Simultaneous Measurement of Contraction and Calcium Transients in Stem Cell Derived Cardiomyocytes

Simul aneous Measu emen o Con ac ion and Calcium T ansien s in S em Cell De i ed Ca diomyocy es A. AHOLA , 1 R.-P. PO ¨LO ¨NEN, 2 K. AALTO-SETA ¨LA ¨, 2,3 and J. HYTTINEN 1 1 BioMediTech Ins i u e and Facul y o Biomedical Sciences and Enginee ing, Tampe e Uni e si y o Technology, Ko keakoulunka u 10, 33720 Tampe e, Finland; 2 BioMediTech Ins i u e and Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e, Finland; and 3 Hea Hospi al, Tampe e Uni e si y Hospi al, Tampe e, Finland (Recei ed 29 June 2017; accep ed 19 Sep embe 2017; published online 3 Oc obe 2017) Associa e Edi o Smada Cohen o e saw he e iew o his a icle. Abs ac —Induced plu ipo en s em cell de i ed ca diomy- ocy es (iPSC-CM) p o ide a powe ul pla o m o disease modeling and d ug de elopmen in i o. T adi ionally, elec ophysiological me hods o fluo escen dyes (e.g. cal- cium) ha e been used in hei unc ional cha ac e iza ion. Recen ly, ideo mic oscopy has enabled non-in asi e anal- ysis o CM con ac ile mo ion. Simul aneous assessmen s o mo ion and calcium ansien s ha e no been gene ally conduc ed, as mo ion de ec ion me hods a e a ec ed by changing pixel in ensi ies in calcium imaging. He e, we p esen o he fi s ime a p o ocol o simul aneous ideo- based measu emen o con ac ion and calcium wi h fluo es- cen dye Fluo-4 ideos wi hou co ec ions, p o iding da a on bo h ionic and mechanic ac i i y. The me hod and i s accu acy a e assessed by measu ing he e ec o fluo escence and backg ound ligh on ansien wid hs and con ac ion eloci y ampli udes. We demons a e he me hod by showing he con ac ion-calcium ela ion and measu ing he ansien ime in e als in ca echolamine gic polymo phic en icula achyca dia pa ien specific iPSC-CMs and heal hy con ols. Ou alida ion shows ha he simul aneous me hod p o ides compa able da a o combined indi idual measu emen s, p o iding a new ool o measu ing CM biomechanics and calcium simul aneously. Ou esul s wi h calcium sensi i e dyes sugges he me hod could be expanded o use wi h o he fluo escen epo e s as well. Keywo ds—Biomechanics, Fluo escence, Mo ion analysis, Calcium-con ac ion coupling. ABBREVIATIONS CaPIV Calcium pa icle image elocime y CD Con ac ion du a ion CM Ca diomyocy e CPVT Ca echolamine gic polymo phic en icu- la achyca dia CTD Calcium ansien du a ion hiPSC Human induced plu ipo en s em cell iPSC Induced plu ipo en s em cell iPSC-CM Induced plu ipo en s em cell de i ed ca - diomyocy e PIV Pa icle image elocime y RYR Ca diac yanodine ecep o MQD Minimum quad a ic di e ence WT Wild ype INTRODUCTION Ca dio ascula diseases ep esen he leading cause o dea h globally. Addi ionally, ad e se effec s o ca - dioac i e and non-ca diac d ugs emain he mos common eason o ma ke wi hd awal. Induced plu ipo en s em cell de i ed ca diomyocy es (iPSC- CM) ha e p esen ed a p omising pla o m o bo h disease modeling and d ug de elopmen wi h heal hy and pa ien specific cells. These s udies ha e been a- di ionally done using elec ical measu emen s, while me hods o measu ing mechanical unc ion o cells wi hou ex ensi e ins umen a ion ha e only ecen ly been eme ging. High h oughpu me hods o d ug sc eening wi h iPSC-CMs a e needed, as he la ge a iabili y in elec ophysiological esul s ega ding iPSC-CM b ing abou he need o a g ea e numbe o expe imen s. 15 These me hods could be used in addi- ion o de elopmen o pe sonalized he apy o ge- ne ic ca diac diseases, such as ca echolamine gic polymo phic en icula achyca dia (CPVT). 10 Add ess co espondence o A. Ahola, BioMediTech Ins i u e and Facul y o Biomedical Sciences and Enginee ing, Tampe e Uni e si y o Technology, Ko keakoulunka u 10, 33720 Tampe e, Finland. Elec onic mail: an i.l.ahola@ u .fi Annals o Biomedical Enginee ing, Vol. 46, No. 1, Janua y 2018 (Ó2017) pp. 148–158 DOI: 10.1007/s10439-017-1933-2 0090-6964/18/0100-0148/0 Ó2017 The Au ho (s). This a icle is an open access publica ion 148 The commonly used me hods o analyzing elec- ophysiological unc ions o ca diomyocy es (CM) include calcium imaging, 18 pa ch clamping 8,11 and mic oelec ode a ay 11 s udies. Howe e , hese mea- su emen s o en equi e manual ope a ion by ained pe sonnel and hus a e no well sui ed owa ds high- h oughpu applica ions. In addi ion, while he ionic unc ion hey measu e con ol he mechanical unc ion o he CMs, he esul ing signals do no ela e di ec ly o he esul ing biomechanics ha is he ac ual CM unc ional aspec . CM biomechanics has adi ionally been measu ed in i o using a omic o ce mic oscopy, 19 mechanical ansduc ion wi h mic op- os s 4,22 and ac ion o ce mic oscopy. 21 Video-based me hods ha e been de eloped and s udied by us 1,2,9 and by se e al o he g oups. 6,7,14 As he me hod is non-in asi e and has easy o use ins umen a ion, i- deo analysis has good high- h oughpu capabili ies. Calcium imaging can be used as a powe ul ool o high h oughpu p ofiling o d ug esponses in i o.In CMs, calcium plays a key ole as a second messenge o exci a ion con ac ion coupling. D ug e ec s a e eflec ed as changes in calcium ansien p ofile. Wi h fluo escen calcium epo e dyes, ei he small mole- cule o gene ically encoded, i is possible o isualize and eco d calcium fluxes inside CMs. Special high h oughpu imaging appa a uses, o example Hama- ma su FDSS, 3 allow simul aneous eading o calcium ansien s om up o hund eds o samples. These pla o ms can be used o s udy se e al disease models and d ugs in a sho ime pe iod compa ed o con- en ional mic oscopes. High con en sc eening p o- duces la ge amoun o da a, which equi e au oma ed so wa e o handling and analyzing he massi e amoun s o da a. 17 The majo benefi o human iPSC (hiPSC) in CM s udies is hei powe o p o ide CMs om specific pheno ypes. He e CPVT cells was selec ed as a ge ed applica ion example o ou no el me hodology. CPVT is a a e bu se e e inhe i ed ca diac disease in s uc u ally no mal hea s, which may lead o sudden ca diac dea h. CPVT pa ien s expe ience a hy hmias du ing men al o physical s ess. CPVT is associa ed wi h mu a ions in RYR2 gene, coding o ca diac yanodine ecep o (RYR), among ew o he disease- associa ed genes. The apy includes d ug ea men wi h be a-blocke s, implan able ca diac defib illa o s o le ca diac sympa he ic dene a ion. 23 The calcium abno mali ies a ising om mu a ions in RYR2 eflec o con ac ile mo ion o CMs. The e o e, CPVT specific iPSC-CMs a e in e es ing pla o m o s udy no only calcium kine ics bu also he con ac ile mo- ion simul aneously. So a , he ionic and mechanical unc ions ha e been s udied simul aneously using specific pla o ms, such as on mic opos s. 4,22 Video-based measu emen s, howe e , do no equi e such sys ems: mic oscopes and ideo came as a e ypical labo a o y ins umen s, which as such enable ideo-based measu emen s. Me hods o p o ide combined ionic and biomechanic assessmen o he CMs could o e no el a enues o look in o disease mechanisms and d ug e ec s. P e i- ously, i has no been able o ack mo ion in b igh field mic oscopy simul aneously wi h fluo escen e- po e s due o he e ec s o fluo escen induced changes in pixel in ensi y on he mo ion de ec ion me hods. In p e ious s udies he fluo escence back- g ound has been sub ac ed om images wi h mixed phase-con as and fluo escence. 6 In ano he s udy, calcium epo e GCaMP6 –gene was used and simul aneous measu emen was conduc ed wi h com- pu ed local in ensi y no maliza ion in each in e oga- ion window o coun e ac he e ec o he in ensi y changes. 7 In his s udy, we combine concu en calcium imaging wi h ou p e iously de eloped ideo analysis me hod, which uses minimum quad a ic diffe ence (MQD) based pa icle image elocime y (PIV) and assess biomechanical and ionic unc ion simul ane- ously om ideos o fluo escen calcium imaging. As in he MQD me hod we used 1 se s equal weigh o all image pixels, 5 we hypo hesized ha he me hod could wo k also in calcium imaging by inc easing back- g ound ligh ing condi ions o make mo ion mo e is- ible. To e i y ou hypo hesis, consecu i e ames wi h fluo escen calcium dye exci a ion and ansmission channels we e measu ed wi h wo backg ound ligh le els. We alida ed he me hod by compa ing con- ac ion and calcium ansien wid hs as well as PIV ampli udes om ames wi h and wi hou he fluo es- cen exci a ion, and wi h and wi hou backg ound ligh . We demons a ed he me hod by measu ing he ela ion o con ac ion and calcium in wild ype (WT) and CPVT cells, and calcula ed he ime in e al be ween he maximum signal onse and o se a es o change in bo h signals. MATERIALS AND METHODS Pa ien -Specific Human iPSC Lines De ails on hiPSC cul u ing and CM diffe en ia ion a e a ailable in supplemen s. In his s udy bo h no mal con ol CMs and diseased CMs p one o a hy hmias we e used. The diseased CPVT cells we e de i ed om pa ien s ha ing mu a- ions in RYR2 gene, associa ed wi h s uc u al changes in RYR, esul ing in abno mal elease o Ca 2+ om sa coplasmic e iculum and inc eased en icula a hy hmias when hea a e inc eases. Two CPVT Concu en Con ac ion and Ca 2+ Measu emen in iPSC CM 149 pa ien specific hiPSC lines we e used: UTA.05605.CPVT ca ying exon 3 dele ion and UTA.05404.CPVT ca ying V4563F mu a ion in RYR2 gene ( e e ed o as CPVTa and CPVTb espec i ely). UTA.04602.WT cell line was used as a con ol. The collec ion o biopsies o gene a ing pa- ien specific hiPSC lines was app o ed by he e hical commi ee o Pi kanmaa Hospi al Dis ic (Aal o-Se- a ¨la ¨R08070) and w i en in o med consen was ob ained om all he dono s. Imaging P o ocol De ails on imaging p o ocol a e a ailable in sup- plemen s. CMs pla ed on glass co e slips we e loaded wi h Fluo-4 AM. Imaging was done using Olympus IX71 in e ed mic oscope wi h Polych ome V and TH4-200 ligh sou ces and Ando iXon 885 EMCCD came a. Cells we e pe used wi h wa med (37 °C) pe usa e medium. Two channels we e eco ded consecu i ely: he Olympus ligh sou ce was u ned on o ob ain i- deo image while s ill being able o disce n Ca 2+ fluc- ua ions. The eco ding p ocess is illus a ed in Fig. 1a. In o al, 25 ideos we e eco ded: 7 CPVTa, 5 CPVTb and 13 WT ideos. The eco dings we e all baseline measu emen s. Video Da a P ocessing The ob ained eco dings, consis ing o ideo imag- ing and calcium imaging da a wi h ligh sou ce off and on, we e ca ego ized and p ocessed. The egion o in e es o each eco ding was selec ed based on bo h in ensi y o he calcium fluc ua ions and obse ed mo ion. MQD based PIV, as desc ibed p e iously, 1,2,9 was used o de e mining eloci y ec o fields in he ideo image se ies. The p ocess was applied o bo h ans- mission and calcium imaging wi h backg ound ligh on, esul ing in wo se ies o eloci y ec o fields om which mo ion signals a e calcula ed: one om he ames wi h only isible ligh — adi ional PIV mea- su emen —and ano he om he ames wi h fluo es- cen emission o calcium indica o , defined as calcium PIV (CaPIV). Only he ames wi h ligh sou ce on we e included in hese se s. Calcium signals we e calcula ed om he pixel in ensi y in each ame. Wi h espec o he ligh sou ce s a us, wo consecu i e da a se ies we e ob ained: calcium wi h fluo escen ligh om Poly- ch ome V—Ca (da k)—and calcium wi h ligh om bo h Polych ome V and Olympus TH4-200 ( ansmis- sion) - Ca (b igh ). Wi h his me hod, ou signal ypes we e ob ained— wo om bo h modali ies: calcium and mo ion. The p ocess is illus a ed in Fig. 1a. Using ou p e iously desc ibed me hod, sho ly, he di ec ional mo ion signals indica ing con ac ion and elaxa ion eloci ies we e ob ained om eloci y ec- o fields by de e mining a CM bea ing ocus poin . The mo ion di ec ed owa ds i was defined as adial mo emen and he pe pendicula mo ion as angen ial, esul ing in mo ion signals. As opposed o he p e ious a icle, each ec o was di ided in hese componen s wi h espec o hei loca ion. The same egion o in e es was used o calcium imaging. Fu he , con- ac ion signals we e calcula ed om he mo ion sig- nals by in eg a ion wi h espec o ime. Linea de ending was applied o he esul ing signals o co - ec he baseline. As a non- a iome ic dye was used and absolu e con ac ion ampli ude was no assessed, calcium, con ac ion and mo ion signals we e all no malized o in e al [0,1] by linea escaling o isualiza ion. The mo ion signal baseline was hen se o ze o, wi h con- ac ion being posi i e and elaxa ion nega i e. Fo calcium and con ac ion signals, one bea long c oss- FIGURE 1. Imaging p o ocol, p ocessing o he ideo channels and de ini ion o ansien du a ion pa ame e s. (a) Fluo escen calcium and isible ligh we e eco ded in consecu i e ames. Midway du ing he eco ding, ligh sou ce was u ned on o inc ease he backg ound ligh in ensi y, allowing ideo signal o be cap u ed. T adi ional PIV was measu ed om he ames wi h only isible ligh and CaPIV signal was measu ed om ames wi h calcium luo escence p esen . Calcium signal was ca ego ized based on he backg ound ligh in ensi y o Ca (da k) and Ca (b igh ). (b) T ansien du a ion pa ame e s a e de ined as pe cen ages om he peak maximum, simila ly as when classi ying ac ion po en ials. AHOLA e al.150 co ela ion based a e aged signal empla es we e cal- cula ed, as desc ibed ea lie . 2 The a e aging was syn- ch onized based on he maximum a e o change du ing he onse o he con ac ion o ansien . A co ela ion coe ficien o 0.5 was used o exclude only he mos abno mal bea s. Signal Compa ison The simila i y o he eco ded signals was measu ed by de e mining pa ame e s desc ibing he wa e o m wid hs and ampli udes. The du a ion o he con ac- ion in he empla es was cha ac e ized by calcula ing con ac ion du a ion (CD) pa ame e s and he du a- ion o calcium by calcium ansien du a ion (CTD) pa ame e s. The du a ions a pe cen ages o he max- imum peak heigh we e defined as shown in Fig. 1- b—an app oach simila o cha ac e iza ion o ac ion po en ials. The used pa ame e s we e CD10, CD25, CD50, CD80 and CD90, ep esen ing he wid h o he peak a each pe cen age measu ed om he peak in each wa e o m. Fo calcium, he same CTD pa ame- e s we e de e mined, espec i ely. Tempo al diffe ences be ween he con ac ion and calcium ansien we e measu ed using he combined signals om he maximum a e o change in he onse and offse o CaPIV and Ca (b igh ). As he calcium ansien s may ha e se e al peaks o diffe en heigh s, his app oach was selec ed ins ead o compa ing he ime diffe ence be ween peaks hemsel es. S a is ics Mean alues and s anda d de ia ions o CDs and CTDs, as well as empo al diffe ences be ween con- ac ion and calcium we e calcula ed. CDs and CTDs we e es ed agains each o he using a wo-sample es wi h all combina ions o PIV, CaPIV, Ca (da k) and Ca (b igh ). In addi ion, du a ions om WT cell ideos (n=13) we e es ed agains hose measu ed om CPVT cell ideos (CPVTa n=7, CPVTb n=5) ega ding all combina ions o signal ypes om he same modali y—calcium and mo ion. Wa e o m ampli ude simila i y was de e mined by calcula ing he symme ic mean absolu e pe cen age e o o con- ac ion mo ion eloci y magni udes o each con ac- ion in PIV and CaPIV signals. Linea eg ession was used o map wid h pa ame e s agains he o he signal ypes, and hei simila i y was es ima ed using linea eg ession R 2 pa ame e . P- alues below 0.05 we e conside ed s a is ically significan . The s a is ical analysis was pe o med in MATLAB R2014a, The Ma hwo ks, Inc. RESULTS Bea ing Cha ac e is ics Fo WT, he a e age bea ing a e was 22 bea s pe min wi h a ±10.8 s anda d de ia ion. Fo CPVTa and CPVTb he alues we e 20.7 ±12.9 and 12.2 ±5.6 bea s pe min, espec i ely. The numbe o bea s in CPVT ideos we e lowe han in WT ideos as some ideos only had wo bea s wi h he backg ound ligh on/off. Rep esen a i e Signals Rep esen a i e signals wi h synch onized mo ion (CaPIV) and calcium signals (Ca (b igh )) we e calcu- la ed. Fo each cell line, bo h mo ion and con ac ion signals a e shown wi h o e laid calcium ansien s in Fig. 2. In mo ion signals (le side), an upwa d peak signifies con ac ile and downwa d peak elaxa ion mo emen eloci y. Mo ion signal is used o illus- a ing he fine de ails o he mo emen . Con ac ion signal ( igh side)— he in eg al o he mo ion signal wi h espec o ime—shows displacemen du ing a con ac ion. The con ac ion signal enables easie compa ison o calcium and con ac ion. Fo his illus a ion, he signals we e linea ly no malized sepa- a ely o isualiza ion. In he isual inspec ion o he displacemen and calcium signals om s udied WT cell popula ions, con ac ion s a ed only a e inc ease in calcium ansien had s a ed. Con ac ion signals ypically had a s eepe offse slope han calcium, which had a longe ail, and s a ed o all sligh ly ea lie . In CPVTa cells, in some signals mo ion ollowed he calcium ansien wi h a longe delay han in WT. No e iden a e de- pola iza ions o ailing o calcium was p esen in he signals. In CPVTb, he con ac ion signal s a ed o all sligh ly la e han he calcium ansien . Me hod Valida ion The simul aneous eco ding me hod was alida ed by compa ing signals ob ained om PIV (mo ion om ideos wi hou fluo escence), CaPIV (mo ion om i- deos wi h fluo escence), Ca (b igh ) (calcium wi h backg ound ligh ) and Ca (da k) (wi hou backg ound ligh ) signals. CTD pa ame e s om he wo calcium signals, CD pa ame e s om he wo con ac ion sig- nals, and ampli udes om he wo mo ion eloci y signals we e calcula ed. PIV and CaPIV did no show s a is ically significan diffe ences be ween he means o CD pa ame e s. This can be seen in Fig. 3, whe e he o e laid CaPIV (in blue) and PIV ( ed) show a e y simila cu e, wi h Concu en Con ac ion and Ca 2+ Measu emen in iPSC CM 151 mino de ia ions showing nea he baseline. Simila ly, Ca (b igh ) and Ca (da k) did no exhibi significan di e ences in CTD pa ame e s. This indica es ha no majo di e ences we e induced by he change in ligh ing o enable he combined signal measu emen . The symme ic mean absolu e pe cen age e o s be ween PIV and CaPIV con ac ion eloci y ampli- udes we e calcula ed o he con ac ions in each eco ding. The diffe ence was ound o be 6.42 ±3.44% in WT, 4.11 ±5.01% in CPVTa and 2.73 ±1.57% in CPVTb. The o al e o o all eco dings was 5.04 ±3.87%. Assessmen o Cell Line Func ions Using Simul aneous Con ac ion and Calcium Con ac ion and Calcium Signal Cha ac e is ics The wid hs o he con ac ion peak in CaPIV and he calcium ansien Ca (b igh ) in he simul aneous measu emen we e cha ac e ized by pa ame e s indi- ca ing hei du a ion a pe cen ages o he peak max- imum heigh . The CD- and CTD-pa ame e s we e calcula ed a signal le els 10, 25, 50, 80 and 90 and shown in Table 1. In bo h CD and CTD pa ame e s, he s anda d de ia ion was e y high indica ing high a iance wi hin he cell lines. Simila i y o Signals Be ween Modali ies Linea eg essions be ween pai s o con ac- ion/calcium ansien du a ions o he ou diffe en signals we e calcula ed. The esul s a e shown in Table 2. Pai s o he same modali y (le mos ables) FIGURE 2. Rep esen a i e mo ion (le side) and displacemen ( igh side) signals plo ed o e calcium ansien s o he h ee cell lines. In he mo ion signals, he con ac ile mo emen eloci y is ep esen ed wi h and upwa d peak and he elaxa ion mo emen eloci y as a downwa d peak in ed, while calcium ansien is shown in blue. In he displacemen signals ( igh side), ep esen ing he mo ion eloci y signal in eg al wi h espec o ime, con ac ion is shown in ed and calcium ansien in blue. Rep esen a i e signals a e shown o he h ee cell lines. FIGURE 3. A ep esen a i e displacemen measu emen showing o e laid CaPIV (blue) wi h luo escence and PIV (shown in ed) wi hou luo escence. AHOLA e al.152 show high R 2 alues: PIV exhibi s e y high coe ficien o de e mina ion (R 2 ) alues wi h CaPIV in bo h WT and CPVT cells. Only in CD90 he alue o CPVTb R 2 was below 0.9. Likewise, high R 2 alues we e ob ained om eg ession be ween Ca (b igh ) and Ca (da k) e en hough he signals we e no concu en like mo- ion signals. In calcium, R 2 alues below 0.9 we e seen in CPVTa CTD25 and CTD90, and a ela i ely low alue in CTD10 o CPVTb. Low R 2 alues we e ypi- cally seen in c oss-modali y pai s nea he baseline (CD/CTD80 and CD/CTD90), especially in CPVTa. These esul s indica e he simul aneous measu emen wo ks well— he signals o he same modali y a e in- deed e y simila in shape. Fu he , he simul aneous mo ion and calcium measu emen s show a disconnec- ion o calcium and mo ion in CPVT lines—especially CPVTa—which is no e iden in calcium measu e- men s. Calcium-Mo ion Coupling The concu en calcium-con ac ion measu emen allows he measu emen o synch onici y in calcium and mo ion dynamics. He e, we looked a he empo al diffe ences be ween hese signals by calcula ing he ime poin s o maximum a e o change in he onse s and offse s o calcium ansien and mo ion om Ca (b igh ) and CaPIV, espec i ely. The ob ained alues a e shown in Table 3. Posi i e ime alues in onse and offse indica e ha mo ion signal is ollowing he calcium ansien . Some nega i e alues in offse we e ob ained indica ing ha mo ion eached i s peak elaxa ion eloci y be o e maximum calcium decline eloci y. In Table 3 his is indica ed by la ge s anda d de ia ion compa ed o he o se alue especially o CPVTa. Al hough CPVTa shows a highe onse and CPVTb a highe o se in mean imes, he findings a e no s a is ically significan due o he la ge a iabili y. Also hese esul s e eal ha CPVTa and CPVTb show di e en con ac ion- calcium coupling dynamics om WT bu also di e en om each o he . The diffe ence be ween he means o mo ion and calcium signal pa ame e s was s a is ically significan in WT in all pa ame e s (p<0.05 in CD/CTD10, p<0.01 in he emaining pa ame e s), bu no in CPVTa o CPVTb. This finding shows ha measu e- men s o mo ion and calcium p o ide complemen a y in o ma ion and insigh on he Ca 2+ and con ac ion coupling dynamics. When compa ing he cell lines, he mo ion modali- ies PIV and CaPIV did no exhibi significan diffe - ences be ween he con ac ion du a ions o WT and CPVT cell lines. Howe e , calcium modali ies Ca (b igh ) and Ca (da k) showed significan diffe ences TABLE 1. Calcium and con ac ion signal cha ac e iza ion. CD10 CD25 CD50 CD80 CD90 CTD10 CTD25 CTD50 CTD80 CTD90 WT 229 ±77 ms 343 ±118 ms 475 ±165 ms 684 ±206 ms 888 ±391 ms 328 ±104 ms 513 ±148 ms 797 ±227 ms 1315 ±420 ms 1663 ±570 ms CPVTa 317 ±185 ms 436 ±200 ms 590 ±266 ms 888 ±250 ms 1340 ±651 ms 356 ±150 ms 581 ±236 ms 797 ±305 ms 1057 ±330 ms 1283 ±414 ms CPVTb 255 ±175 ms 379 ±181 ms 515 ±202 ms 714 ±190 ms 919 ±301 ms 273 ±150 ms 413 ±244 ms 589 ±268 ms 788 ±296 ms 905 ±415 ms The cha ac e iza ion o con ac ion and calcium in he simul aneous measu emen . The pa ame e s CD10–CD90 e e o con ac ion peak wid hs and CTD10–CTD90 o calcium ansien wid hs a pe cen ages o maximum peak heigh . Concu en Con ac ion and Ca 2+ Measu emen in iPSC CM 153 be ween means o WT and CPVTb in CTD80 and CTD90. Nei he calcium no mo ion modali ies dis- played significan diffe ences be ween WT and CPVTa. DISCUSSION We de eloped and demons a ed a me hod o simul aneous CM mo ion and calcium ansien eco ding om fluo escen ideos. The me hod uses simul aneous applica ion o bo h ansmission and fluo escen ligh sou ce. In adi ional me hods he calcium based fluo escen signal could ep esen a sou ce o e o o PIV analysis and u he , he b igh field imaging could affec he calcium imaging ansien s. We es ed he me hod suscep ibili y o bo h e o sou ces by sepa a ing hese ac o s wi h in e - lacing b igh field and fluo escen ideo ames and obse ed ha a simul aneous measu emen o calcium and con ac ion is easible. We eco ded consecu i e ames o b igh field imaging and calcium fluo escen imaging in iPSC-CMs on ideo, wi h backg ound ligh on and off. Fou diffe en signals we e ob ained: mo ion om b igh - field ideo (PIV), mo ion om ideo wi h fluo escence (CaPIV), calcium imaging wi hou backg ound ligh (Ca (da k)) and calcium imaging wi h backg ound ligh (Ca (b igh )). We cha ac e ized he signals o bo h alida ion and demons a ion o he me hod. Wi h his p ocedu e, we aimed o demons a e ha CM con ac ion could be quan ified eliably while unning a simul aneous fluo escen measu emen . Fu he , we demons a ed he use ulness o he simul aneous measu emen by compa ing he calcium and con ac ion signals, calcula ing ime in e als be ween calcium and mo ion, and analyzing he di - e ences be ween WT and wo diffe en CPVT cell lines. Effec o Fluo escen Ligh on Mo ion Analysis P e iously, he in ensi y changes caused by fluo es- cence ha e p e en ed simul aneous mo ion measu e- men s. MQD, howe e , does no pu emphasis on b igh pixels hus making i a plausible me hod o he ask. Ou esul s show ha he in ensi y fluc ua ion effec by fluo escen ligh is negligible o bo h iming and magni ude o he mo ion signal. This is he fi s ime such measu emen and assessmen o accu acy TABLE 2. Linea eg ession o CD and CTD pa ame e s. PIV/CaPIV PIV/Ca (b igh ) CaPIV/Ca (b igh ) WT CPVTa CPVTb WT CPVTa CPVTb WT CPVTa CPVTb 10 0.983 0.997 0.975 0.663 0.128 0.430 0.698 0.136 0.511 25 0.943 0.997 0.996 0.737 0.540 0.897 0.698 0.537 0.888 50 0.965 0.997 0.995 0.637 0.731 0.732 0.612 0.684 0.752 80 0.967 0.991 0.976 0.640 0.129 0.451 0.630 0.091 0.384 90 0.966 0.989 0.817 0.710 0.000 0.186 0.724 0.003 0.057 Ca (b igh )/Ca (da k) PIV/Ca (da k) CaPIV/Ca (da k) WT CPVTa CPVTb WT CPVTa CPVTb WT CPVTa CPVTb 10 0.914 0.913 0.567 0.518 0.138 0.958 0.525 0.150 0.945 25 0.939 0.753 0.859 0.685 0.605 0.834 0.612 0.616 0.793 50 0.951 0.970 0.991 0.552 0.741 0.742 0.507 0.702 0.760 80 0.968 0.946 0.992 0.573 0.155 0.539 0.564 0.119 0.473 90 0.950 0.778 0.961 0.686 0.070 0.274 0.685 0.111 0.082 Linea eg ession be ween he con ac ion (CD)/calcium ansien du a ions (CTD) a pe cen ages o he maximum peak heigh (10, 25, 50, 80 and 90) o each pai o measu ed signals. TABLE 3. Time in e als be ween calcium and con ac ion. WT CPVTa CPVTb Ca (b igh ) and CaPIV di e ence on onse 163 ±132 ms 238 ±136 ms 126 ±102 ms Ca (b igh ) and CaPIV di e ence on o se 40 ±91 ms 24 ±142 ms 113 ±75 ms Mean ime di e ences and s anda d de ia ions be ween he calcium ansien and con ac ion in di e en cell lines, in milliseconds. The di e ences be ween cell lines a e no s a is ically signi ican . AHOLA e al.154 ha e been done wi hou using co ec ions wi h a flu- o escen dye. Ou esul s indica e ha MQD based analysis o CM mo ion is minimally affec ed by he p esence o a fluo escence based measu emen . In - es s, he e we e no significan diffe ences be ween he means o mea- su ed CD pa ame e s. The R 2 alues in linea eg es- sion s udies be ween no mal b igh field PIV and CaPIV wi h fluo escence we e e y high in all h ee cell lines indica ing he e y low impac o fluo escence p esence. The e we e small di e ences in he lowe pa s o he peaks as well as a mino 5% di e ence in magni udes o con ac ion eloci ies. We belie e his measu emen se up allows measu ing o mo ion wi h a a ie y o di e en fluo escen indica o s, as long as mo emen is p esen . De e mining a su ficien le el o ligh ing o he measu emen s may be a key ac o in he acquisi ion o wo simul aneous high quali y eco dings. Effec o B igh Ligh on Calcium Imaging Applying ansmission ligh had only small effec on he calcium imaging wa e o ms. In linea eg ession analysis, high R 2 alues we e ob ained be ween Ca (b igh ) and Ca (da k), in all h ee cell lines. Howe e , in CPVT some lowe eg ession alues we e ob ained due o a numbe o possible ac o s: calcium ansien s Ca (b igh ) and Ca (da k) we e no om he same bea s, he numbe o eco ded cells in CPVT lines was low and as he bea ing a e was lowe , he numbe o bea s in each signal was smalle han in WT. As illus a ed in he CPVTa signals shown in Fig. 2, he calcium ansien s could ha e mul iple small peaks (i.e. a hy hmias/abno mal ansien s) making he di ec compa ison o peak wid h pa ame e s di ficul . Fo WT cells, he peaks we e mo e egula and Ca (b igh ) and Ca (da k) we e bo h capable o cha ac e izing he cell lines simila ly, as shown in - es s. Ca (da k) and Ca (b igh ) ampli udes we e no compa ed, as he compa ison would no ha e been alid due o he bleaching o fluo escen dye du ing he measu emen . Mo eo e , i is e iden ha inc eased backg ound ligh inc eases he backg ound noise le el in fluo escence measu emen . Howe e , su ficien ansmission ligh o cons an fluo escen ligh is needed o de ec ion o mo ion signal om he ideo. A he same ime, he fluo escen emission o he dye should be s ong en- ough o be de ec ed while he ansmission ideo is acqui ed. Inc easing he ligh in ensi y o exposu e ime o he sample may no be he bes solu ions since hey esul in highe pho obleaching o fluo opho es and slowe ame a es. Ra iome ic dyes ha e diffe en emission spec a o bound and unbound s a es wi h he indica ed ion, o example wi h Fu a-2. The e o e, i akes wo consec- u i e channel eco dings o cap u e he a io be ween hese s a es, which akes longe and esul s in lowe ame a es in acquisi ion. In his case, as an addi ional channel o b igh field ideo is measu ed, non- a io- me ic dyes, such as Fluo-4, a e be e since hei emission needs o be eco ded only on one wa eleng h a he han a iome ic dyes. The esul s indica e ha in calcium s udies, wi h a ocus on he wa e o ms in- s ead o absolu e Ca 2+ concen a ion, i is easible o apply also he b igh field ligh o ob ain addi ional in o ma ion on he mechanical ac i i y wi hou a ec ing he calcium measu emen . Cell Line Diffe ences The e was no s a is ically significan diffe ences in - es s be ween mo ion in WT and CPVT cell lines when conside ing PIV o CaPIV. CPVT cells ypically p e- sen a hy hmias unde ad ene gic s imula ion, bu as only he baseline bea ing is analyzed he e, he mo ion signals co espond o hose o WT. These esul s would indica e ha he CPVT cells do no di ec ly exhibi bea ing pheno ypes di e en om WT cells a baseline condi ions. Ea lie , using PIV we ha e ound di e ing bea ing pheno ypes in LQT cells, bu o cou se also he clinical p esen a ion o he diseases a e di e en . 9 Ca (b igh ) and Ca (da k) showed s a is ically sig- nifican diffe ences be ween WT and CPVTb in CTD80 and CTD90, bu no be ween WT and CPVTa. These diffe ences could be caused by a highe a ia ion in dias olic calcium le els in CPVTb caused by a leakage o calcium h ough RYRs in he cells. Howe e , he ampli ude o hese dias olic e en s is so low compa ed o he ac ual calcium peak so he e we e no diffe ences in he pa ame e s associa ed wi h he highe pa s o he peak (i.e. CTD10). The combina ion o mo ion and calcium measu e- men s showed a a ying disconnec ion o he wo, as shown by he R 2 alues (Table 2) especially o CPVTa. The mos p e alen cause o CPVT lies in mu a ions in RYR2 gene coding o RYR. 20 I is lo- ca ed a he memb ane o sa coplasmic e iculum in CMs and ac s as a channel eleasing calcium om sa coplasmic e iculum o cy osol. Mu a ions in RYR2 cause spon aneous calcium leakage h ough he chan- nel and al e ed calcium ansien s inside CMs. 13 In CPVTb popula ion, which ca ies a V4653F mu a ion in RYR, he subs i u ed amino acid is loca ed in he channel domain o he p o ein. 16 I is hough o be a gain-o - unc ion poin mu a ion, as mos o he RYR mu a ions, and al e he channels sensi i i y o pe - meabili y o calcium ions. 23 CPVTa popula ion, which ca ies an exon 3 dele ion, has he s uc u al de ec on he cy osolic pa o he p o ein in he N- e minal Concu en Con ac ion and Ca 2+ Measu emen in iPSC CM 155 domain and is associa ed wi h mo e se e e clinical ou come. 12 The di e ence in he na u e o hese wo mu a ions could explain he di e ences in ou esul s. The disconnec ion be ween calcium and bea ing mo- ion could be ela ed o he ea lie finding by Tang e al. whe e he exon 3 dele ion was ound o be esul ing in an abno mal e mina ion o calcium e- lease h ough he RYR. 24 Simul aneous Calcium and Mo ion Measu emen s The combined simul aneous eco ding o mo ion and calcium as shown he e p o ides new ools o s udying he CM con ac ion mechanism. In - es s be ween mo ion and calcium modali ies, s a is ical diffe ences we e seen in WT cell lines, bu no in CPVTa/CPVTb. The lack o s a is ical diffe ence be ween mo ion and calcium in CPVT was su p ising and i may again be caused by he low numbe o eco ded cells o mo e he e ogeneous bea ing pa e ns and calcium ansien s. In linea eg ession analysis o mo ion and calcium modali ies, ep esen ed in Table 2, CPVTa had lowe R 2 alues han WT and CPVTb: he e was p ac ically no ela ion be ween he mea- su emen s in peak pa ame e s cha ac e izing he low ampli udes. Mean ime diffe ences o signal onse and offse a e o change maxima did no show s a is ically significan diffe ences be ween he cell lines due o high a iabili y and low numbe o cells measu ed. Howe e , CPVTa did show indica ion o longe ime in e als be ween calcium and mo ion in onse , and CPVTb in offse . O e all, he esul s indica e ha calcium and mo ion go hand in hand as is expec ed, bu one canno di ec ly be deduced om he o he . Al hough he low co e- la ion alues in CPTVa may be caused by he low numbe o samples, high co ela ion alues we e no p esen o WT ei he whe e he sample numbe was highe . This ein o ces he idea ha e en i Ca 2+ is he d i ing o ce in CM con ac ion, calcium measu emen alone canno be used o desc ibe he mechanical unc ion in ull de ail. Fu he , mo ion alone does no necessa ily eflec he ionic unc ions, emphasizing he need o he combined me hod sugges ed he e as a new ool o CM analysis. Me hod Ad an ages Video-based measu emen s a e minimally in asi e and ha e low ins umen a ion equi emen s. These aspec s make hem a easible candida e o high- h oughpu s udies. Fo labo a o ies, his me hod can be used in conjunc ion wi h o he s udies, as mic o- scopes and ideo came as a e mains ay labo a o y equipmen . The h eshold o including con ac ion measu emen s in s udies is lowe han measu emen s wi h o he me hods, because no addi ional equipmen is needed. Combining he me hod wi h fluo escen beads, subs a e de o ma ion and hus con ac ion o ce could be measu ed as well. While he e we ha e conside ed indi idual cells and small clus e s, he me hod should be applicable o la ge samples o ca diac issue, as long as 2-dimensional app oach is s ill easible. As we ha e shown he e, he possibili y o measu ing CM con ac ion om calcium imaging da a also enables e isi ing p e ious s udies and expanding on hei esul s. S udy Limi a ions and Sou ces o E o To alida e he me hod we used in e laced ideo. The ime diffe ence be ween PIV and CaPIV i- deos—by being consecu i e ames and he ela i ely low ame a e—mus be aken in o accoun when es ima ing hei diffe ences and simila i ies. The ame in e lacing p oduces in insic ime diffe ences o 21.7 and 8.8 ms o ou minimum ame a e 46 and max- imum ame a e 114, espec i ely. A highe ame a e wi h slow bea ing cells would gi e a mo e accu a e eading. Howe e , he obse ed diffe ences be ween no mal b igh field PIV and fluo escence CaPIV we e small, indica ing ha MQD-based mo ion measu e- men s a e minimally obs uc ed by fluo escence sig- nals. We ha e p e iously explo ed wa e o m a e aging in b ie , 2 sugges ing a ame a e o 60. Fu he s udies wi h a highe ame a e and mo e bea s pe ideo would be e quan i y he di e ence in ampli udes o measu emen s whe e absolu e p ecision o magni ude is equi ed. Due o he low numbe o eco dings, he findings ela ed o he disconnec ion be ween calcium and mo ion in CPVT cells should be conside ed en- a i e. A s udy ocusing on calcium-mo ion coupling in CPVT cells is equi ed o eliabili y. In his s udy, CMs we e cul u ed on glass co e slips whe e hei mo phology is diffe en om ma u e human CMs. The e o e, esul s may be diffe en using i.e. pa e ned CMs. Howe e , ou me hod should be applicable o hose s udies also. Ou me hod uses simul aneous op imized ans- mission and fluo escen ligh sou ces. The esul s showed ha he added ligh sou ce had e y small e - ec on he de ec ed fluo escen signal. Howe e , he me hod is p oducing addi ional ligh load o he fluo- escen dye and cells and hus pho obleaching o he dye and pho o oxici y should be conside ed. As ou me hod uses no fluo escen signal backg ound co ec- ion, i is s aigh o wa d o compu e and ee om possible co ec ion e o s. AHOLA e al.156