggccctc3'; S1653S sense primer5'tgccctcatgatgtcActgcctgccctcttcaacatcg3' and anti-sense primer5'tgaagagggcaggcagTgacatcatgagggcaaagagc3'; R689H sense primer:5'tgctggaaccAtctcgcccagcgctacctg and anti-sense primer5'ctgggcgagatggttccagcatggtggac. The mutated plasmids were completely sequencedon both strands to ensure the presence of the mutation as well as the absence of othersubstitutions introduced by PCR. Human embryo kidney (HEK)293cells weretransfected with the WT-or mutated-SCN5A constructs.Electrophysiological AnalysisThe identified SCN5A variants in SQTS were expressed heterologously in HEK293cells and characterized by whole-cell patch clamp recordings as previouslydescribed. We characterized the membrane current and the voltage dependence ofsteady-state activation (SSA) and steady-state inactivatin(SSI). The effects ofantiarrhythmic drugs-amiodarone, mexiletine and propafenone on the mutated channel were evaluated to screen the effective anti-arrhythmic drugs for SQTS.Recording were made at room temperature.Results:Clinical FindingsThe range of QTc was from261ms to362ms in10unrelated probands with agefrom15to48years old. Three of10were females, five of10got VT, three of10gotsyncope and two of10had family history of SCD. All the individuals recruited in thestudy refused to peform the intracardiac electrophysiological studies.Genetics and Corresponding phenotypeThree heterozygous missense mutations were found in SCN5A generespectively. One mutation E428G consisting of A-to-G substitution at nucleotidesite1283(c.1283A＞G), predicted a substitution of Glutamic for Glycine at codon site428(p.Glu428Gly). The mutation R689H consisting of G-to-A substitution atnucleotide site2066(c.2066G＞A), predicted a substitution of Glutamic for Glycineat codon site689(p. Arg689His).The mutation V1951L consisting of G-to-Tsubstitution at nucleotide site5851(c.5851G＞T), predicted a substitution of Valinefor Leucine at codon site1951(p.Val1951Leu). The three mutations were notdetected in the200healthy control chromosomes of the same ethnic background.E428G proband showed recurrent syncope caused by high incidence of VT with shortQT interval (QTc270ms). Unfortunately, the DNA from her relatives could not beobtained because they refused to do genetic screening. V1951L proband had thehistory of recurrent palpitation and syncope with short QT interval (QTc347ms) andearly repolarization in I,aVL,II,III, aVF,V4-V6lead on ECG. The same mutation waspresent in his father, who had the history of palpation. Howerver, the ECG of hisfather was normal. A co-inherited synonymous variation S1653S (nucleotide c.4959C＞A; amino acid p.Ser1653Ser) in SCN5A, was found in V1951L proband and hisfather. R689H broband with short QT interval (QTc348ms), early repolarization inII,III and aVF, and Brugada-like ECG, had no related symptoms, but had a familyhistory of SCD.The scores of the three probands with SCN5A mutation were larger than4points, indicating the consistent diagnosis of SQTS. The rest of patients had nomutations in SCN5A gene and the known genes for SQTS.Electrophysiological PropertiesMammalian cell line HEK293cells were transfected by Wild type (WT)-SCN5Aand mutated-SCN5A (E428G,V1951L,S1653S,R689H). Compared to WT, E428Gchannel produced significant reduction of peak INaand no effect on voltagedependence of SSA and SSI of INa. V1951L channel showed it was a decreasetendency of INadensity, however there were no statistic difference between WT andV1951L channel in current density, voltage dependence of peak conductance andSSI of INa. There were no statistic difference between WT and S1653S channel incurrent density, voltage dependence of peak conductance and SSI of INa. In order toexplain a possible pathogenic role of the V1951L mutation, in vitro analyses wereperformed using co-expression of V1951L with synonymous variant S1653S(V1951L and S1653S) in Nav1.5. To our great surprise, V1951L and S16523Sproduced significant reduction of INadensity and hyperpolarization shift of bothvoltage dependence of SSA and SSI of INacomparing with WT. There was no INawiththe R689H mutated channel indicating loss-of-function of the mutation. Presence ofamiodarone(50μM), mexiletine(50μM), and propafenone(30μM) in vitro, onlypropafenone could significantly increase the peak INaof E428G channel. Theamiodarone, mexiletine, and propafenone induced a hyperpolarization shift of the I-Vrelationship curve, voltage dependence of SSA and SSI of INa. Both amiodarone andmexiletine had no effect on the peak current amplitude of E428G channel, whilepropafenone increased about ten percent of E428G channel current.Discussion and Conclusion:In the present study,we provide the first report that three mutations E428G,V1951L and R689H, and the new synnonymous variant S1653S in SCN5A gene,which on the basis of our functional studies is most likely pathogenic. Three probandswith SCN5A mutation display the short QT interval(QTc:270ms,347ms and348msrespectively) on ECG, a high predisposition to SCD, and no underlying structuralheart disease. Meanwhile, the acquired causes of QT interval abbreviation have been excluded, such as hyperkalemia, acidosis, hypercalcemia, hyperthermia, tachycardiaand QT-shortening medications. According the2011SQTS diagnostic criteria, theE428G proband get the score of7, and V1951L and S1653S proband get the score of5, R689H proband get the score of6,which indicates the consistent diagnosis ofSQTS. Through the above analysis, mutations in SCN5A gene are contributed to anew variant of SQTS.The proband10, in this study with short QT interval ECG, had no symptoms ofheart disease.Through the genetic screening, a SCN5A mutation R689H was found,which showed the loss-of-function. There were no mutations in the known genes forSQTS. So it was reasonable to think that R689H was associated with the short QTand the family history of SCD. The lack of cardiac symptoms may result from thephenotypes of SQTS being only partially penetrant. The ECG of this patient alsoshow the early repolarization in II,III and aVF, and Brugada-like ECG, whichindicated the concomitance of short QT and early repolarization. In fact, rencentlyresearchers begin to realize that a short QT interval in individuals with earlyrepolarization and vice versa. To date, the mechanism of this phenomenon was poorlyunderstood.E428G mutation in SCN5A, previously not reported in the inheritedchannelpoathy, was firstly detected in patient with SQTS in present study. E428Gsites in the intracellular linker connecting the Domain I and Domain II of Nav1.5. Theelectrophysiological property of E428G in the heterologous expression system showsthe significant decreasing of peak INa, indicating the obvious loss-of-function, whichis in accord with the overt arrhythmogenic phenotype of E428G carrier. The resultalso indicates the importance of Domain I-Domain II linker of Nav1.5in maintainingpeak current of sodium channel. The gating properties including voltage dependenceof SSA and SSI of E428G have no differences compared with WT.V1951L mutation in SCN5A has been found in patients with LQTS and suddeninfant death syndrome, and reported as a pathogenic mutation in Brugada syndrome.However, Akerman MJ et al identified that V1951L was a common variant (6.7%) inhealthy Hispanics and not found in non-Hispanic individuals, which indicated thatV1951L was relatively ethnic-specific polymorphism. In our study, the variant V1951L was not detected in200control chromosomes from healthy Chinese(incidence<0.5%), demonstrating a rare mutation in Chinese, which was inaccordance with the results of previous study. Our study results show the biophysicalfunction of V1951L no significant differences compared with WT, which is consistentwith the results from another study. However, the V1951L proband exhibitsprominent proarrhythmic phenotype, which is not accordance with the unalteredbiophysical function of V1951L. This findings prompted us to make furtherexploration whether S1653S would modify the biophysical function of V1951L. Withthis question, we perform the function analysis of co-expression of V1951L withS1653S(V1951L and S1653S). To our great surprise, V1951LandS1653S mutatedNav1.5exhibits a significant reduction of INadensity, which indicats that S1653Splays a noteworthy synergistic modulation effect to the biophysical function ofV1951L mutation and acts as a genetic modifier of disease. Meanwhile,the notablealteration of biophysical function caused by the compound variants should in theroyexpain the obvious proarrhythmic phenotype of V1951L and S1653S carrier. It isinteresting that our observation is not consist with the traditional theory thatsynonymous variant do not change the function of protein because of the no alteringfinal amino acid sequences. Recent study have proved a new theory that synonymousvariant can also have an strong effect on subsequent function of the protein. Thoughthe exact mechanism is not clear and a further research is need, our observation doprovide another strong evidence supporting the theory that synonymous variant canchange the function of protein and implicate in diseases.Both the E428G and the V1951L and S1553S variants in SCN5A exhibit theobviously loss-of-function in mutant Nav1.5and cause the overt arrhythmogenicphenotype-SQTS. This make us firstly realize the association between theloss-of-function in Nav1.5and SQTS. Though the exact mechanism that theloss-of-function mutation in SCN5A cause SQTS is not clear, it can be supported bythe following:1. The loss-of-function in Nav1.5in cardiomyocytes causes decreasedtotal inward ion, breaks the precise balance between the inward and outward ion inAP, and leads to the reduced depolarization reserve. If the repolarization rate is notchanging, the repolarization time will decrease and the AP and QT interval on ECG become shortened.2. It has been witnessed that the association between QT intervaland SCN5A variations through large population-based association studies.3. INareduction caused by the loss-of-function mutations in SCN5A may predispose to theVT/VF in SQTS by slowing the electrical conduction velocity,which is an essentialdeterminant for maintaining reentrant excitation waves.4. Both the sodium channeland calcium channel are responsible for the inward ion in AP. It has been proved theassociation beween the loss-of-function in calcium channel and SQTS.5. Thegain-of-function in both sodium and calcium channel can cause LQTS, and theloss-of-function in calcium channel caused the opposite phenotype of LQTS-SQTS.6.Rencently, researchers begin to realize that a short QT interval in individuals withearly repolarization and vice versa. While it has been proved that the relationshipbetween early repolarization and loss-of-fucntion mutations in SCN5A.7. There isincreasing evidence that BrS and SQTS is the same allele gene disease. It is knownthat SCN5A mutation can cause the BrS.In the present study, We find that propafenone, can increase the peak INaofE428G channel, which indicate propafenone may be a new quite potential drug totreat SQTS caused by the E428G mutation. It is not clear that whether propafenonecan be effective for SQTS caused by other SCN5A mutations. The gating property ofE428G channel intervened by propafenone shows SSA and SSI curve remarkablehyperpolarization shift, which means that the increment of INacaused byhyperpolarzation shift of SSA, is partly balanced by the decrement of INacaused byhyperpolarization shift of SSI. So the gating property is not the main reasonresponsible for the increasing INacaused by propafenone in E428G channel. Themechanism that propafenoe can increase the E428G mutant sodium current is notclear. Whether the propanone also have the specular role on the inactivated E428Gsodium channel and counteract the suppression role of E428G on sodium channel,and through what mechanism mitigate the INadecreasing caused by E428G mutation,remain to be clarified in the further research.In conclusion, our study is the first to show the loss-of-function mutations inSCN5A gene contributed to SQTS and causes a new variant of SQTS. Thesynonymous variation S1653S in SCN5A has significant modification role to the biophysical function of V1951L mutation and causes the loss-of-function in Nav1.5.Propafenone may be a quite potential drug to treat SQTS caused by E428G mutation.