Effects of cardioactive drugs on human induced pluripotent stem cell derived long QT syndrome cardiomyocytes

Kuusela e al. Sp inge Plus (2016) 5:234 DOI 10.1186/s40064-016-1889-y RESEARCH E ec s o ca dioac i e d ugs onhuman induced plu ipo en s em cell de i ed long QT synd ome ca diomyocy es Jukka Kuusela1†, Ville J. Kujala1,2†, Anna Ki iaho1, Ma isa Ojala1, Heikki Swan3, Kimmo Kon ula3 and Ka iina Aal o‑Se älä1,4,5* Abs ac Human induced plu ipo en s em cells (hiPSC) ha e enabled a majo s ep o wa d in pa hophysiologic s udies o inhe i ed diseases and may also p o e o be aluable in in i o d ug es ing. Long QT synd ome (LQTS), cha ac e ized by p olonged ca diac epola iza ion and isk o sudden dea h, may be inhe i ed o esul om ad e se d ug e ec s. Using a mic oelec ode a ay pla o m, we in es iga ed he e ec s o six di e en d ugs on he elec ophysiological cha ac e is ics o human emb yonic s em cell‑de i ed ca diomyocy es as well as hiPSC‑de i ed ca diomyocy es om con ol subjec s and om pa ien s wi h ype 1 (LQT1) and ype 2 (LQT2) o LQTS. A baseline he epola iza ion ime was signi ican ly longe in LQTS cells compa ed o con ols. Isop enaline inc eased he bea ing a e o all cell lines by 10–73 % bu did no show any a hy hmic e ec s in any cell ype. Di e en QT‑in e al p olonging d ugs caused p olonga ion o ca diac epola iza ion by 3–13 % (cisap ide), 10–20 % (e y h omycin), 8–23 % (so alol), 16–42 % (qui‑ nidine) and 12–27 % (E‑4031), bu we did no ind any sys ema ic di e ences in sensi i i y be ween he con ol, LQT1 and LQT2 cell lines. So alol, quinidine and E‑4031 also caused a hy hmic bea s and bea ing a es s in some cases. In summa y, he d ug e ec s on hese pa ien ‑speci ic ca diomyocy es appea o ecapi ula e clinical obse a ions and p o ide u he e idence ha hese cells can be applied o in i o d ug es ing o p obe hei ulne abili y o a hy hmia. Keywo ds: Induced plu ipo en s em cell, Pa ien ‑speci ic, Long QT synd ome, Ca diomyocy es, Mul ielec ode a ay, Ca dioac i e d ug, A hy hmia © 2016 Kuusela e al. This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. Backg ound Inhe i ed long QT synd ome (LQTS) is a po en ially se e e a hy hmic disease ha a ec s he elec ical epo- la iza ion o he myoca dium and mani es s as an abno - mally long QT in e al on elec oca diog am (ECG) eco dings. LQTS may esul in polymo phic en icula achyca dias known as o sades de poin es (TdP), which can ul ima ely lead o li e- h ea ening en icula ib illa- ion and sudden ca diac dea h (Schwa z e al. 2013). Inhe i ed o ms o LQTS ypically esul om mu a- ions in ca diac ion channel coding genes. In he case o LQT ype 1 (LQT1), he mu a ions a e p esen in he KCNQ1 gene which encodes he α-subuni o he slow componen o he delayed ec i ie po assium cu - en (K 7.1) (Wang e al. 1996), whe eas LQT2 is caused by mu a ions o he KCNH2 gene which encodes he α-subuni o he apid componen o he delayed ec i ie po assium cu en (K 11.1) (Sanguine i e al. 1995). Fou ounde mu a ions in hese wo genes ha e been ound o explain he high (0.4%) p e alence o LQTS in he Finn- ish popula ion (Ma jamaa e al. 2009). The mos p e a- len o hese ounde mu a ions is he C- e minal KCNQ1 G589D missense mu a ion, esul ing in a dys unc ional assembly domain and p e en ing he o ma ion o ully unc ional po assium channel e ame s (Piippo e  al. Open Access *Co espondence: ka iina.aal o‑[email p o ec ed] †Jukka Kuusela and Ville J. Kujala con ibu ed equally o his wo k 1 BioMediTech, Uni e si y o Tampe e, Finn‑Medi 5, Bioka u 12, 33014 Tampe e, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle Page 2 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 2001). Ano he p e alen bu appa en ly milde ounde mu a ion, he KCNH2 R176W missense mu a ion, esides in he N- e minus o he channel and accele a es he deac i a ion o he IK ail cu en , hus leading o educed po assium e lux (Fods ad e al. 2006). Resea ch on he exac e ec s o hese mu a ions on he unc ional p ope ies o ca diomyocy es has been hampe ed by he lack o sui able echniques ha can adequa ely ecapi u- la e he ull spec um o he disease in i o. Human induced plu ipo en s em cell (hiPSC) ech- nology has ecen ly made i possible o p oduce human cell lines wi h cha ac e is ics simila o hose o emb y- onic s em cells om soma ic cells o a ious o igins (Takahashi e al. 2007; Yu e al. 2007). To da e, mul iple disease-speci ic hiPSC lines ha e al eady been es ab- lished (Wu and Hochedlinge 2011). Se e al s udies ha e ocused on LQTS using he hiPSC echnology (Bellin e al. 2013; I zhaki e al. 2011; Ki iaho e al. 2015; Lah i e al. 2012; Ma sa e al. 2011, 2014; Mo e i e al. 2010). These s udies sugges ha hiPSC echnology may be use- ul o modeling LQTS in i o. Un il now, la ge-scale d ug e ec s ha e been s udied wi h human emb yonic s em cell-de i ed ca diomyocy es (hESC-CMs) and wild- ype hiPSC-CMs (B aam e al. 2010; Caspi e al. 2009; Ha is e al. 2013; Meh a e al. 2013; Na a e e e al. 2013; Yokoo e al. 2009). In his s udy, we in es iga ed he e ec o six clinically used d ugs in con ol hESC- and wild- ype hiPSC-CMs and in hiPSC-de i ed LQT-CMs. Addi- ional alue will be gained by s udying d ug esponses a clinically ele an concen a ions (Schulz and Schmold 2003), since many o he mu a ion ca ie s can ha e a la en , subclinical o m o LQTS ha may be exace ba ed upon pha macological challenge (Roden 2004). We c ea ed hiPSC lines om he de mal ib oblas s o pa ien s wi h KCNQ1 G589D and KCNH2 R176W mu a- ions. We in es iga ed how hese pa ien -speci ic hiPSCs espond o pha macological challenges in compa ison wi h ca diomyocy es de i ed om con ol hESCs and hiPSCs. Me hods Pa ien ‑speci ic plu ipo en s em cell induc ion The LQT1- and 2-speci ic hiPSCs we e induced as p e i- ously desc ibed (Takahashi e al. 2007) om ib oblas s o pa ien s ca ying he KCNQ1 (G589D) o he HERG (R176W) mu a ion (Fods ad e  al. 2004) (see Addi- ional ile1: Ma e ials and me hods). The s udy has been app o ed by Pi kanmaa Hospi al Dis ic e hical com- mi ee (R08070). De ails o he hiPSC lines a e lis ed in Table1. The subjec s olun ee ed o he s udy ga e hei consen o skin biopsy o be aken. S em cell cul u e The con ol plu ipo en s em cell lines used in his s udy we e H7 (hESC, WiCell), UTA.00112.hFF [ e al wild- ype om ATCC (WT)-hiPSC], and UTA.01006.WT (WT- hiPSC de i ed om a heal hy adul ). Fo LQT1, 4 pa ien - speci ic hiPSC lines we e used: UTA.00208.LQT1, UTA.00211.LQT1, UTA.00313.LQT1 and UTA.00303. LQT1. Fo LQT2 2 pa ien -speci ic hiPSC lines we e used: UTA.00514.LQT2 and UTA.00525.LQT2. The plu ipo en s em cells we e cul u ed on a mouse emb y- onic ib oblas (MEF, Millipo e) eede cell laye in KSR medium consis ing o KnockOu Dulbecco’s Modi ied Eagle’s Medium (KO-DMEM, In i ogen) supplemen ed wi h 20% KnockOu se um eplacemen (KO-SR, In i - ogen), 1% non-essen ial amino acids (NEAA, Lonza), 1% Glu amax (In i ogen), 50U/mL penicillin/s ep o- mycin (Lonza), and ecombinan human basic ib oblas g ow h ac o (R&D Sys ems). The cells we e passaged once a week by ea ing he plu ipo en s em cell colo- nies wi h collagenase IV (Gibco) and seeding hem on o a esh MEF eede laye . Table 1 De ails o  he con ol andpa ien -speci ic plu ipo en s em cell lines used o d ug es ing ICM inne cell mass o blas ocys , hFF human o eskin ib oblas , hADF human adul de mal ib oblas , hESC human emb yonic s em cell, hiPSC human induced plu ipo en s em cell *Siblings ‡,θ,ψ clonal lines om same indi idual ssymp oma ic aasymp oma ic Cell line Line ype Mu a ion Disease A ec ed cu en Sou ce Pu pose H7 (hESC) hESC Wild ype None None ICM hESC con ol UTA.00112.hFF (WTa) hiPSC Wild ype None None hFF Con ol UTA.01006.WT (WTb) hiPSC Wild ype None None hADF Adul con ol UTA.00208.LQT1θ,*,s hiPSC G589D LQT1 IKs hADF Symp oma ic, pa ien ‑speci ic UTA.00211.LQT1θ,*,s hiPSC G589D LQT1 IKs hADF Symp oma ic, pa ien ‑speci ic UTA.00303.LQT1ψ,*,a hiPSC G589D LQT1 IKs hADF Asymp oma ic mu a ion ca ie ‑speci ic UTA.00313.LQT1ψ,*,a hiPSC G589D LQT1 IKs hADF Asymp oma ic mu a ion ca ie ‑speci ic UTA.00514.LQT2‡,a hiPSC R176W LQT2 IK hADF Asymp oma ic mu a ion ca ie ‑speci ic UTA.00525.LQT2‡,a hiPSC R176W LQT2 IK hADF Asymp oma ic mu a ion ca ie ‑speci ic Page 3 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 Ca diomyocy e di e en ia ion Ca diomyocy es we e di e en ia ed om hESCs as well as om he con ol and LQT1 and -2-speci ic hiPSCs using he mouse isce al endode m-like (END-2) cell co- cul u e me hod, as desc ibed elsewhe e (Mumme y e al. 2003). B ie ly, mi omycin C (Sigma) -inac i a ed END-2 cells we e ypsinized and pla ed on o 12-well pla es in 0% KO-SR hES medium. The plu ipo en s em cells we e pla ed on op o he END-2 cell monolaye he ollowing day. The 0% KO-SR hES medium was e eshed on days 5, 8 and 12. A day 14, he medium was changed o 10% KO-SR hES medium, and i was e eshed h ee imes a week he ea e . Pha macological es ing D ugs Cisap ide monohyd a e, quinidine and (±)-so alol hyd o- chlo ide we e ob ained om Sigma-Ald ich and dis- sol ed in dime hyl sul oxide (Sigma-Ald ich) a 10mM. E-4031 (Alomone Labs) and e y h omycin (Abbo icin i. ., Amdipha m) we e dissol ed in s e ile H2O a 1 and 50mg/mL, espec i ely. Isop enaline (Isup el i. ., Hos- pi a) was supplied in eady- o-use ampoules. The d ug concen a ions we e chosen based on hei he apeu ic blood se um concen a ion ange (Schulz and Schmold 2003). The ollowing concen a ions o cisap ide mono- hyd a e, quinidine, (±)-so alol hyd ochlo ide, and e y h- omycin we e es ed: hal he concen a ion o he lowe limi o he he apeu ic blood se um concen a ion, he lowe limi o he he apeu ic blood se um concen a- ion ange, he a e age he apeu ic se um concen a- ion, he uppe limi o he he apeu ic ange, and wice he uppe limi o he he apeu ic concen a ion (Schulz and Schmold 2003). The isop enaline concen a ions we e chosen acco ding o p e iously published alues and mul iples he eo (Pekkanen-Ma ila e  al. 2009). The d ugs we e se ially dilu ed in emb yoid body (EB)- medium wi h 5% e al bo ine se um (FBS, see below). The es concen a ions o each d ug a e lis ed in Table2. Mul ielec ode a ay eco dings The hESC-CMs and he con ol, LQT1-, and LQT2- speci ic hiPSC-CM agg ega es we e mechanically excised om he cul u es and pla ed on o e al bo ine se um (PAA) and 0.1% gela in- (Sigma-Ald ich) coa ed 6-well-MEAs (6wellMEA200/30iR-TI-m , Mul iChan- nel Sys ems MCS GmbH). The FPs we e eco ded wi h a USB-MEA1060 ampli ie , and he empe a u e was kep a +37°C using a TC02 hea ing elemen (bo h om Mul i Channel Sys ems MCS GmbH). The MEAs we e co - e ed wi h gas-pe meable MEA memb anes (ALA MEA- SHEET, ALA Scien i ic) du ing he eco dings. A wash-in ime o 2min was allowed o he d ugs be o e eco d- ings. The bea ing a eas we e cul u ed in EB medium consis ing o KO-DMEM (In i ogen) supplemen ed wi h 20% FBS (In i ogen), 1% NEAA (Lonza), 1% Glu- amax (In i ogen), and 50U/mL penicillin/s ep omycin (Lonza). The d ug es s we e pe o med in 5% FBS con- aining EB medium; he medium was changed a leas an hou p io o he baseline eco dings, and he cells we e allowed o s abilize in he incuba o (+37°C, 5% CO2). The ield po en ial signals we e eco ded wi h MC_Rack .4.0.0 so wa e (Mul i Channel Sys ems MCS GmbH). Da a ands a is ical analysis The ield po en ial du a ions (FPDs) om he eco ded iles we e measu ed o line using ou in-house de el- oped Ca dioMDA so wa e (P adhapan e al. 2013) and AxoScope10 (Molecula De ices). The baseline da a a e p esen ed as mean±s anda d de ia ion (SD). Lin- ea mixed-e ec models wi h ei he Baze ’s co ec ed ield po en ial du a ion (cFPD) o bea ing a e (BR) as a dependen a iable we e i ed using he unc ion lme in R (So wa e en i onmen o s a is ical compu ing and g aphics, e sion 2.13.0, The R Founda ion o S a is ical Compu ing). The ac o ial a iable o he concen a ion le el was used as an independen a iable. The lowes concen a ion le el was used as a e e ence g oup. Ran- dom in e cep o di e en agg ega es was used oge he wi h independen andom e o s. Resul s LQT pa ien cha ac e is ics Skin biopsies we e ob ained om wo siblings, a 28-yea old emale asymp oma ic LQT1 pa ien (QTc in e al, 428ms) and a 41-yea old emale symp oma ic LQT1 pa ien (QTc in e al, 456 ms), bo h wi h he KCNQ1 G589D mu a ion. In addi ion, skin biopsies we e ob ained om a 61-yea old male asymp oma ic LQT2 pa ien wi h he KCNH2 R176W mu a ion (QTc in e - al, 437ms). The symp oma ic LQT1 pa ien had expe- ienced seizu es be o e be a-blocke medica ion, while he asymp oma ic sibling has ne e had any ca diac Table 2 D ugs and hei concen a ions used wi hca dio- myocy es de i ed om human emb yonic and induced plu ipo en s em cells D ug Concen a ions Isop enaline nM: 81, 402 Cisap ide monohyd a e nM: 40, 83, 120, 165, 330 E y h omycin µM: 1.5, 3, 5.5, 8, 16 So alol µM: 0.8, 1.6, 5.7, 9.7, 19.4 Quinidine µM: 1.5, 3, 9, 15, 30 E‑4031 nM: 10, 100, 300, 500, 700 Page 4 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 symp oms. The asymp oma ic LQT2 indi idual has only expe ienced occasional palpi a ions, which al hough a e no speci ic symp oms o LQTS. Molecula cha ac e is ics o ca diomyocy es Re e se- ansc ip ion PCR and immuno luo escence mic oscopy showed ha he con ol and LQT-hiPSCs exp essed he plu ipo ency ma ke s NANOG, OCT3/4, SOX2, SSEA4, TRA-1-60, and TRA-1-81 (Addi ional ile1: Fig. S1A, Table S1). Fu he mo e, he hiPSC colony mo phology esembled hose o hESCs, he hiPSCs di - e en ia ed in o ca diomyocy es a e iciencies p e iously epo ed o hESCs upon END-2 co-cul u e (da a no shown) (Pekkanen-Ma ila e al. 2009), showed nuclea localiza ion o he ca diac-speci ic ansc ip ion ac o NKX2.5 which is also exp essed in hESC-CMs (Asp e al. 2010), and exhibi ed a dis inc s ia ed pa e n in ca - diac oponin T s aining (Addi ional ile1: Fig. S1B). The elec ophysiological p ope ies and d ug esponses we e eco ded wi h mic oelec ode a ays (MEAs). E ec s o ca dioac i e d ugs o he cFPD o hiPSC‑CMs We compa ed he d ug esponses o he con ol ca dio- myocy es de i ed om hESC line H7 (he ea e called hESC) and wo wild- ype hiPSC lines UTA.00112.hFF (he ea e called WTa) and UTA.01006.WT (he ea e called WTb) o he hiPSC-de i ed LQT-CMs ( o he cell line p ope ies, see Table1). The esul s o he clonal cell lines de i ed om he same indi idual we e pooled oge he (UTA.00208.LQT1 and UTA.00211.LQT1 [=LQT1A], UTA.00303.LQT1 and UTA.00313.LQT1 [=LQT1B], UTA.00514.LQT2 and UTA.00525.LQT2 [=LQT2]). LQT1A we e de i ed om a symp oma ic pa ien , LQT1B om an asymp oma ic mu a ion ca ie and LQT2 om asymp oma ic mu a ion ca ie . Baseline cha ac e is ics The baseline bea ing a es (BR) showed la ge di e ences be ween cell popula ions, anging om 59 o 105 bea s pe minu e (Fig. 1a). Howe e , hese BRs did no di - e signi ican ly among ca diomyocy es di e en ia ed om hESCs, con ol-, o LQT-hiPSCs. In con as , he bea ing a e-co ec ed ield po en ial du a ion (cFPD, Baze ´s) showed signi ican di e ences be ween con ol- and LQT-CMs (Fig.1b). A baseline le el, he LQT-CMs showed signi ican ly longe cFPD (586–600 ms) com- pa ed o con ol-CMs (340–400ms) (p<0.001). Isop enaline e ec s To es whe he he di e en ia ed ca diomyocy es had p ope β-ad ene gic esponses, we subjec ed hem o an isop enaline challenge. This ea men had ch ono opic e ec on ca diomyocy es and inc eased he BR a he concen a ion o 81 and 402nM (Fig.2). Isop enaline did no induce any a hy hmic bea s in any o he CMs, includ- ing hose de i ed om LQT1- and LQT2-hiPSC lines. E ec s o non‑ca diac d ugs cisap ide ande y h omycin Cisap ide inc eased he epola iza ion ime in all CMs s udied, al hough his e ec did no each s a is ical sig- ni icance in he LQT2-CMs, appa en ly due o la ge expe imen al a ia ion (Fig.3a). E y h omycin likewise induced a signi ican a e age inc ease o cFPD in all cell lines s udied, in a concen a- ion ange o 1.5–16µM (Fig.3b). No a hy hmic bea s o bea ing a es s we e obse ed wi h cisap ide o wi h e y h omycin. E ec s o  he an i‑a hy hmic d ugs so alol andquinidine Upon so alol (0.8–19.4µM) challenge, he maximal cFPD p olonga ion anged om o 8% (WTb-CMs) o 23% (LQT1A-CMs) (Fig.4a). Quinidine also caused a concen- a ion-dependen (1.5–30µM) a e age cFPD p olonga- ion in all he ca diomyocy es (Fig.4b). O all clinically used d ugs es ed (cisap ide, e y h omycin, so alol, qui- nidine), quinidine showed he la ges cFPD-inc easing e ec (31–42%) seen in LQT1-CMs. The e ec o  he hERG channel blocke E‑4031 E-4031 also caused concen a ion-dependen cFPD p o- longa ion in all ca diomyocy e ypes (Fig.4c). I is o no e ha in hESC-CMs all bu one o he ca diac agg ega es became a hy hmic o s opped bea ing a concen a ions o 500–700 nM. Thus, e o ba s indica ing s anda d de ia ion could no be calcula ed o hese expe imen al poin s. The cFPD p olonga ion in he di e en cell lines eached s a is ical signi icance a E-4031 concen a ion o 100nM indica ing ha he sensi i i ies o he di e en cell lines o E-4031 we e e y simila (Fig.4c). In summa y, Fig.5 illus a es he ep esen a i e aces o cisap ide, e y h omycin, so alol, quinidine and E-4031 in con ol- and LQT-CMs. The p oa hy hmic e ec s o so alol, quinidine andE‑4031 So alol, quinidine and he hERG blocke E-4031 showed p oa hy hmic e ec s on ce ain indi idual cell lines (Table3). Wi h so alol, only spon aneous a e depola iza- ion (AD)-like wa e o m abno mali ies (bu no bea ing a es s) we e obse ed (Table3). So alol caused AD-like wa e o ms anging om concen a ions 5.7 µM (a e - age he apeu ic se um concen a ion) o 19.4µM ( wice he uppe he apeu ical se um concen a ion). None o he LQT1B-CMs showed any a hy hmic beha io upon so alol challenge. Quinidine showed also bea ing a es s in addi ion o AD-like wa e o m abno mali ies (Table3). The AD-like Page 5 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 wa e o ms we e obse ed a he he apeu ical se um concen a ion ange (3–15µM) in hESC-, LQT1A- and LQT2-CMs bu no in WT- o LQT1B-CMs (Table3). On he o he hand, he bea ing a es s occu ed anging om 9µM (a e age he apeu ic se um concen a ion) o 30µM ( wice he uppe he apeu ical se um concen a- ion) (Table3). The compound E-4031 also caused AD-like wa e- o m abno mali ies and bea ing a es s (Table 3). E-4031 caused AD-like wa e o m abno mali ies in ca - diomyocy es di e en ia ed om all o he lines excep WTb-CMs. Fu he mo e, bea ing a es s we e seen in all ca diomyocy e ypes excep LQT1A-CMs. Among he LQTs, he highes equency o AD-like wa e o m abno mali ies was ound in LQT1A-CMs (50% o ca dio- myocy e agg ega es). On he o he hand, he highes e- quency o bea ing a es s we e seen in LQT2-CMs (31% o ca diomyocy e agg ega es). Discussion The p esen s udy u he demons a es he use ulness o he hiPSC-de i ed ca diomyocy es o se e as a model Fig. 1 Baseline ca diac pa ame e s. a The bea ing a e (BR) o he ca diomyocy e clus e s di e en ia ed om each human induced plu ipo en s em cell (hiPSC) line. b The a e‑co ec ed ield po en ial du a ion (cFPD) o he same ca diomyocy e clus e s. Signi icance le el is indica ed by ***p < 0.001 Page 6 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 o long QT synd ome, showing ha he QT in e al- p olonging e ec s o clinically used d ugs a e eplica ed in his in i o sys em, using concen a ions ha a e wi hin o close o hei he apeu ic ange (Schulz and Schmold 2003). We showed ha LQT-CMs co ec ed ield po en ial du a ion is signi ican ly longe han in con- ol-CMs a baseline condi ions, which sugges s ha long QT synd ome is ecapi ula ed in hese cells. This s udy also ep oduces in cell clus e s he ea lie obse a ion wi h single cells by pa ch clamp wi h p olonged APDs in LQT-CMs compa ed o con ols (Ki iaho e al. 2015; Lah i e al. 2012). In he cu en s udy LQT- and con ol CMs we e u he exposed o a ious ca dioac i e d ugs o in es iga e hei e ec s. The ca diomyocy es bea ing a es we e inc eased by isop enaline, which is a β-ad ene gic agonis ha a ec s calcium (ICa,L) channel ac i a ion ia β1 ecep o s. This is in line wi h he p e ious s udies showing posi i e ch ono opic e ec o isop enaline on bo h hESC- and hiPSC-CMs (Mo e i e al. 2010; Pekkanen-Ma ila e al. 2009; Yokoo e al. 2009). β-ad ene gic s imula ion wi h isop enaline did no ha e any a hy hmic e ec s on he LQT1 lines al hough LQT1 pa ien s o en expe ience symp oms a inc eased hea a es. Cisap ide is a se o onin 5-HT4 ecep o agonis ha has been epo ed o block he hERG po assium chan- nel (Mohammad e al. 1997). I is known o induce TdP in pa ien s (Da pö 2001). Caspi e  al. (2009) epo ed signi ican (>50 nM) dose-dependen p olonga ion o cFPD in hESC-CMs (B aam e al. 2010; Caspi e al. 2009). We ob ained simila esul s as Caspi e al. (2009) as he hESC-CMs used in ou s udy showed dose-dependen and signi ican cFPD p olonga ion s a ing a 83 nM. In con as , B aam e al. (2010) epo ed no signi ican cisap ide-induced cFPD p olonga ion in hESC-CMs. Howe e , hey judged he he apeu ic concen a ion o be 2.6–4.9nM (Red e n e al. 2003), whe eas we chose he he apeu ic ange as 83–165nM (Schulz and Schmold 2003), which may explain he ac ha hey we e no able o de ec signi ican cisap ide-induced cFPD p olonga- ion. Meh a e al. (2013) ha e also in es iga ed he e ec o cisap ide o he cFPD o 150–170days old EB-based hiPSC-de i ed CMs. Likewise, hey also ound con- cen a ion-dependen cFPD p olonga ion (Meh a e  al. 2013). Meh a e al. (2013) did no obse e any a hy h- mias wi h cisap ide, which is in line wi h ou s udy he e. E y h omycin is a mac olide an ibio ic ha p o- longs he QT in e al in pa ien s (Sha e e al. 2002). The he apeu ic plasma concen a ion o e y h omycin has been epo ed o be 3–8µM (Schulz and Schmold 2003). In ou s udy, e y h omycin p olonged signi ican ly cFPD in hESC-CMs al eady a he i s es concen a- ion (1.5 µM). Ou da a may sugges a sligh ly highe sensi i i y o LQT2-CMs compa ed o LQT1-CMs, bu high expe imen al a ia ion p ecludes i m conclusions. E y h omycin did no induce any a hy hmic e ec s in he ca diomyocy es. So alol is a β-ad ene gic ecep o an agonis and a Vaughan–Williams class III an i-a hy hmic agen ha po en ly blocks he po assium ion channels in Fig. 2 Isop enaline esponse o he con ol and long QT synd ome (LQT) ype 1 and 2‑speci ic ca diomyocy e clus e s (hESC, WTa, WTb, LQT1A, LQT1B and LQT2). Da a is p esen ed mean ± SD. Mean bea ing a es (BR) we e no malized o baseline BR. As e isks ma k he s a is ical signi icance o mean BR inc ease compa ed o baseline alues. *p < 0.05, **p < 0.01, ***p < 0.001 Page 7 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 ca diomyocy es (Ed a dsson e al. 1980). In hESC-CMs, B aam e  al. (2010) epo ed cFPD p olonga ions o 15–20% a so alol concen a ions o 1.8–14µM (B aam e al. 2010). In a subsequen s udy, hey also obse ed 20% cFPD p olonga ion in hESC-CMs a so alol con- cen a ion o 10 µM (B aam e  al. 2013). Ou s udy esul ed in simila da a. A concen a ions o 6–10µM, we obse ed a cFPD p olonga ion o 12–18 % in he hESC-CMs, inc easing o 21% a 19µM. We obse ed ha so alol inc eased cFPD in all he CMs a he he a- peu ical concen a ion ange (1.6–9.7µM). In line wi h ou da a, Meh a e al. (2013) epo ed simila obse a- ions om so alol-induced cFPD p olonga ion in hiPSC- de i ed CMs (Meh a e al. 2013). Fu he mo e, acco ding o ou da a so alol signi ican ly inc eased cFPD a ea lie concen a ions in LQT-CMs han in WTa o WTb. This may sugges ha LQT-CMs may be mo e ulne able o he cFPD-p olonging e ec o so alol han hiPSC-de i ed con ol cells. Quinidine is a class Ia an ia hy hmic d ug ha p ima - ily blocks sodium ion channels a high concen a ions and hERG channel a low concen a ions and is associ- a ed wi h TdP (Bauman e  al. 1984). Quinidine p o- longed he cFPDs in all o ou cell lines anging om 16 o 42%. B aam e al. (2010) epo ed a p olonga ion o 20–50% in hESC-CMs wi h quinidine concen a ions o up o 3.2µM and inc eases in he cFPD a highe con- cen a ions, and Caspi e  al. (2009) epo ed a cFPD Fig. 3 Non‑ca diac d ug esponses o he con ol and long QT synd ome (LQT) ype 1 and 2‑speci ic ca diomyocy e clus e s (hESC, WTa, WTb, LQT1A, LQT1B and LQT2). The d ugs used we e a cisap ide and b e y h omycin. Changes in he a e‑co ec ed ield po en ial du a ion (cFPD) we e calcula ed as ela i e changes om he baseline o each ca diomyocy e clus e . Da a is p esen ed mean ± SD. Mean ΔcFPDs we e no mal‑ ized o baseline cFPD. As e isks ma k he s a is ical signi icance o mean ΔcFPD p olonga ion compa ed o baseline alues. *p < 0.05, **p < 0.01, ***p < 0.001 Page 8 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 Fig. 4 Class I and class III an i‑a hy hmic d ug esponses o he con ol and long QT synd ome (LQT) ype 1 and 2‑speci ic ca diomyocy e clus e s (hESC, WTa, WTb, LQT1A, LQT1B and LQT2). The d ugs used we e a so alol, b quinidine and c E‑4031. Du ing E‑4031 challenge all bu one o he ca diac agg ega es in hESC‑CMs became a hy hmic o s opped bea ing a concen a ions o 500–700 nM. Thus, e o ba s indica ing s anda d de ia ion could no be calcula ed o hese expe imen al poin s. Changes in he a e‑co ec ed ield po en ial du a ion (cFPD) we e calcula ed as ela i e changes om he baseline o each ca diomyocy e clus e . Da a is p esen ed mean ± SD. Mean ΔcFPDs we e no malized o baseline cFPD. As e isks ma k he s a is ical signi icance o mean ΔcFPD p olonga ion compa ed o baseline alues. *p < 0.05, **p < 0.01, ***p < 0.001 Page 9 o 13 Kuusela e al. Sp inge Plus (2016) 5:234 p olonga ion o sligh ly <30% in hESC-CMs wi h 8µM quinidine (B aam e al. 2010; Caspi e al. 2009). Meh a e al. (2013) epo ed a 20 and 55% cFPD p olonga ion in 150–170days old hiPSC-CMs a 1 and 10µM concen a- ions, espec i ely. He e, we no iced an 18% inc ease in he cFPD o he hESC-CMs a a concen a ion o 9µM. Quinidine induced bea ing a es s which eme ged a he a e age o uppe he apeu ic se um concen a ion. Qui- nidine is known o inhibi sodium ion channels a high concen a ions (Snyde s and Hondeghem 1990), which may be he eason o highe equency o bea ing a es han AD-like wa e o m abno mali ies. The LQT2-CMs, showed signi ican ly p olonged cFPD al eady a he hal o he lowe limi o he apeu ic concen a ion (1.5µM), Fig. 5 Rep esen a i e aces o cisap ide, e y h omycin, so alol, quinidine and E‑4031 in con ol‑ and long QT‑ca diomyocy es (LQT‑CMs)