ARTICLE Received2Jun 2015 |Accepted 16Feb 2016 |Published 11Apr 2016 DOI:10.1038/ncomms11067 OPEN Splicing misregulation of SCN5A contributes to cardiac-conduction delay and heart arrhythmia in myotonic dystrophy Fernande Freyermuth1,*,w, Fre´de´rique Rau2,*, Yosuke Kokunai3, Thomas Linke4, Chantal Sellier1, Masayuki Nakamori3, Yoshihiro Kino5, Ludovic Arandel2, Arnaud Jollet2, Christelle Thibault1, Muriel Philipps1, Serge Vicaire1, Bernard Jost1, Bjarne Udd6,7,8, John W. Day9, Denis Duboc10, Karim Wahbi10, Tsuyoshi Matsumura11, Harutoshi Fujimura11, Hideki Mochizuki3, Franc¸ois Deryckere12, Takashi Kimura13, Nobuyuki Nukina14, Shoichi Ishiura15, Vincent Lacroix16, Amandine Campan-Fournier17, Vincent Navratil18, Emilie Chautard19, Didier Auboeuf19, Minoru Horie20, Keiji Imoto21, Kuang-Yung Lee22, Maurice S. Swanson23, Adolfo Lopez de Munain24, Shin Inada25, Hideki Itoh20, Kazuo Nakazawa25, TakashiAshihara20,EricWang23,ThomasZimmer4,DenisFurling2,MasanoriP.Takahashi3&NicolasCharlet-Berguerand1 Myotonic dystrophy (DM) is caused by the expression of mutant RNAs containing expanded CUG repeats that sequester muscleblind-like (MBNL) proteins, leading to alternative splicing changes.Cardiacalterations,characterizedbyconductiondelaysandarrhythmia,arethesecond most common cause of death in DM. Using RNA sequencing, here we identify novel splicing alterationsinDMheartsamples,includingaswitchfromadultexon6Btowardsfetalexon6Ain thecardiacsodiumchannel,SCN5A.WefindthatMBNL1regulatesalternativesplicingofSCN5A mRNA and that the splicing variant of SCN5A produced in DM presents a reduced excitability comparedwiththecontroladultisoform.Importantly,reproducingsplicingalterationofScn5ain mice is sufficient to promote heart arrhythmia and cardiac-conduction delay, two predominant featuresofmyotonicdystrophy.Inconclusion,misregulationofthealternativesplicingofSCN5A maycontributeto asubsetofthecardiacdysfunctionsobservedin myotonicdystrophy. 1DepartmentofTranslationalmedicineandneurogenetics,IGBMC,CNRSUMR7104,INSERMU964,Universite´deStrasbourg,Illkirch67400,France.2SorbonneUniversite´s UPMCUnivParis06,Inserm,CNRS,CentredeRechercheenMyologieUMRS974/FRE3617,InstitutdeMyologie,GHPitie´-Salpeˆtrie`re,Paris75013,France.3Departmentof Neurology,OsakaUniversityGraduateSchoolofMedicine,Osaka565-0871,Japan.4DepartmentofPhysiology,FriedrichSchillerUniversityHospital,Jena07743,Germany. 5DepartmentofBioinformaticsandMolecularNeuropathology,MeijiPharmaceuticalUniversity,Kiyose205-8588,Japan.6NeuromuscularResearchCenter,Tampere UniversityandUniversityHospital,Tampere33520,Finland.7DepartmentofMedicalGenetics,Folkha¨lsanInstituteofGenetics,HelsinkiUniversity,Helsinki00250,Finland. 8DepartmentofNeurology,VaasaCentralHospital,Vaasa65130,Finland.9DepartmentofNeurology,StanfordUniversity,Stanford,California94304,USA.10Servicede Cardiologie,Universite´Paris-Descartes,HoˆpitalCochin,AP-HP,Paris75014,France.11DepartmentofNeurology,ToneyamaNationalHospital,Toyonaka560-8552,Japan. 12CNRSUMR7175,EcoleSupe´rieuredeBiotechnologiesdeStrasbourg,Illkirch67400,France.13DivisionofNeurology,HyogoMedicalCollege,Nishinomiya663-8501,Japan. 14LaboratoryofStructuralNeuropathology,DoshishaUniversityGraduateSchoolofBrainScience,Kyoto610-0394,Japan.15GraduateSchoolofArtsandSciences,University ofTokyo,Tokyo153-8902,Japan.16Universite´Lyon1,CNRS,UMR5558LBBE,Villeurbanne69622,France.17HospicescivilsdeLyon,Laboratoiredecytoge´ne´tique constitutionelle,Bron69500,France.18PoˆleRhoˆneAlpesdeBioinformatique,Universite´Lyon1,BaˆtimentGregorMendel,Villeurbanne69100,France.19CentredeRecherche enCance´rologiedeLyon,Lyon69373,France.20DepartmentofCardiovascularandRespiratoryMedicine,ShigaMedicalUniversity,Otsu520-2192,Japan.21Departmentof InformationPhysiology,NationalInstituteforPhysiologicalSciences,Okazaki444-8585,Japan.22DepartmentofNeurology,ChangGungMemorialHospital,Keelung20401, Taiwan.23DepartmentofMolecularGeneticsandMicrobiology,CenterforNeuroGeneticsandtheGeneticsInstitute,UniversityofFlorida,CollegeofMedicine,Gainesville, Florida32610,USA.24DepartmentofNeurology,HospitalUniversitarioDONOSTIA,NeuroscienceArea,InstituteBiodonostiaCIBERNEDandUniversityofBasqueCountry UPV-EHU,SanSebastia´n20014,Spain.25LaboratoryofBiomedicalSciencesandInformationManagement,NationalCerebralandCardiovascularCenterResearchInstitute, Osaka565-8565,Japan.*Theseauthorscontributedequallytothework.wPresentaddress:MassachusettsGeneralHospital,MassGeneralInstituteforNeurodegenerative Diseases,Charlestown,Massachusetts02129,USA.CorrespondenceandrequestsformaterialsshouldbeaddressedtoD.F.(email:[email protected])orto M.P.T.(email:[email protected])ortoN.C-B.(email:[email protected]). NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications 1 ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 M yotonicdystrophy(DM),themostcommonadult-onset observed in DM. These results suggest that altered splicing of muscular dystrophy, includes two genetically distinct SCN5A mRNA may participate to the electrical cardiac forms. DM of type 1 (DM1) and its severe congenital abnormalities observed in DM. form(CDM1)arecausedbyanexpansionofCTGrepeatsinthe 30-untranslated region (UTR) of the DMPK gene1–3. In contrast, DMoftype2(DM2)iscausedbyanexpansionofCCTGrepeats Results withinthefirstintronoftheCNBP(alsoknownasZNF9)gene4. Identification of splicing changes in DM heart samples. To The pathogenesis of DM involves a RNA gain-of-function determine novel splicingabnormalities in DMheart samples, we mechanism caused by expression of mutant RNAs containing first used whole-genome microarrays (GeneChip Human Exon hundred to thousands of CUG or CCUG repeats that interfere 1.0STarray)onpolyadenylatedRNAextractedfromleftventricle with the splicing of other pre-mRNAs through dysfunction samples of three adult DM1 patients compared with three age- of two classes of RNA-binding proteins. MBNL proteins matched control individuals. Bioinformatic analyses predicted (MBNL1, MBNL2 and MBNL3) are sequestered within nuclear significant (Fold Change Z2, Sudent t-test, P value r0.01) RNA foci formed by expanded CUG and CCUG repeats5,6, changes in the splicing of 24 exons between control and DM1 whereas expression and phosphorylation of CUG-binding samples (Supplementary Table 1), including a misregulation of protein 1 (CUGBP1, encoded by the CELF1 gene) are increased thealternativesplicingoftheSCN5Apre-mRNA.Toextendthis in DM1 heart samples7. MBNL and CUGBP1 proteins regulate analysis, we performed paired-end RNA sequencing (RNA-seq) alternative splicing, and alterations of their functional levels in on the same DM1 and control heart samples, yielding 1,611 myotonic dystrophic tissues results in reversion to fetal million of mapped 100bp reads. DESeq and Cuffdiff were then splicing patterns for several mRNAs, such as the insulin applied to estimate differential gene expression and over or receptor (INSR) (ref. 8), the muscle chloride channel (CLCN1) under-expressed mRNAs were selected by using the Benjamini (refs 9,10), dystrophin (DMD) (refs 11,12) and key components and Hochberg adjusted P values (false discovery rate (FDR) of the skeletal muscle excitation–contraction coupling r0.1). A total of 9 and 19 upregulated genes were predicted process, including amphiphysin2 (BIN1) (ref. 13), ryanodine differentially expressed with DESeq and Cuffdiff, respectively, receptor 1 (RYR1) (ref. 14), sarcoplasmic/endoplasmic reticulum but none were confirmed by quantitative real-time RT-qPCR Ca2þ-ATPase SERCA1 (ATP2A1) (ref. 14) and the muscle analyses. This low number of differentially expressed mRNAs calcium channel Ca 1.1 (CACNA1S) (ref. 15). Misregulation of suggests that cardiac pathology in DM is not associated with V the alternative splicing of the insulin receptor INSR, CLCN1 and drastic modifications of gene expression levels. In contrast, DMD mRNAs are associated with the insulin resistance8, DEXSeq(ref.22),whichtestsdifferentialexonusagebetweentwo myotonia9,10,16 and dystrophic process12, respectively, while conditions, predicted 134 significant (Log2 Fold Change Z1.2, alterations of the alternative splicing of BIN1, RYR1, ATP2A1 FDR r0.1) alternative splicing changes between control and and CACNA1S may contribute to the skeletal muscle weakness DM1 heart samples (Supplementary Data 1). Similarly, MISO observed in DM13–15. (ref. 23) analysis, which computes the fraction of mRNA that In contrast, the molecular mechanisms underlying the cardiac includes a given cassette alternative exon, predicted 259 defects, which affect 80% of individuals with DM and represent significant (DPSI Z0.3; Z-score Z1.2) alternative splicing the second most common cause of death in this disease17,18, are changes between control and DM1 heart samples (Fig. 1a and yettobedefined. CardiacinvolvementsinDMarecharacterized Supplementary Data 2), including a robust misregulation of the by cardiac-conduction delay that may result in fatal atrio- alternative splicing of SCN5A (Fig. 1b). MISO and DEXSeq pre- ventricular block, and by atrial or ventricular tachycardia17,18. dictions overlapped, but with some exceptions, such as the Electrocardiography (ECG) analyses in DM patients indicate skipping of the consecutive exons 18, 19 and 20 of CAMK2B prolonged conduction time from the sinoatrial node to the predicted by DEXSeq but not by MISO; or the shift of SCN5A ventricles (PR interval) and elongated ventricular depolarization exon6Btowardsexon6AidentifiedbyMISObutnotbyDEXSeq. (QRS duration). Interestingly, cardiac dysfunctions in DM are Thesedifferencesareinherenttotheircomputationmodels,since reminiscent in some aspect to an alteration of the cardiac MISOdoesnotdetectalterationsofsuccessiveexonsandDEXSeq sodiumcurrent.Thea-subunitofthecardiacvoltage-gatedNaþ does not identify mutually exclusive exons, highlighting that channel, Na 1.5, is encoded by the SCN5A gene and plays a key MISO and DEXSeq are complementary bioinformatics v role in the excitability of cardiomyocytes and for rapid approaches. Next, we tested by PCR with reverse transcription propagation of the impulse through the cardiac-conduction (RT–PCR)fortycandidatemRNAshavingthehighestprobability system. Mutations in SCN5A lead to a variety of arrhythmic ofmisregulationinDEXSeqand/orMISOanalyses.Wevalidated disorders, including long QT3, progressive and non-progressive splicingalterationsfor32ofthem,includingsomethathavebeen cardiac-conduction disease (also known as Lev-Lene`gre disease), identified in previous studies (TNNT2, TNNT3, ABLIM1, LDB3, atrial fibrillation, sick sinus syndrome, Brugada syndrome and MBNL1, CAMK2B, MAPT and so on)24–26, and 20 others that numerous overlapping syndromes19–21. represent, to the best of our knowledge, novel alterations of Using transcriptomic approaches, we identified various novel alternative splicing (ADD3, GOLGA4, CRTC2, ARHGEF10L, splicingchangesinheartsamplesofDM1individuals.Analysisof ANK3, DCLK2, EPN2, UNC13B, TECR, ARVCF, SOCS7, CELF1 the RNA motifs enriched in the vicinity of these misregulated and so on) in DM1 heart samples (Fig. 1c). Of interest, some of exons indicates that sequestration of the MBNL proteins is these splicing alterations may be of pathological consequence probably the main cause of splicing misregulation in heart of inDM.Forexample,knockoutoftheSocs7geneinmouseresults individuals with DM. Among these novel splicing alterations, in insulin resistance27. Whether the splicing misregulation of we focused on misregulation of alternative splicing of the SOCS7 in DM contributes to insulin resistance remains to be SCN5A pre-mRNA. This splicing alteration results in expression tested. Also, RNA sequencing predicts an increased retention of of a fetal isoform of SCN5A with altered electrophysiological thepenultimateintronofFCGRT,whichencodestheFcfragment properties. Of importance, we demonstrate that reproducing of the IgG receptor transporter a (FCRN) protein, involved in the splicing alteration of Scn5a in mouse is sufficient to cause IGGrecycling28.WhethersplicingalterationofFCGRTinDMis heartarrhythmiaandcardiac-conduction delaywithelevatedPR responsible to the decreased level of IGG in blood of these interval, which are key characteristics of the heart alterations patients is an attractive hypothesis that remains to be tested. 2 NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 a b Exons excluded in DM Exons included in DM SCN5A 3 5 6A 7 SCN5A 6B RPKM321000 12Z score RPKMRPKMRPKM321321321000000000 CTL RPKM321000 DM1 RPKM321000 –1 –0.5 0 0.5 1 38662462 38658666 38654924 38651227 Δ PSI Genomic coordinate (chr3), “-” strand c CTL DM1 CTL DM1 CTL DM1 CTL DM1 320000 6B 200 + 14 320000 + 8 435000000 ++ 1189..1290.20 100 6A 100 – 14 100 – 8 120000 – 18.19.20 SCN5A ABLIM1 COPZ2 CAMK2B CTL DM1 CTL DM1 CTL DM1 CTL DM1 300 300 200 200 200 + 16 + 6 + 24 + 10 – 16 200 – 6 100 – 24 – 10 100 100 ADD3 CLTB GOLGA4 SUN1 CTL DM1 CTL DM1 CTL DM1 CTL DM1 200 300 300 400 + 6 + 13 + 4 300 + 8 – 6 200 – 13 200 – 4 200 – 8 100 100 100 MYH11 CRTC2 MXRA7 ARHGEF10L CTL DM1 CTL DM1 CTL DM1 CTL DM1 300 300 200 300 + 47 200 + 20 200 + 40 100 + 5 200 - 47 – 20 – 40 – 5 100 100 NCOR2 NUMA1 ANK3 DCLK2 CTL DM1 CTL DM1 CTL DM1 CTL DM1 300 400 300 + 6B 300 + 5 200 + 38 200 + 5 200 – 6A 200 – 5 – 38 100 – 5 100 100 TPM2 EPN2 UNC13B SOCS7 CTL DM1 CTL DM1 CTL DM1 CTL DM1 200 300 200 + 2 + 6 200 + 19 + 2 200 100 – 2 100 – 6 100 – 19 100 – 2 TECR ZFYVE21 ARVCF CELF1 Figure1|IdentificationofnovelsplicingmisregulationsinDM1heartsamples.(a)D-PSIversusZ-scoreplotofexoncassettesmisregulationspredicted byMISOanalysis.(b)ExonsstructureandcoverageofRNA-seqreadsacrossSCN5Aexons5–7showincreasedinclusionofexon6Aanddecreased inclusionofexon6BinheartsamplesofthreeDM1patients(bottom,blue)versusthreecontrolsamples(top,red).(c)Validationby RT–PCRofRNA-seqpredictionsinhumanheartsamplesofnormaladultindividuals(CTL,black)versusadultDM1patients(DM1,red).Molecularsize markersinbpsarereportedtotheleftofeachRT–PCRgels.bp,basepair. Finally,RNAsequencingpredictsmisregulationofthealternative these RNA motifs are indeed present in the vicinity of exons splicing of a cardiac-specific exon located in the 50-UTR of misregulated in DM, we determined all 4-mer RNA motifs CELF1, which encodes CUGBP1. Whether this alternative enrichedwithin,upstreamordownstreamoftheexonspredicted splicing may contribute to the increase levels of CUGBP1 asmisregulated byMISOinDM1heartsamples,comparedwith protein observed in DM1 hearts remains also to be evaluated. 2,000 control exons (Fig. 2). Most RNA motifs significantly enriched (binomial test, P value o1.0 10(cid:2)7) contained YGC sequences, while none were found to contain UGU sequences. Splicing altered in DM are enriched for MBNL-binding sites. Furthermore, YGC sequences were enriched upstream of Mutants RNAs containing expanded CUG or CCUG repeats exons abnormally included in DM1, while YGC motifs were interfere with the functional levels of CUGBP1 and MBNL enriched downstream of exons repressed in DM1. These results proteins. Earlier studies determined thatMBNLproteins bind to matchedtheMBNLsplicingregulatorymapdeterminedbyCLIP YGCRNAmotifs(whereYisapyrimidine)29–32,whileCUGBP1 experiments,wherebindingofMBNLupstreamofanexontends binds to UGU-enriched sequences33,34. To determine whether toinhibitexoninclusionwhereasbindingofMBNLdownstream NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications 3 ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 UGCU (2×10–26) GCUU (2×10–19) CUGC (2×10–15) UUGC (2×10–9) CUGC (2×10–11) Exons included in DM GAAG (5.1 10–7) CGCU (4×10–9) UGCU (3×10–9) GCUC (7×10–7) GCUG (6×10–8) CCCC (2×10–18) Exons excluded in DM CUGC (9×10–15) CCGC (2×10–22) UGCU (2×10–14) CGCC (8×10–17) CUAA (5×10–10) CGCU (9×10–10) UUGC (5×10–8) UCGC (4×10–7) CCUG (4×10–7) GCGC (7×10–7) UGCC (7×10–7) CCCG (8×10–7) Figure2|MBNL-bindingmotifsareenrichedinvicinityofexonsmisregulatedinDM1.SequenceandbinomialtestPvaluesof4-merRNAmotifs enricheddownstream,withinandupstreamofexonsmisregulatedinDM1heartsamples.SequencesenrichedinexonsexcludedinDMareindicatedinred, whilesequencesenrichedinexonsincludedinDMareindicatedinblue. of an exon generally stimulates exon inclusion35,36. In contrast, alternative splicing of SCN5A resulting in expression of a fetal we found no enriched motifs for other RNA-binding proteins, form of this channel in adult DM heart. These results are includingCUGBP1,rbFOX1,hnRNPHorStaufen.Theseresults, consistent with previous studies where alternative splicing as well as previous data36–38, support a model in which titration changes in DM resume a MBNL-dependent fetal splicing of MBNL proteins is the main cause of splicing change in DM1 pattern that persist in adult tissues26,37. heart, while misregulation of other RNA-binding proteins may contribute to a subset of splicing alterations. Alternative splicing of SCN5A is regulated by MBNL1. To determine the mechanisms underlying misregulation of SCN5A Splicing of SCN5A is misregulated in DM heart samples. Both splicing,wefirstdetermineditssplicingpatternincellmodelsof microarray and RNA-seq predicted misregulation of alternative DM.SinceSCN5Aisexpressedatlowlevelincultureofimmature splicingofSCN5Apre-mRNAinDM1heartsamples.Splicingof skeletal muscle cells, we investigated its splicing in primary SCN5A is developmentally regulated, such that exon 6A is cultures of differentiated skeletal muscle cells originating from included in fetal heart but rapidly replaced by exon 6B after muscle biopsies of control and DM1 individuals. RT–PCR birth39.Consequently,SCN5Aexon6Aisnamedasembryonicor experiments determined a switch of exon 6B towards exon 6A fetal, while exon 6B is known as adult. Exons 6A and 6B are in DM1 muscle cells compared with control, reproducing the mutually exclusive exons encoding part of the voltage sensor, splicing alteration observed in cardiac tissue (Fig. 4a). Of segments3and4locatedinthedomainIofthesodiumchannel technical interest, the basal level of exon 6A inclusion was (Fig. 3a,b). These are key segments for the electrical activity of higher in muscle cell cultures than in adult heart samples the sodium channel, and inclusion of either fetal exon 6A or (compareFig.4atoFig.3c),whichprobablyreflecttheimmature adult exon 6B results in channel isoforms, named, respectively, aspect of cell cultures. Since mutant RNAs containing expanded hNa 1.5e and hNa 1.5, with different electrophysiological CUGorCCUGrepeatsinterferewithalternativesplicingthrough v v properties39–41. We confirmed our microarray and RNA-seq titration of MBNLproteins, we testedwhetherMBNL1 regulates predictions by RT–PCR and found that adult SCN5A exon 6B is SCN5A splicing. Reduction of MBNL1 expression through a partlyreplacedbyitsfetalexon6Ainheartsamplesofindividuals siRNA-mediated approach in human control primary muscle with DM, including adult DM1 and adult DM2 cases (Fig. 3c). cellsmimickedtheeffectofCUGrepeatsandpromotedaswitch Note that to differentiate exon 6B from exon 6A that have from adult exon 6B towards fetal exon 6A (Fig. 4b). Western the exact same length of 92bp, we took advantage of a BstbI blotting analysis confirmed the successful depletion of MBNL1 restriction site present only in exon 6A, which thus appears as a expression(SupplementaryFig.3A).Next,weassessedalternative BstbI-digested doublet band in Fig. 3c. These results are splicing of Scn5A in heart samples of Mbnl knockout mice43. consistent with the recent report of a splicing misregulation of RT–PCRanalysisshowsthatinclusionoftheexon6AofScn5ais SCN5AinoneDM1heartsample42.AlthoughsplicingofSCN5A increased in heart samples of mice with no Mbnl1 and reduced ismisregulatedinDM1,weobservednocorrelationbetweenthe level of Mbnl2 (Mbnl1(cid:2)/(cid:2), Mbnl2þ/(cid:2))(Fig. 4c). The increased percentage of SCN5A exon 6A inclusion and the increased inclusion of Scn5a exon 6A in Mbnl knockout mice is duration of the PR interval and only a very limited, if any, significant (Student t-test, P value r0.01) but rather mild, correlation between misregulation of SCN5A exon 6A splicing probablyreflectingdifferenceinregulationofalternativesplicing and alteration of the QRS duration in individuals with DM1 betweenhumanandmouseorthecompensatoryeffectofresidual (R2 of 0.2 with six DM1 samples; Supplementary Fig. 1). Mbnl2 expression43. This hypothesis is consistent with the mild Misregulation of SCN5A splicing was specific to DM, as we did splicingalterationofScn5AobservedinthesoleMbnl1knockout not observe inclusion of exon 6A in heart samples from mice44.Overall,theseresultssuggestthatMBNLproteinsregulate individual affected with Duchenne muscular dystrophy (DMD), thealternativesplicingofSCN5Aexons6Aand6B.Todetermine amyotrophic lateral sclerosis (ALS) or dilated cardiomyopathy whether this regulation is direct or indirect, we constructed a (DCM) (Fig. 3d). Moreover, misregulation of splicing in DM1 minigene containing exons 6A and 6B of SCN5A bordered by wasspecificandnotglobal,asweobservednosplicingchangesof their intronic regions. Expression of this construct in mouse SCN5A alternative exon 18, of CACNA1C mutually exclusive C2C12 myoblasts reproduced a fetal pattern with mainly exons 8A and 8B, of KCNAB1 alternative exons 2 and 11, or of inclusion of exon 6A (Fig. 4d). Since inclusion of exon 6B was KCNQ1alternativeexons2and5(SupplementaryFig.2).Finally, repressed, reduction of Mbnl1 activity through siRNA or we observed no significant alteration of the expression level of expression of expanded CUG repeats had no further repressive SCN5A mRNA by quantitative real-time RT-qPCR (Fig. 3e). effectonexon6B.Incontrast,expressionofMBNL1promoteda Overall, these results indicate a specific misregulation of switch from fetal exon 6A towards adult exon 6B, while 4 NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 a b I II III IV Fetal 5 6A 6B 7 12345 6 12345 6 12345 6 12345 6 BstBI Adult NH2 OH O C c bp Adult CTL Adult ALS Fetal CTLCongenital DM1 Adult DM1 Adult DM2 300 Exon 6B 200 Exon 6A 100 d e 100 A % Exon 6Ainclusion 24680000 CN5A mRNexpression 0.51 S 0 0 Fetal CATdLult CTL ALS DCCMongDeMniDtal DAMd1ult DAMd1ult DM2 CTL DM1 Figure3|SplicingofSCN5Aexon6AisalteredinDMheartsamples.(a)Schematicrepresentationofmutuallyexclusiveexons6Aand6BofSCN5A. SCN5AmRNAincludesexon6A(red)infetalheart,whileSCN5AmRNAexpressesexon6B(blue)inadultheart.(b)SchematicrepresentationofSCN5A topologyexpressingexon6A(red).Exons6Aor6Bencodespartofsegment3,connectingloopbetweenS3andS4andmostpartofthevoltage-sensitive segment4ofdomain1ofthesodiumchannelSCN5A.(c).RepresentativeBstBI-digestedRT–PCRanalysisofendogenousSCN5AmRNAfromhumanheart samplesofnormaladult(CTL),adultALS,non-DMfetuses(20,24and35weeks),congenitalDM1fetuses(CDM1of22,25and28weeks),adultsDM1 andDM2individuals.Molecularsizemarkerisindicatedinbp.(d)GraphicalrepresentationofRT–PCRanalysisdepictingthepercentageofSCN5AmRNA includingexon6Ainleftventricularheartsamplesfromfetalandadultcontrol,ALS,DCM,DMD,CDM1andadultDM1andDM2individuals.(e)Graphical representationofquantitativereal-timeRT-qPCRdepictingthemRNAexpressionofSCN5ArelativetoRPLP0incontrolnormaladults(n¼5)versusadult DM1(n¼5)heartsamples.Barsindicates.e.m.bp,basepairs. expression or siRNA-mediated depletion of CUGBP1 had no (Fig. 5a and Supplementary Table 2). Injection of RNA effect (Fig. 4d). Western blotting analysis confirmed that siRNA encoding hNav1.5e, which is the splicing isoform of SCN5A transfection efficiently reduced endogenous Mbnl1 or Cugbp1 found in DM, indicated a significant reduction of the sodium expression (Supplementary Fig. 3B and C). Next, gel-shift assays current amplitude of 45%, compared with hNa 1.5, the normal v determined that recombinant purified GST-tagged MBNL1 adult SCN5A exon 6B form (Fig. 5b,c). Since, the extent of bound to UGC RNA motifs located upstream of exon 6A splicing misregulation varies among DM individuals, which (Fig.4e).Ofinterest,thisUGCsequenceisabsentfromthemouse typicallyexpressamixofSCN5Asplicingformscontainingeither genome,whichmayexplainthemildsplicingalterationofScn5A exon 6A or exon 6B (cf. Fig. 3c), we analysed sodium currents observedinmiceknockoutforMbnlproteins.Mutationof these generated by a mix of both SCN5A isoforms (Fig. 5a). Injecting UGC motifs abolished MBNL1 binding (Fig. 4f), as well as the Xenopus oocytes with an equimolar mix of RNA encoding each regulatory effect of MBNL1 on a mutant SCN5A minigene channel, namely 50% of hNa 1.5e (SCN5A containing fetal exon v (Fig. 4g). Overall, these results establish that MBNL1 regulates 6A) and 50% of hNa 1.5 (SCN5A expressing adult exon 6B), v directly alternative splicing of SCN5A exons 6A/6B. resultedinareductionof30%ofthecurrentamplitudecompared with the control hNa 1.5 (Fig. 5b,c). Next, two-electrode voltage v clamp recording experiments revealed that the steady-state SCN5A splicing forms present different electrical properties. activation of the fetal hNa 1.5e was shifted by 7mV towards v SCN5A encodes Na 1.5, the main cardiac voltage-gated sodium depolarized potential compared with the control adult hNa 1.5 v v channel, and loss-of-function mutations in SCN5A lead to a form(Fig.5dandSupplementaryTable2).Thisshiftisconsistent variety of arrhythmic disorders, which share some common with the shift observed previously in transfected mammalian pathological features with DM. Furthermore, exons 6A and 6B cells39–41, thus validating our approach in Xenopus oocytes. To differatsevenaminoacidpositions,resultinginchannelvariants better reproduce the situation observed in DM, we injected in with different electrophysiological properties39–41. To investigate Xenopus oocytes an equimolar mix of DM (hNa 1.5e, fetal exon v the consequences of the switch from SCN5A exon 6B towards 6A) and control (hNa 1.5, adult exon 6B) RNA isoforms of v exon 6A observed in DM, we first examined in Xenopus oocytes SCN5A. Importantly, this mix of splicing forms also presented a the sodium currents generated by either hNa 1.5e, the splice significant shift of steady-state activation towards depolarized v variant of SCN5A containing the fetal exon 6A, or hNa 1.5, potentials by 3.8mV, compared with the control hNa 1.5 form v v encoded by SCN5A containing the adult control exon 6B (Fig.5d,SupplementaryTable2).Correspondingly,asimilarshift NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications 5 ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 a b c Mbnl1–/– bp CTL DM1 bp siCTL siMBNL1 bp CTL Mbnl2+/– 300 6B 6B 6B 300 200 200 200 6A 6A 100 6A 100 100 B 60 B 60 A30 n 6 40 n 6 40 n 620 ** o o o x x x % E 20 *** % E 20 *** % E10 0 0 0 d e CMV PolyA bp Control CUG 960xMBNL1 siMbnl1 CUGBP1siCelf1 5 6A 6B 200 UUUGCUAUGCUGUGCUAUGCCUUGCAG 6B SCN5A minigene WT # MBNL1 100 6A Bound B 60 6 n 40 *** o Ex 20 Free % 0 f g CMV PolyA 5 XXX 6A 6B bp Control CUG 960xMBNL1 siMbnl1 200 UUU_CUAU_CUGU_CUAU_CCUU_CAG 6B SCN5A minigene MUT MBNL1 100 6A Bound B60 6 n 40 o Free Ex20 % 0 Figure4|MBNL1regulatesalternativesplicingofSCN5A.(a)Upperpanel,RT–PCRanalysisofendogenousSCN5AmRNAfromdifferentiatedprimary musclecellculturesderivedfrombiopsiesofcontrolorDM1individuals.(lower)QuantificationofthepercentageofSCN5AmRNAincludingexon6B. (b,upper)RT–PCRanalysisofendogenousSCN5AmRNAfromhumandifferentiatedculturesofcontrolprimarymusclecellstransfectedwithascrambled siRNA(siCTL)orasiRNAtargetingMBNL1mRNA(siMBNL1).(lower)PercentageofSCN5AmRNAincludingexon6B.(c,upper)RT–PCRanalysisof endogenousScn5amRNAinheartsamplesofwild-typeandcompoundMbnl1(cid:2)/(cid:2),Mbnl2þ/(cid:2) doubleknockoutmice.(lower)PercentageofScn5amRNA includingexon6A.(d,upper)RT–PCRanalysisofexogenousSCN5AmRNAfromdifferentiatedC2C12musclecellsco-transfectedwithaSCN5Aminigene containingexons6Aand6Bborderedbytheirintronsandwitheitheraplasmidexpressing960CTGrepeats,MBNL1,CUGBP1orwithasiRNAdirected againstMbnl1(siMbnl1)orCelf1(encodingCugbp1;siCelf1).#Indicatesusageofacrypticsplicesiteinherenttotheminigene.(lower)PercentageofSCN5A mRNAincludingexon6B.(e,upper)SchematicrepresentationofSCN5Aminigene,includingtheUGC-richsequenceusedforbindingassays.(lower)Gel- shiftassayswereperformedusing5–1,000nMofpurifiedbacterialrecombinantGST-MBNL1D101andauniformly32P-CTPlabelledRNA.(f,upper) SchematicrepresentationofmutantSCN5Aminigene,includingthemutantsequence,usedforbindingassays.(lower)Gel-shiftassayperformedasin e.(g,upper)RT–PCRanalysisofexogenousSCN5AmRNAfromdifferentiatedC2C12musclecellsco-transfectedwithmutantSCN5Aminigeneandwitha plasmidexpressing960CTGrepeatsorMBNL1orwithasiRNAdirectedagainstMbnl1(siMbnl1).(lower)PercentageofSCN5AmRNAincludingexon6B. Alltransfectionandgel-shiftexperimentswererepeatedthreetofivetimes.Molecularsizemarkersareindicatedinbp.Barsindicates.e.m.Studenttest,** indicatesPo0.01,***indicatesPo0.001.bp,basepairs. was observed for the time constant of inactivation (Fig. 5e). 6A,presentsareducedexcitabilitycomparedwithhNa 1.5,which v Consistent with previous electrophysiological studies39–41, no is the adult control SCN5A isoform containing exon 6B. significant differences were observed between hNa 1.5 and v hNa 1.5e regarding steady-state inactivation and recovery from v inactivation (Fig. 5f,g). Overall, our results are consistent with AlterationofSCN5Asplicingleadstoheartconductiondefects. previousstudies39–41,anddemonstratethathNa 1.5e,thesplicing Misregulation of thealternative splicingof SCN5A in DMisone v form of SCN5A expressed in DM and containing the fetal exon alteration identified among many others, thus questioning the 6 NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 a hNav1.5 (SCN5A exon 6B) hNav1.5e (SCN5A exon 6A) 50% hNav1.5 + 50% hNav1.5e 1 μA 2 ms b c Voltage (mV) 4 –60 –40 –20 0 20 40 60 A) 0 μnt ( 3 *** –1 e *** eak curr 12 μnt (A) –2 P e urr –3 0 C hNa1.5 hNav1.5 hNav1.55e0% h% NhaNv1.a5v1.5e ––45 h55N00%%avv 1hh.NN5aaevv 11..55 e+ 50 d e 1.0 ms) 14 Steady-state activation 00000.....02468 hh55NN00%%aavv 11hh..NN55aaevv11..55 e+ nactivation time constant ( 110246802 h55hNN00%%aavv 11hh..NN55aa evv 11..55 e+ –60–50–40–30–20–10 0 10 20 I –40 –30 –20 –10 0 10 Voltage (mV) Voltage (mV) f g n 1.0 1.0 o Steady-state inactivati 0000....2468 5h5hNN00%%aavv 11hh..NN55aaevv11..55 e+ Fractional recovery 0000....2468 5h5hNN00%%aavv 11hh..NN55aaevv11..55 e+ 0.0 0.0 –130 –110 –90 –70 –50 –30 0 20 40 60 80 100 Voltage (mV) Recovery interval Δt (ms) Figure5|ElectrophysiologicalpropertiesofhNa1.5andhNa1.5echannels.(a)RepresentativeNaþ currentsgeneratedinXenopusoocytesbyhNa1.5 v v v (encodedbySCN5Acontainingtheadultexon6B),hNa1.5e(encodedbySCN5Aincludingthefetalexon6A),andsimultaneouslyexpressedNa1.5and v v Nav1.5echannelsata1:1ratio.(b)Peakcurrentamplitudesatthetestpotentialof (cid:2)10mVinXenopusoocytesinjectedwithequimolaramountofcRNA encodinghNa1.5,hNa1.5eor1:1combinationofNa1.5andNa1.5echannels.(c)Current–voltagerelationships.(d)Steady-stateactivationcurves. v v v v (e)Inactivationtimeconstantsth(ms)atdifferenttestpulses.(f)Steady-stateinactivationcurves.(g)Fractionalrecoverycurves.Datawereobtained from11differentbatchesofoocytes.Toillustratesteady-stateactivation,steady-stateinactivationandrecoveryfrominactivation,weused3–5 representativemeasurements.Fortotalnumberofmeasurements(n¼25–27)andforstatisticaldataevaluation(Vm,s)seetheSupplementaryTable2. Barsindicates.e.m.Studenttest,***indicatesPo0.001. contribution of SCN5A misregulation to the cardiac symptoms producedadeno-associatedvirus(AAV2/9)expressingoptimized observedinDM.Totest thephysiologicalimportanceof SCN5A U7-snRNA fused to Scn5a antisense sequences (U7-ASScn5a). splicingmisregulation,weartificiallyforcedtheswitchfromadult Splicing analysis revealed that combination of two U7-AS exon 6B towards fetal exon 6A into wild-type adult mouse heart constructs, spanning intron 6/exon 6B junction and exon using an exon-skipping strategy (Fig. 6a). To insure efficient 6B of Scn5A, promoted a switch from inclusion of adult exon transduction of the cardiac muscle and continuous expression 6B towards inclusion of the fetal exon 6A (Supplementary of nuclear antisense oligonucleotides, we engineered and Fig. 4A). Thus, AAV2/9 expressing both U7-AS constructs NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications 7 ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 a Fetal Adult 5 6A 6B 7 5 6A 7 Forced inclusion of Scn5A antisense exon 6A in adult sequences b c bp CTL U7-ASScn5a CTL 300 6B U7-ASScn5a 200 2 6A A1.5 100 N on 6A4600 *** mR0.51 Ex20 0 % 0 Scn5a Scn1b Gja1 d e CTL 50 20 0.058 50 P P *** 40 15 40 QRS QRS ms 30 10 30 U7-ASScn5a 20 20 P P 10 5 10 0 0 0 QRS QRS 25 ms PR QRS QT CTL U7-ASScn5a f g CTL 225 R150 R 75 CTL 0 U7-ASScn5a 0 0.5 1 1.5 2 225 R150 R 75 U7-ASScn5a 0 0 0.5 1 1.5 2 mn h i Control U7-ASScn5a CTL U7-ASScn5a 2 ** 1.5 * A RN 1 m 0.5 0 Cola1a Col3a1 Tgfb1 Figure6|AlterationofScn5asplicingcausesheartconductiondefectsandarrhythmias.(a)Schematicrepresentationofmutuallyexclusiveexons6A and6BofScn5aandofantisensesequencesdrivenbyoptimizedU7-snRNAs(U7-ASScn5a)toforcefetalexon6Ainclusioninadultwild-typemouseheart. (b,upper)RT–PCRanalysisofthealternativesplicingofendogenousScn5amRNAfromheartsamplesofmiceinjectedwithAAV2/9expressing U7-ASScn5acomparedwithcontrolinjectedmice.Molecularsizemarkerisindicatedinbp.(lower)PercentageofScn5amRNAincludingexon6A.(c)Real- timeRT-qPCRquantificationoftheexpressionofScn5a,Scn1bandGJja1(connexin43)mRNAsinheartsamplesofmiceexpressingU7-ASScn5a(n¼6) comparedwithcontrolinjectedmice(n¼6).(d)RepresentativeECGtracesshowprolongationofthePRintervalinU7-ASScn5a-injectedmicecompared withcontrolmice.(e)ECGmeasuresofPRinterval,QRSandQTintervalsin4-month-oldmiceinjectedwithAAV2/9expressingU7-ASScn5a(n¼25) comparedwithage-matchedcontrolmice(n¼17).(f)RepresentativeECGtracesrevealatrialfibrillationinU7-ASScn5a-injectedmicecomparedwith controlmice.(g)VariationoftheRRintervalindicatesevidencesofheartarrhythmiasinU7-ASScn5a-injectedmice(n¼25)comparedwithcontrolmice (n¼17).(h)RepresentativeimageofsixanalysedheartsamplesshowingmildfibrosisrevealedbyRedSiriushistologystaininginAAV-U7-ASScn5a-injected mice.Scalebar,100mm.(i)Real-timeRT-qPCRquantificationoftheexpressionofCola1a,Col3a1andTgfbmRNAsinheartofcontrol(n¼6)orAAV-U7- ASScn5a-injectedmice(n¼6).Barsindicates.e.m.Studenttest,*indicatesPo0.5,**indicatesPo0.01,***indicatesPo0.001.bp,basepair. (AAV-U7-ASScn5a) were injected systemically into newborn aconcomitant30–40%increaseoftheinclusionofexon6A,thus wild-type mice and cardiac functions were investigated 4 and 6 reproducingthesituationobservedinDM(Fig.6b).Quantitative monthspostinjection.Controlanimalsinjectedeitherwithsaline RT–PCR demonstrated no changes in the expression of Scn5a oremptyAAV2/9presentednosplicingalterationsofScn5aand mRNA or of its associated subunit Scn1b between control- normal cardiac functions. In contrast, mice injected with AAV- and AAV-U7-ASScn5a-injected mice (Fig. 6c). Importantly, U7-ASScn5apresentedadecreasedinclusionofadultexon6Bwith AAV-U7-ASScn5a-injected mice reproduce some of the key 8 NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 pathological features of DM, including conduction defects and predicted a change of the QRS duration from 72ms with heart arrhythmias. Indeed, ECG performed 4 months post control adult hNa 1.5to 88mswith fetalhNa 1.5e,hence a 22% v v injection revealed a significant prolongation of the PR intervals increase(Fig.7aandSupplementaryFig.5).Furthermore,wealso (Student t-test, P value r0.001) in AAV-U7-ASScn5a-injected tested extent of atrio-ventricular change50. Mathematical micecomparedwithcontrolinjectedmice(Fig.6d,e).Incontrast, simulation predicted a change of the atrium-His interval from QTintervalwasnotsignificantlyaltered,andweidentifiedonlya 81ms with control hNa 1.5 to 143ms with fetal hNa 1.5e v v trendtowardsincreasedQRSduration(Studentt-test,Pvalueof (Fig. 7b). Overall, these results support our mouse results and 0.058 with 8 AAV-U7-ASScn5a-injected mice on 25 presenting a provide additional evidences that misregulation of SCN5A QRS higher than 19ms versus 16.5ms in control mice) (Fig. 6e alternative splicing causes cardiac-conduction abnormalities, andSupplementaryFig.4B).Similarly,analysisofheartfunctions which is a key pathological feature of DM (Fig. 7c). in 6-month-old animals showed that AAV-U7-ASScn5a-injected mice present a consistent increase of the PR interval compared with control injected mice (40.5ms versus 34.8ms respectively; Discussion Student t-test, P value r0.05), without significant changes of Cardiacdefectsaffect80%ofindividualswithDMandrepresent theQRSandQTintervals(SupplementaryFig.4B).Ofinterest,a the second most common cause of death in this disease17,18. similarelongation of thePRinterval wasobserved inScn5aþ/(cid:2) However, the molecular mechanisms responsible for mice, which are hemizygote for Scn5a expression and represent cardiac-conduction delay and ventricular tachycardia in DM are an established model for cardiac-conduction disease45–47. unclear. Using RNA sequencing we identified various novel Furthermore, ECG analyses also revealed that 44% of splicing misregulation events in DM1 heart samples. Among AAV-U7-ASScn5a-injected mice develop significant (Student these changes, the splicing switch from adult exon 6B to fetal t-test, Po0,001) heart arrhythmia at 4 months post injection exon 6A in SCN5A mRNA is of particular interest. Previous withanaverageoffivearrhythmicevents,definedasvariationof studies39–41 as well as ours indicate that hNa 1.5e, the splicing v the RR interval, per minute whereas control injected animals variant of SCN5A found in DM and that contains the fetal exon showed no alterations (Fig. 6f,g). We did not detect 6A, possesses a reduced excitability compared with the normal ventricular fibrillations or second and third-degree heart adult splicing form of SCN5A containing the exon 6B. blocks in any injected animals. In contrast, we observed Consequently, the switch from the hNa 1.5 to the hNa 1.5e v v supraventricular premature contractions and atrial fibrillation in channel in DM may cause a slower upstroke velocity of the AAV-U7-ASScn5a-injected mice (Fig. 6f), and five of these cardiac action potential, leading to conduction slowing. injected mice died suddenly between 4 and 6 months post Importantly, this hypothesis is supported by mathematical injections (none of the control mice died). These electrical simulation as well by animal model, since imposing a switch alterations were specific and not caused by global cardiac from inclusion of the control adult exon 6B towards using the remodelling since we observed neither systolic nor diastolic fetal exon 6A of Scn5A in adult mouse heart led to alterationsbydopplerechocardiography(SupplementaryTable3) cardiac-conduction delay and heart arrhythmias, two key and no change in heart/body weight ratio (4.4±0,1mgg(cid:2)1 features of DM. Moreover, clinical evidence also supports an in control, n¼9, versus 4,7±0,2mgg(cid:2)1 in AAV-U7-ASScn5a- alteration of the sodium current in DM. Indeed, the injected animals, n¼14). As further control, H&E-staining electrophysiological features39–41 of the fetal isoform of SCN5A revealed normal heart structures with no evident expressed in DM are similar to the electrophysiological cardiomyopathy or dilation at 6 months post AAV injections characteristics observed with loss-of-function mutations of (Supplementary Fig. 4C). Similarly, quantitative RT–PCR SCN5A causing cardiac-conduction disease51–53. Also, there are experiments show no alteration in the expression levels of somesimilaritiesofECGrecording,includingprolongationofthe Nppa, Nppb (encoding Anp and Bnp, respectively) and PR interval and of the QRS duration, between individuals with Myh7 mRNAs (Supplementary Fig. 4D), suggesting no overt DM and individuals affected by cardiac-conduction disease cardiacremodellinginantisenseAAV-U7-ASScn5a-injectedmice. caused by loss-of-function mutations in SCN5A42,54. Finally, the Moreover,SiriusRedstainingconfirmednormalheartstructures inductionof abnormalECGpatterninDMpatientstreatedwith but also revealed some mild fibrosis (Fig. 6h), which was ajmaline55,56,aclassIaantiarrhythmicagentactingonthecardiac confirmed by increased expression of collagen Cola1a and Tgfb1 sodium channel andtheabnormalsodium currentobserved ina mRNAs (Fig. 6i). Interestingly, mild fibrosis is also observed in mousemodelofDM57,arealsoevocativeofadysfunctionofthe DM cardiac samples17,18, as well as in individuals and mice sodium channel in DM. Overall, our results suggest that models with loss-of-function mutations of the SCN5A misregulation of the splicing of SCN5A participates in a subset gene20,21,46,47. Overall, heart arrhythmias and prolonged PR of electrical cardiac alterations observed in DM, namely the interval in AAV-U7-ASScn5a-injected animals demonstrate that cardiac-conductiondelayandtheheartarrhythmias.However,it inclusion of the fetal exon 6A of Scn5a is inappropriate to adult is likely that other alternative splicing alterations and/or mouse heart physiology. However, while we found a clear mechanisms58–61 are participating to the full pattern of cardiac elongation of the PR interval, we did not detect a significant alterations in DM since knockout of Mbnl1 and Mbnl2 in mice alteration of the QRS duration as only a third of AAV-U7- leads to only mild alteration of Scn5A splicing, while these mice ASScn5a-injectedmicepresentincreasedQRSduration(419ms). show severe conduction disease and cardiac dilatation43,44. Interestingly, similar findings have been described in Scn5aþ/(cid:2) In conclusion, this work may also have some clinical mice, which all show elongation of the PR interval, while only a importance such as considering with caution the treatments of subset of Scn5aþ/(cid:2) animals present a prolongation of the QRS DM patients with pharmaceutical agents that reduce the activity interval45. Hence, elongation of the PR interval is not of the cardiac sodium channel, including mexiletine, flecainide systematically associated with increased duration of the QRS in and other antiarrhythmic drugs of class I. In that aspect, this mousemodelofScn5adysfunction.Thus,tostrengthenourdata, studymayprovideamolecularexplanationtotheadversecardiac we mathematically tested whether human cardiac parameters reactionofsomepatientswithmyotonicdystrophictotreatment wouldbealteredbytheelectrophysiologicaldifferencescausedby withdrugsreducingactivityofSCN5A(refs62,63). Involvement the switch from adult exon 6B towards fetal exon 6A of SCN5A. of the cardiac sodium channel in DM might also highlight the Simulation based on a modified O’Hara-Rudy model48,49 importance ofconsideringpolymorphismin theSCN5A gene,as NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications 9 ARTICLE NATURECOMMUNICATIONS|DOI:10.1038/ncomms11067 a b Atrium hNav1.5e hNav1.5e V ECG m 0 0 1 200 ms hNa1.5 v Nodal-His hNa1.5e v APs Endo M hNav1.5 Epi hNa1.5e v 200 ms APs c G CUG Control Myotonic dystrophy U MBNL1C C U – + UG GC C MBNL1 U 5 MBNL1 6A 6B 7 5 MBNL1 6A 6B 7 UG GC Adult Fetal UGC MBNL1 UGC C U G MBNL1 G U C I II III IV I II III IV 12345 6 12345 6 12345 6 12345 6 12345 6 12345 6 12345 6 12345 6 hNa1.5 hNa1.5e v v Normal adult heart conduction Conduction delay and heart arrhythmia Figure7|ComputersimulationpredictscardiacconductionabnormalitiesinDM.(a)SimulationsofECG(upper)andactionpotential(AP)alterations (middleandlower)causedbytheswitchfromadultSCN5Aexon6Btowardsfetalexon6A,usingamodifiedhumanventricularORdmodel.(b)Simulations oftheatrio-ventricularchangescausedbyinclusionofSCN5Afetalexon6Ainsteadofadultexon6B,employinganatrio-ventricularnodalmodel.APsof atrialandAVnodecellsweresimulatedandatrium-HisintervalwasmeasuredasthedifferenceinthelatencyofAPsbetweenatrialandnodal-Hiscells. (c)Modelofsplicingalterationofthecardiacsodiumchannel,SCN5A,inDM.MBNLproteinsregulatetheswitchfromSCN5Aexon6Ainfetalheartto exon6Binadult.InDM,titrationofMBNLproteinsbymutantRNAcontainingexpandedCUGrepeatsleadstoexpressionofafetalsplicingformofSCN5A, inappropriatetoadultheartphysiology,ultimatelyresultingincardiac-conductiondelayandheartarrhythmias,whicharetwokeysfeaturesofDM. wellasinothergenessuchasSCN10A,asapossiblecauseofthe alteration of the alternative splicing of SCN5A, one may hope clustering of cardiac alterations in some families of DM64, an that correction of splicing misregulations through approaches hypothesis supported by the recent report that DM exacerbates releasing MBNL1 from CUG expanded repeats or through Brugada syndrome caused by a mutation in SCN5A (ref. 65). approaches reducing the expression of pathogenic CUG Inversely, it remains to be tested whether mutations in intronic RNA66,67 may alleviate the cardiac symptoms of DM. regions regulating alternative splicing of SCN5A exon 6A/6B might be considered as a cause of cardiac-conduction Methods abnormalities in patients in whom no mutations were identified Humansamples.Allsampleswereheartleftventriclesthatweresampledwiththe in the coding sequence of SCN5A. Also, if dysfunction of the informedconsentofindividualsandapprovedbytheInstitutionalReviewBoardof cardiacsodiumchannelinDMiscomparableinsomeaspectsto thePitie´-Salpeˆtrie`rehospital,oftheNeuromuscularResearchCenteroftheTam- cardiac-conduction disease caused by mutations in SCN5A, an pereUniversityHospital,oftheHospitalDonostiaandoftheToneyamaNational Hospital.Non-affectedheartsamples(CTL#1to#3)werepurchasedatAmbion attractive supposition would be that therapeutic approaches andStratagene,respectively.DCM#1and#2werepatientssufferingfromDCMof developedforthesediseasescouldbeconsideredforDM.Finally, uncharacterizedgeneticorigin.DMD#1to#4werepatientsaffectedwithDMD. if cardiac-conduction defects in DM are associated with an ALS#1to#3werepatientswithALSdescribedpreviously68.DM1samples#1and 10 NATURECOMMUNICATIONS|7:11067|DOI:10.1038/ncomms11067|www.nature.com/naturecommunications