..................................................................................................................................................................................... ..................................................................................................................................................................................... BASIC SCIENCE Derivation and cardiomyocyte differentiation of induced pluripotent stem cells from heart failure patients Limor Zwi-Dantsis 1,2 , Irit Huber 1 , Manhal Habib 1 , Aaron Winterstern 1 , Amira Gepstein 1 , Gil Arbel 1 , and Lior Gepstein 1,3 * 1 Sohnis Research Laboratory for Cardiac Electrophysiology and Regenerative Medicine, The Bruce Rappaport Faculty of Medicine, Technion–Israel Institute of Technology, POB 9649, Haifa 31096, Israel; 2 Biotechnology Interdisciplinary Unit, Technion–Israel Institute of Technology, Haifa, Israel; and 3 Cardiology Department, Rambam Medical Center, Haifa, Israel Received 4 April 2011; revised 20 February 2012; accepted 22 March 2012 Aims Myocardial cell replacement therapies are hampered by a paucity of sources for human cardiomyocytes and by the expected immune rejection of allogeneic cell grafts. The ability to derive patient-specific human-induced pluripotent stem cells (hiPSCs) may provide a solution to these challenges. We aimed to derive hiPSCs from heart failure (HF) patients, to induce their cardiomyocyte differentiation, to characterize the generated hiPSC-derived cardiomyocytes (hiPSC-CMs), and to evaluate their ability to integrate with pre-existing cardiac tissue. Methods and results Dermal fibroblasts from two HF patients were reprogrammed by retroviral delivery of Oct4, Sox2, and Klf4 or by using an excisable polycistronic lentiviral vector. The resulting HF-hiPSCs displayed adequate reprogramming properties and could be induced to differentiate into cardiomyocytes with the same efficiency as control hiPSCs (derived from human foreskin fibroblasts). Gene expression and immunostaining studies confirmed the cardiomyocyte phenotype of the differentiating HF-hiPSC-CMs. Multi-electrode array recordings revealed the de- velopment of a functional cardiac syncytium and adequate chronotropic responses to adrenergic and cholinergic stimulation. Next, functional integration and synchronized electrical activities were demonstrated between hiPSC- CMs and neonatal rat cardiomyocytes in co-culture studies. Finally, in vivo transplantation studies in the rat heart revealed the ability of the HF-hiPSC-CMs to engraft, survive, and structurally integrate with host cardiomyocytes. Conclusions Human-induced pluripotent stem cells can be established from patients with advanced heart failure and coaxed to differentiate into cardiomyocytes, which can integrate with host cardiac tissue. This novel source for patient-specific heart cells may bring a unique value to the emerging field of cardiac regenerative medicine. ----------------------------------------------------------------------------------------------------------------------------------------------------------- Keywords Induced pluripotent stem cells † Cardiomyocytes † Heart failure † Cell therapy Introduction Recent advancements in stem cell biology and tissue engineering have paved the way for the introduction of a new discipline in bio- medicine: regenerative/reparative medicine. One of the goals of this emerging cardiovascular discipline is to repopulate scar tissue with new contractile cells in an attempt to assist the failing heart. 1,2 This exciting approach has been hampered, however, by the paucity of cell sources for human cardiomyocytes. The derivation of human embryonic stem cells (hESCs) 3 pro- vided a possible solution to this cell-sourcing problem because of their capacity to differentiate into cardiomyocytes 4,5 and their ability to improve cardiac function in animal models of myocardial infarction. 6,7 Yet, a major obstacle to the clinical utilization of these cells is their restricted availability and the anticipated immune rejection following transplantation of allogeneic cells. An important breakthrough was made with the introduction of the human-induced pluripotent stem cells (hiPSCs) technology. 8 – 10 The * Corresponding author. Tel: +972-4-8295303, Fax: +972-4-8524758, Email: [email protected]Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2012. For permissions please email: [email protected]European Heart Journal doi:10.1093/eurheartj/ehs096 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110
Derivation and cardiomyocyte differentiationof induced pluripotent stem cellsfrom heart failure patientsLimor Zwi-Dantsis1,2, Irit Huber1, Manhal Habib1, Aaron Winterstern1,Amira Gepstein1, Gil Arbel1, and Lior Gepstein1,3*
1Sohnis Research Laboratory for Cardiac Electrophysiology and Regenerative Medicine, The Bruce Rappaport Faculty of Medicine, Technion–Israel Institute of Technology,POB 9649, Haifa 31096, Israel; 2Biotechnology Interdisciplinary Unit, Technion–Israel Institute of Technology, Haifa, Israel; and 3Cardiology Department, Rambam Medical Center,Haifa, Israel
Received 4 April 2011; revised 20 February 2012; accepted 22 March 2012
Aims Myocardial cell replacement therapies are hampered by a paucity of sources for human cardiomyocytes and by theexpected immune rejection of allogeneic cell grafts. The ability to derive patient-specific human-induced pluripotentstem cells (hiPSCs) may provide a solution to these challenges. We aimed to derive hiPSCs from heart failure (HF)patients, to induce their cardiomyocyte differentiation, to characterize the generated hiPSC-derived cardiomyocytes(hiPSC-CMs), and to evaluate their ability to integrate with pre-existing cardiac tissue.
Methodsand results
Dermal fibroblasts from two HF patients were reprogrammed by retroviral delivery of Oct4, Sox2, and Klf4 or byusing an excisable polycistronic lentiviral vector. The resulting HF-hiPSCs displayed adequate reprogrammingproperties and could be induced to differentiate into cardiomyocytes with the same efficiency as controlhiPSCs (derived from human foreskin fibroblasts). Gene expression and immunostaining studies confirmed thecardiomyocyte phenotype of the differentiating HF-hiPSC-CMs. Multi-electrode array recordings revealed the de-velopment of a functional cardiac syncytium and adequate chronotropic responses to adrenergic and cholinergicstimulation. Next, functional integration and synchronized electrical activities were demonstrated between hiPSC-CMs and neonatal rat cardiomyocytes in co-culture studies. Finally, in vivo transplantation studies in the rat heartrevealed the ability of the HF-hiPSC-CMs to engraft, survive, and structurally integrate with host cardiomyocytes.
Conclusions Human-induced pluripotent stem cells can be established from patients with advanced heart failure and coaxed todifferentiate into cardiomyocytes, which can integrate with host cardiac tissue. This novel source for patient-specificheart cells may bring a unique value to the emerging field of cardiac regenerative medicine.
IntroductionRecent advancements in stem cell biology and tissue engineeringhave paved the way for the introduction of a new discipline in bio-medicine: regenerative/reparative medicine. One of the goals ofthis emerging cardiovascular discipline is to repopulate scartissue with new contractile cells in an attempt to assist the failingheart.1,2 This exciting approach has been hampered, however, bythe paucity of cell sources for human cardiomyocytes.
The derivation of human embryonic stem cells (hESCs)3 pro-vided a possible solution to this cell-sourcing problem becauseof their capacity to differentiate into cardiomyocytes4,5 and theirability to improve cardiac function in animal models of myocardialinfarction.6,7 Yet, a major obstacle to the clinical utilization of thesecells is their restricted availability and the anticipated immunerejection following transplantation of allogeneic cells.
An important breakthrough was made with the introduction of thehuman-induced pluripotent stem cells (hiPSCs) technology.8–10 The
Strictly embargoed until 00.05hrs BST Wednesday 23rd May 2012
hiPSCs were derived by the reprogramming of adult somatic cellswith a set of transcription factors, yielding pluripotent cells closelyresembling hESCs. More recent studies demonstrated the ability todifferentiate hiPSCs into the cardiac lineage11–13 and to developin vitro models of inherited cardiac disorders.14–16
While the aforementioned studies confirmed the ability toderive hiPSCs from young or healthy individuals, it is still notclear whether such lines could be established, and differentiatedinto cardiomyocytes also in elderly and diseased patients. Toaddress this important issue, we aimed to derive hiPSCs from is-chaemic cardiomyopathy patients suffering from advanced heartfailure (HF) (representing the prototype candidates for futureautologous myocardial cell-replacement procedures). Our resultsshow the ability to efficiently derive hiPSC lines from such patientsand to differentiate them into cardiomyocytes. The generatedhiPSC-derived cardiomyocytes (hiPSC-CMs) were demonstratedto functionally integrate with pre-existing cardiac tissue in vitroand to engraft, survive, and integrate structurally with the hostmyocardium following in vivo transplantation.
Methods
Generation and cardiomyocytedifferentiation of patient-derivedhuman-induced pluripotent stem cellsThe study was approved by the Rambam Medical Center Helsinki com-mittee. Patient-derived hiPSC clones were established by retroviral de-livery of three reprogramming factors (Sox2, Klf4, and Oct4) followedby valproic acid (VPA) treatment as previously described.14 Toinduce cardiomyocyte differentiation, hiPSCs were dispersed intosmall clumps with collagenase-IV (Life-Technologies, 1 mg/mL), culti-vated in a suspension for 10 days as embryoid bodies (EBs), andplated on 0.1% gelatin-coated culture dishes.13
Generation of transgene-free heartfailure-human-induced pluripotent stem cellsA single polycystronic lentiviral vector containing the four-factor ‘stemcell cassette’ (STEMCCA),17 which can be excised after integrationusing Cre-recombinase, was used for transduction of the fibroblasts(kind gift from Prof. Mostoslavsky). The STEMCCA cassette wasco-transfected with the plasmid Gag-Pol and the helper plasmid en-coding VSVG into HEK293T cells to produce viruses. After infection,cells were cultured in ES medium starting from 1 day post-infectionand treated with 0.9 mM VPA for 2 weeks.
For Cre-recombinase-mediated vector excision, hiPSCs wereharvested with 0.05% trypsin/EDTA (Invitrogen), re-suspended inPBS (1 × 106cells), and transfected by electroporation withpCAG-Cre-EGFP (10 mg). Cre-recombinase eGFP-expressing cellswere selected from a single-cell suspension by FACS sorting (FAC-SAria, BD-Biosciences) 72 h after electroporation. Selected cellswere re-plated at low density in ES medium containing the ROCKinhibitor (Sigma).
ImmunostainingSpecimens were fixed with 4% paraformaldehyde, permeabilized with1% Triton-X-100 (Sigma), blocked with 5% horse serum, and incu-bated overnight at 48C with primary antibodies targeting: Tra-1-60,Oct-4, connexin-43 (Cx43) (all from Santa-Cruz), Nanog (Abcam),
TRA-I-81, human mitochondria antigens (Millipore), cardiac troponin-I(cTnI, Chemicon), SSEA-4, cardiac troponin-T (cTnT) (R&D), andsarcomeric-a-actinin (Sigma). Preparations were incubated withsecondary antibodies for 1 h and examined using a laser-scanningconfocal microscope (Zeiss LSM-510-PASCAL).
Teratoma formationUndifferentiated hiPSCs were injected subcutaneously to immunodefi-cient SCID-beige mice. Teratomas, developing 4–8 weeks after injec-tion, were harvested, cryosectioned (10 mm), and H&E stained.
Bisulfite sequencingGenomic DNA (1 mg) was bisulfite converted with theMethylamp-DNA-Modification-kit (Epigentek) and amplified usingFaststart Taq-polymerase (Roche). PCR products were TA-clonedinto pTZ57R/T plasmid (Fermentas). Inserts were sequenced withM13 universal primers. Primer sequences appear in Table 1.
Karyotype analysisKaryotype analysis was performed using the standard G-bandingchromosome analysis at the institution’s cytogenetic laboratory.
Gene expression studiesUndifferentiated hiPSCs and beating EBs (30–40 days) were frozen inliquid nitrogen. RNA was isolated using the RNeasy-plus mini-kit(Qiagen). Reverse transcription into cDNA was conducted using ahigh capacity cDNA-reverse-transcription kit (Applied-Biosystems).The PCR program used was 3 min at 938C, 30 s at 938C, 30 s at608C, and 30 s at 728C (30 cycles) using DreamTaqTM DNA Polymer-ase (Fermentas).
SYBR green quantitative-PCR studies were performed in triplicatesusing SYBR Green PCR Master Mix and the ABI-7000 SequenceDetector (Applied-Biosystems). Samples were cycled 40 times (2 minat 508C, 15 min at 958C) followed by 40 cycles of 15 s at 958C, 30 sat 608C, and 30 s at 728C. The threshold cycle (CT) was calculatedunder default settings for the real-time sequence detection software.Expression levels were normalized to b-tubulin transcript levels.Primers for RT–PCR and quantitative-PCR are detailed in Table 1.
Multi-electrode array recordingsContracting EBs were microdissected and plated on fibronectin-coatedMEA plates (multi-channel systems). Local activation times (LATs)were determined at each of the 60 electrodes and used to generatecolour-coded activation maps as described.18
Co-culture studiesPrimary monolayer cultures of neonatal rat (Sprague-Dawley) ven-tricular myocytes (NRVMs) were prepared as described18 and culturedon MEA plates (2 × 106cells/mL). Contracting EBs were then added tothe NRVM cultures.
Cell transplantationFemale Sprague-Dawley rats (250–300 g) were anaesthetized (keta-mine/xylazine), intubated, and ventilated. The hiPSC-derived cardio-myocytes (1.5 × 105) were pre-labelled with a fluorescent cell tracer(Qtracker, Invitrogen) and injected into the rat’s left ventricular myo-cardium (n ¼ 3). To prevent graft rejection, animals were treated withcyclosporine-A (15 mg/kg/day) and methylprednisolone (2 mg/kg/day).Hearts were harvested 7–10 days following grafting, frozen in liquidnitrogen, and cryo sectioned (10 mm) for histological examination.
L. Zwi-Dantsis et al.Page 2 of 12
115
120
125
130
135
140
145
150
155
160
165
170
175
180
185
190
195
200
205
210
215
220
225
Statistical analysisResults are reported as mean+ SEM. To compare between the differ-ent hiPSC clones (for differentiation efficiency and beating rates),one-way ANOVA was used followed by post hoc Bonferroni analysis.Drug effects on beating rates were analysed by paired T-tests. All stat-istical methods were two sided. The statistical software used for ana-lysis was Sigma-Stat (version 3.5). P , 0.05 was considered statisticallysignificant.
Results
Derivation and characterization of theheart failure-human-induced pluripotentstem cellsDermal fibroblasts were obtained from two patients withadvanced HF due to ischaemic cardiomyopathy (both males, ages51 and 61). The fibroblasts were reprogrammed to generate theHF-hiPSCs by retroviral infection of three reprogrammingfactors: Oct4, Sox2, and Klf4. Control hiPSCs lines were generatedby reprogramming of foreskin fibroblasts of a healthy individual.
The reprogramming efficiency was evaluated by immunostainingfor the pluripotency marker Tra-1–60 after 21 days. The efficiencyranged between 0.014 and 0.02% (resulting in 15.5+2.1 colonies/105cells). This efficiency is higher than previously reported for thereprogramming of healthy fibroblasts with the retroviral or lenti-viral three-factor approaches.19 This improved efficiency can beexplained by the supplementation of the chromatin modifier val-proic acid, previously shown to enhance hiPSC reprogramming.20
For each patient, ten HF-hiPSC clones were generated, two ofwhich (HF2-5, HF2-8, HF3-2, and HF3-3) were continuously pro-pagated, characterized, and used for cardiomyocyte differentiation.All hiPSC clones exhibited the characteristic hESC morphology,
expressing the pluripotent markers NANOG, OCT4, Tra-I-60,Tra-I-81, and SSEA4 (Figure 1A, Supplementary material online,Figure S1A), and maintained a stable karyotype. Quantitative real-time PCR analysis revealed the re-expression of the endogenouspluripotency genes OCT4, NANOG, and SOX2 in the HF-hiPSCs,at similar levels to control hiPSCs (Figure 1B). Next, bisulfitegenome sequencing demonstrated that the NANOG promoterwas hypomethylated in both HF and control hiPSCs, in contrastto the hypermethylated state in the parental fibroblasts(Figure 1C, Supplementary material online, Figure S2A). Finally, injec-tion of control and HF-hiPSCs to immunodeficient SCID-beigemice led to the formation of teratomas, containing advancedtissue derivatives of all the three germ layers (Figure 2, Supplemen-tary material online, Figure S2B). Thus, our data show thatHF-hiPSCs are properly reprogrammed and pluripotent.
Cardiomyocyte differentiationThe two HF-hiPSC lines were coaxed to differentiate into cardio-myocytes using the EB differentiation system (Supplementarymaterial online, Movie S1). Immunostaining confirmed the cardiacphenotype of enzymatically dispersed cells derived from thesecontracting EBs. As depicted in Figure 2A, both control andHF-hiPSC-CMs demonstrated a typical striated pattern followingstaining for the sarcomeric proteins cTnI, cTnT, and a-actinin. Simi-larly, the gene expression analysis of differentiating hiPSC-CMs(Figure 2B) confirmed the expression of cardiac-specific transcrip-tion factors (NKX2–5) and structural genes including cTnI (TNNI),a and b myosin heavy chains (MYH6, MYH7), ventricular myosinlight chain (MLC-2V), and myosin light chain 2 atrial isoform(MYL7). This cardiac gene-expression profile was similar in bothhealthy and HF-hiPSC-CMs but was undetectable in undifferenti-ated hiPSCs, where the expression of pluripotency genes(OCT4 and NANOG) was dominant.
Figure 1 Heart failure-human-induced pluripotent stem cells derivation and characterization. (A) Phase-contrast and immunostainings ofhealthy control, HF2-8, and HF3-3 human-induced pluripotent stem cells colonies for the pluripotency markers: OCT4, NANOG, SSEA4,TRA-1-60, and TRA1-81. (scale bars: 20 mm). (B) Quantitative real-time PCR showing reactivation of the endogenous OCT4, SOX2, andKLF4 genes in control and heart failure human-induced pluripotent stem cell clones (HF2-5, HF2-8, HF3-2, and HF3-3) in comparison withthe parental fibroblasts. Values were normalized to b-tubulin and presented as mean+ SEM. Expression values are relative to the correspond-ing dermal fibroblasts. (C) Bisulfite sequencing of the NANOG promoter in control-human-induced pluripotent stem cell, HF2-8-hiPSCs,HF3-3-hiPSCs, and parental fibroblasts. Open and closed circles indicate unmethylated and methylated CpGs, respectively. (D) H&E stainingof teratomas formed following injection of control and heart failure human-induced pluripotent stem cells into SCID-beige mice. Note thepresence of pigmented epithelium (ectoderm), gastrointestinal epithelium (endoderm), and hyaline cartilage (mesoderm) (scale bars: 200 mm).
L. Zwi-Dantsis et al.Page 4 of 12
345
350
355
360
365
370
375
380
385
390
395
400
405
410
415
420
425
430
435
440
445
450
455
We next sought to compare the cardiac differentiationcapacity of the HF and control hiPSCs. As appreciated inFigure 2C, the average number of beating EBs/plate after30 days of differentiation was not significantly differentbetween all HF-hiPSC clones and control hiPSCs (P ¼ 0.673).To strengthen these results, we used quantitative real-timePCR for selected cardiac genes (GATA4, MYH6, MYH7, TNNI,and ACTC) as an additional method to compare the cardio-myocyte yield. These studies also revealed comparable
differentiation efficiencies between controls and HF-hiPSCclones (Figure 2D).
Functional characterization of heartfailure-human-induced pluripotent stemcell-cardiomyocytesA multi-electrode array (MEA) mapping system was used to studythe electrophysiological properties of the HF-hiPSC-CMs.
Figure 2 Cardiomyocyte differentiation of the human-induced pluripotent stem cell-cardiomyocytes. (A) Immuostaining of control and heartfailure-human-induced pluripotent stem cell-cardiomyocytes for sarcomeric-a-actinin, cTnI, and cTnT (scale bars: 50 mm). (B) Semi-quantitativeRT–PCR studies revealing the expression of NKX2-5, MYH-6, MYH-7, MLC-2v, TNNI, and MYL7 by heart failure and control human-inducedpluripotent stem cell-cardiomyocytes. Undifferentiated human-induced pluripotent stem cells, in contrast, robustly expressed OCT4 andNANOG. (C) Comparison of cardiomyocyte differentiation of control and heart failure human-induced pluripotent stem cells. Shown areaverage number of beating EBs/plate from two clones of each patient (HF2-5, n ¼ 4; HF2-8 n ¼ 8; HF3-2, n ¼ 5; HF3-3 n ¼ 6) and healthycontrol human-induced pluripotent stem cells (n ¼ 4). (D) Real-time quantitative PCR showing comparable expression levels of cardiogenicmarkers (MYH6, MYH7, TNNI, ACTC, GATA4) in heart failure and control human-induced pluripotent stem cell clones.
Cardiomyocyte differentiation of iPSC Page 5 of 12
460
465
470
475
480
485
490
495
500
505
510
515
520
525
530
535
540
545
550
555
560
565
570
Microdissected hiPSC-derived cardiac tissues were cultured on theMEA plates (Figure 3A). The extracellular potentials recorded fromelectrodes underlying these beating EBs (Figure 3B) were used togenerate detailed activation maps (Figure 3C). These maps revealedthe development of a functional cardiac syncytium, with the activa-tion wavefront propagating from the pacemaker area (red) to thelatest (dark blue) activation sites (Figure 3C). When evaluatingbeating frequencies, we found no significant differences betweenthe different HF and control hiPSC-CMs (n ¼ 10, P ¼ 0.28;Figure 3D).
We next assessed the response of HF-hiPSC-CMs to neurohu-moral stimulation. Administration of the b-agonist isoproterenol(1 mM) led to a comparable increase in the beating frequency ofcontrol, HF2-8, and HF3-3 hiPSC-CMs (by 37+ 10, 48+6, and55+8%, respectively; n ¼ 5; Figure 4A). Similarly, application ofthe muscarinic agonist carbamylcholine (1 mM) led to comparablenegative chronotropic responses (by 26+ 4, 45+7, and 32+11%, respectively; n ¼ 5; Figure 4B). The chronotropic changesinduced by isoproterenol and carbamylcholine (Figure 4) werestatistically significant for each of the hiPSC lines studied but themagnitudes of the effects did not differ between control andHF-hiPSC-CMs (P ¼ 0.63 for isoproterenol; P ¼ 0.26 forcarbamylcholine).
Generation of transgene-free heartfailure-human-induced pluripotentstem cellTo derive transgene-free HF-hiPSCs, we initially reprogrammedthe HF3 fibroblasts with an excisable single polycistronic lentiviralvector (STEMCCA) encoding all four factors (Oct4, Sox2, Klf4, andc-Myc).17 Fifteen clones were isolated and subjected to the South-ern blot analysis to determine the number of genomic integrationsites of the stem cell cassette. Eight of these clones contained asingle integration site, of which one (HFS-5-hiPSC) was continu-ously propagated.
The HFS-5-hiPSC line expressed the pluripotent markersNANOG, OCT4, Tra-I-60, Tra-I-81, and SSEA4 (Figure 5A) andmaintained a normal karyotype. Next, we aimed to remove theSTEMCCA cassette in order to generate HF-hiPSCs free of theintegrated transgenes. Since the STEMCCA cassette was designedto contain lox-P sites flanking the reprogramming transgenes, weattempted to excise the integrated vector by expression ofCre-recombinase. To this end, the HFS-5-hiPSCs were transientlytransfected with an expression vector encoding Cre-recombinaseand EGFP (pCAG-Cre-EGFP). Cre-recombinase expressing cells(identified by eGFP) were then selected from a single-cell
Figure 3 Multi-electrode array recordings. (A and B) Extracellular recordings (B) from the heart failure-human-induced pluripotent stem cell-derived cardiac tissue cultured on the MEA plate (A). (C ) The resulting activation map. (D) Comparison of beating rates between control andheart failure human-induced pluripotent stem cell-cardiomyocytes.
L. Zwi-Dantsis et al.Page 6 of 12
575
580
585
590
595
600
605
610
615
620
625
630
635
640
645
650
655
660
665
670
675
680
suspension by FACS sorting 72 h after electroporation. Selectedcells were re-plated at low density and the generated cloneswere isolated 14 days later. Successful excision of the STEMCCAcassette was confirmed by PCR analysis of genomic DNA usingprimers targeting the WPRE sequence of the lentiviral vector(Figure 5B).
In a similar manner to the HF-hiPSCs derived by retroviral trans-duction, the HFS-5-hiPSC line could be differentiated into cardio-myocytes using the EB differentiation method. Semi-quantitativeRT–PCR studies (Figure 5C) revealed the expression of cardiac-specific genes by the differentiating EBs. Similarly, immunostaininganalysis for cTnT and of sarcomeric-a-actinin confirmed thecardiac identity of the differentiating myocytes (Figure 5D).
Finally, we also compared the cardiomyocyte differentiationcapacity of the generated transgene-free hiPSCs with that of thetransgene-containing HF-hiPSC clones (prior to Cre-excision).Comparable differentiation potentials were noted in both clonesas demonstrated by the similar quantitative expression ofcardiac-specific genes (MYH6, MYH7, TNNI, and ACTC) by thedifferentiating cells (Figure 5E).
In vitro integration with pre-existingcardiomyocyte culturesThe HF-hiPSC-derived cardiac tissues were co-cultured with pre-existing cardiac tissue (NRVMs) (Figure 6A). Within 24–48 h, we
could already detect synchronous mechanical activity (Supplemen-tary material online, Movie S2) between the human and rat cardio-myocytes (n ¼ 12). We next utilized the MEA mapping techniqueto evaluate the functional interactions within the co-cultures. Byrecording extracellular field potentials (Figure 6B) simultaneouslyfrom all electrodes, we were able to create activation maps thatshowed the development of synchronized activity in all hybrid cul-tures. Note in the example in Figure 6C that electrical activationinitiated within the rat tissue (red) and then propagated to therest of the co-culture, activating also the HF-hiPSC-CMs. Theelectrophysiological integration observed was continuous. Hence,electrograms recorded simultaneously from the human (blue elec-trode) and rat (red electrode) tissues showed synchronized activityand tight temporal coupling (Figure 6D) over several beats.
Next, immunostaining studies targeting the major gap junctionprotein Cx43 were performed in the co-cultures. As depicted inFigure 6E, a positive punctuate Cx43 immunosignal (suggestingthe development of gap junctions, yellow arrows) was identifiedat the border between the plated EB (containing theHF-hiPSC-CMs) and the NRVMs.
Finally, to evaluate the long-term coupling within theco-cultures, graphs were created, sequentially plotting the cyclelengths (CLs) of the electrical activity in the HF-hiPSC-derivedcardiac tissue vs. that in the NRVMs. Similarly, the activationtime difference between the two tissue types (DLAT) was mea-sured and plotted for each beat. As can be seen in Figure 6F, the
Figure 4 Pharmacological studies. Positive (A) and negative (B) chronotropic responses induced in the control and heartfailure-human-induced pluripotent stem cell-cardiomyocytes by adrenergic (isoproterenol, A) and muscarinic (carbamylcholine, B) stimulationrespectively. *P , 0.05 when compared with baseline values.
Cardiomyocyte differentiation of iPSC Page 7 of 12
long-term electrical synchronization in the co-cultures wasreflected by the establishment of fixed DLAT between the EBsand NRVMs (Figure 6E, left) and by convergence of their CLplots (linear correlation, Figure 6E right). This coupling persistedfor several days in culture.
In vivo tranplantation of the heartfailure-human-induced pluripotent stemcell-cardiomyocytesWe next evaluated the ability of the HF-hiPSC-CMs to engraft inthe in vivo heart. Beating EBs were mechanically dissected, enzyma-tically dispersed into small clusters, and transplanted into the LVmyocardium of healthy rat hearts. As can be viewed in Figure 7,the engrafted HF-hiPSC-CMs (pre-labelled with Qtracker, red)were identified within the rat myocardium in all cases, theirhuman origin verified by immunostaining with anti-human mito-chondrial antibodies (Figure 7A), and their cardiomyocyte identityconfirmed by the positive immunostaining for sarcomeric-a-actinin(Figure 7B). In a similar manner to previous hESC-CMs transplant-ation studies,6 the hiPSC-CMs at 7–10 days following engraftment
were smaller in the size than host rat cardiomyocytes, werearranged isotropically, and still displayed an early-striated stainingpattern. Finally, to evaluate for structural coupling between hostand donor cells, immunostainings were performed for Cx43(Figure 7C). The resulting punctuate Cx43-immunosignal (white),identified at the interphase between the HF-hiPSC-CMs and ratcardiomyocytes (red arrows), suggests the formation of gapjunctions between host and donor cells.
DiscussionMyocardial cell replacement therapies have emerged as noveltherapeutic paradigms for myocardial repair but have been ham-pered by the paucity of sources for human cardiomyocytes, bythe lack of direct evidence for functional integration betweendonor and host cells, and by the anticipated immune rejectionassociated with allogeneic cell transplantation. In this study, weaimed to target these obstacles by establishing and coaxing thecardiomyocyte differentiation of hiPSCs from patients withischaemic cardiomyopathy.
Figure 5 Generation of transgene-free heart failure-human-induced pluripotent stem cell. (A) Phase-contrast and immunostaining ofHFS-5-hiPSC colonies for the pluripotency markers OCT4, NANOG, SSEA4, TRA-1-60, and TRA1-81 (scale bars: 50 mm). (B) GenomicPCR (using primer set flanking the WPRE region of the vector) verifying the excision of the lentiviral vector, originally present in theHFS-5-hiPSC clones. Note that clone Cre-HFS-5-hiPSCs showed no detectable WPRE in the genomic DNA. HF3-hiPSC and HF2-hiPSClines, generated by retroviral infection were used as negative controls. (C) Semi-quantitative RT–PCR studies demonstrating the expressionof pluripotent markers (OCT4 and NANOG) in undifferentiated cells and the expression of cardiac-specific markers (NKX2–5, MYH-6,MYH-7, MLC-2v, TNNI, MYL7) by the differentiating EBs. (D) Positive staining of HFS-5-hiPSC-CMs for a-actinin and cTnT (scale bars:50 mm). (E) Real-time quantitative PCR showing comparable expression levels of cardiogenic genes (MYH6, MYH7, TNNI, ACTC) during thedifferentiation of the transgene-free human-induced pluripotent stem cells derived cardiomyocytes (Cre-HFS-5-hiPSC-CMs) and the transgene-containing clone derived cardiomyocytes (HFS-5-hiPSC-CMs).
L. Zwi-Dantsis et al.Page 8 of 12
800
805
810
815
820
825
830
835
840
845
850
855
860
865
870
875
880
885
890
895
900
905
910
Figure 6 Functional and structural integration in co-cultures. (A) Co-culturing of the heart failure-human-induced pluripotent stem cell-derived cardiac tissue (arrows) with the NRVMs. (B and C) MEA recordings (B) and the resulting colour-coded activation map (C) showingthe development of functional integration, with electrical activity originating (red) in the rat tissue and propagating to the human cardiomyo-cytes. (D) Extracellular potentials, recorded from two electrodes underlying the heart failure-human-induced pluripotent stem cell-cardiomyocytes (red) and NRVMs (blue), showing synchronized activity. (E) Structural integration between the human-induced pluripotentstem cell-cardiomyocytes and NRVMs. Left panel: DAPI and DIC staining showing the hiPSC-derived cardiac tissue (EB) in the centre andNRVMs in the periphery. Right panel: High-magnification Cx43 immunostaining (red punctuate staining) image of the co-culture. Note the de-velopment of gap junctions (yellow arrows) at the interphase between the human-induced pluripotent stem cell-cardiomyocytes and NRVMs.(F) DLAT (left) and CL (right) plots showing long-term synchronized activity between the heart failure-human-induced pluripotent stem cell-cardiomyocytes and NVRMs at 1 and 4 days.
Cardiomyocyte differentiation of iPSC Page 9 of 12
915
920
925
930
935
940
945
950
955
960
965
970
975
980
985
990
995
1000
1005
1010
1015
1020
1025
Our results show: (i) that hiPSCs can be established from dermalfibroblasts of patients with advanced HF using a three-factor repro-gramming approach that does not include cMyc; (ii) that similartransgene-free HF-hiPSCs can be derived using an excisable poly-cistronic lentiviral vector; (iii) that the HF-hiPSCs can be differen-tiated into cardiomyocytes with appropriate molecular, structural,and functional properties; (iv) that the reprogramming efficiency,hiPSC cardiomyocyte differentiation capacity, and phenotype prop-erties of the generated hiPSC-CMs are comparable when usingfibroblasts from HF patients or healthy foreskin fibroblasts; (v)that the HF-hiPSC-CMs can functionally integrate with pre-existingcardiac tissue in a co-culture model; and (vi) that theHF-hiPSC-CMs can engraft, survive, and integrate structurallywith host cardiac tissue following in vivo transplantation.
Differentiation, once believed to be a one-way process, was re-cently shown to be a dynamic process that can be reversed bytransduction of stemness transcription factors into somatic cells.9
The reprogrammed cells, also known as iPSCs, highly resembleESCs in terms of morphology, gene expression, pluripotency,and epigenetic status. The added value of the iPSCs lies in the un-precedented opportunity to generate patient-specific pluripotent
stem cells from adult individuals. In this study, we contributed tothis rapidly developing field by showing, for the first time, theability to establish hiPSCs from ischaemic cardiomyopathy patientswith advanced HF, who represent the typical candidates for futureautologous myocardial cell replacement procedures. We thendemonstrated that the patient-specific hiPSC lines fulfil the criteriadefining fully reprogrammed pluripotent hiPSCs and that they candifferentiate into bone fide cardiomyocytes.
We next compared the HF-hiPSC-CMs to cardiomyocytesderived from healthy control hiPSCs and found comparable prop-erties in terms of cardiomyocyte differentiation efficiency, beatingrates, up-regulation of cardiac genes and down-regulation of plur-ipotency genes, structural organization, electrical activity, and re-sponse to neurohormonal triggers. Moreover, the differentiationsystem in both hiPSC types was not limited to the generation ofisolated cardiomyocytes, but rather a functional cardiac syncytiumwas generated.
The ‘healthy’ cardiogenic phenotype observed in theHF-hiPSC-CMs is not surprising given that the disease process inthese HF patients was acquired. Hence, the patients studied didnot have any abnormal genetic background that could impact the
Figure 7 In vivo transplantation of the heart failure-human-induced pluripotent stem cell-cardiomyocytes. (A and B) The grafted heartfailure-human-induced pluripotent stem cell-cardiomyocytes (pre-labelled with Qtracker, red) were identified within the rat myocardium,their human origin verified by immunostaining with anti-human mitochondrial antibodies (green, A), and their cardiomyocyte phenotype con-firmed by the positive immunostaining for sarcomeric-a-actinin (green, B). Note the early striated pattern of the engrafted human-inducedpluripotent stem cell-cardiomyocytes (B). Scale bars (A—100 mm, B—50 mm). (C) Development of gap junctions between donorhuman-induced pluripotent stem cell-cardiomyocytes and host rat cardiomyocytes. Left panel: Immunostaining for sarcomeric-a-actinin(green) identifying the rat host cardiomyocytes (in cross section) as larger and structurally more mature cells (R); when compared with theengrafted human-induced pluripotent stem cell-cardiomyocytes (H ), which display an early-striated staining pattern. Note that several ofthe engrafted cells are labelled with Qtracker (red). Right panel: Immunostaining of the same specimen for Cx43. Note the positive punctuateCx43 immunosignal (white) between the transplanted human-induced pluripotent stem cell-cardiomyocytes themselves and also at the inter-face between donor and host cardiomyocytes (red arrows) (scale bars: 50 mm).
L. Zwi-Dantsis et al.Page 10 of 12
1030
1035
1040
1045
1050
1055
1060
1065
1070
1075
1080
1085
1090
1095
1100
1105
1110
1115
1120
1125
1130
1135
1140
properties of the reprogrammed fibroblasts or their differentiatedcardiomyocyte progeny. This is in contrast to recent hiPSC studiesin inherited monogenic cardiac disorders (such as long QT syn-drome)14– 16 in which hiPSC-CMs were established from dermalfibroblasts carrying the mutation, consequentially leading to adisease-specific abnormal phenotype.
Another important finding of the current study is the ability toproduce hiPSCs and to differentiate them into cardiomyocyteswithout c-Myc. One of the major hurdles that may hamper theintroduction of the hiPSC technology into the clinic is the potentialrisk for tumorgenicity.21 This concern stems from the initialrequirement for the use of oncogenic transcription factors (suchas cMyc) for reprogramming, the random transgene insertioninto the host’s genome, and the use of viral vectors. Okitaet al.,22 for example, reported that the chimeras and progenyderived from mouse iPSCs frequently showed tumour formation;and that in these tumours, retroviral expression of cMyc was reac-tivated. In contrast, when cMyc-free-iPSC lines were used, notumours were found in the offspring.
Beyond the use of a cMyc-free reprogramming approach, severalstrategies are being developed to generate transgene-free hiPSCs,which are further expected to decrease the tumorogenic risk aswell as potential alterations in hiPSC characteristics because of re-sidual transgene expression. Such strategies can be grouped into:(i) methods that use non-integrating DNA vectors such as adeno-viruses,23 conventional plasmids,24 and minicircles; (ii) DNA-freemethods such as by the administration of synthetic-modifiedmRNA25 or protein delivery;26 and (iii) methods that use integrat-ing but excisable systems in which the transgenes are removedfrom the genome through the piggyBac transposon/transposasesystem27 or by the Cre/loxP approach.17
In the current study, we used the latter strategy to derivetransgene-free hiPSCs from the HF patients. Our protocol con-sisted of reprogramming the HF fibroblasts into hiPSCs with anexcisable single polycistronic lentiviral vector (STEMCA) that con-tained all four factors, screening to select the hiPSC clones thatpossess a single integration site, and subsequent excision of thevector by transient introduction of Cre-recombinase.
Another important finding of this study is the ability of theHF-hiPSC-CMs to integrate electrically and mechanically with pre-existing cardiac tissue. Numerous reports suggest that cell trans-plantation can improve the cardiac performance in animalmodels of myocardial infarction.1,2 However, it is not entirelyclear whether this functional improvement is due to the directcontribution to contractility by the transplanted myocytes or by al-ternative indirect mechanisms such as attenuation of the remodel-ling process, amplification of an endogenous repair process, orinduction of angiogenesis. True systolic augmentation wouldrequire structural, electrophysiological, and mechanical couplingof donor and host tissue so that the transplanted cell graftwould participate actively in the synchronous contraction ofthe ventricle.
The observations that HF-hiPSC-CMs could couple with hostcardiomyocytes and generate a single functional syncytium in vitroand to form a gap junction with host cardiomyocytes in vivo arein agreement with previous studies showing the ability of cardio-myocytes derived from embryonic hearts or ESCs to couple
both in vitro and in vivo with host cardiomyocytes.18,28– 30 This cap-ability may be crucial for future use of these cells not only for thetreatment of HF but also for conduction system repair (biologicalpacemaker approach).18
Nevertheless, long-term application of the hiPSC-CMs relymore on the prospect of in vivo use. In this study, we demonstratedthat HF-hiPSC-CMs can survive following transplantation, formstable cell grafts, and establish electromechanical junctions withhost cardiomyocytes in the healthy rat heart. Future studies willhave to evaluate whether similar engraftment could also occur insmall and large animal models of cardiac injury and whether suchan engraftment could lead to a functional benefit. To conductsuch studies, however, more efficient cardiomyocyte differenti-ation systems and scaling up procedures should be developed(as already reported for hESCs)7,31,32 to derive a clinically relevantnumber of purified cardiomyocytes.
Supplementary materialSupplementary material is available at European Heart Journalonline.
AcknowledgementsWe thank Dr Sara Selig and Dr Annie Rebibo-Sabbah for their helpin bisulfite sequencing and southern blot analyses. We thankDr Edith Suss-Toby for her assistance with confocal imaging andDr Ofer Shenker and Yaakov Sakoury for their assistance withflow cytometry.
FundingThis work was supported in part by the Israel Science Foundation(1449/10, Mina and Oto Shpirman fund); by the European ResearchCouncil (ERC-2010-StG-260830-Cardio-iPS); and by the J&J-Technionresearch grant; and by the Nancy and Stephen Grand philanthropicfund.
Jones JM. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147.
4. He JQ, Ma Y, Lee Y, Thomson JA, Kamp TJ. Human embryonic stem cells developinto multiple types of cardiac myocytes: action potential characterization. Circ Res2003;93:32–39.
5. Kehat I, Kenyagin-Karsenti D, Snir M, Segev H, Amit M, Gepstein A, Livne E,Binah O, Itskovitz-Eldor J, Gepstein L. Human embryonic stem cells can differen-tiate into myocytes with structural and functional properties of cardiomyocytes.J Clin Invest 2001;108:407–414.
6. Caspi O, Huber I, Kehat I, Habib M, Arbel G, Gepstein A, Yankelson L,Aronson D, Beyar R, Gepstein L. Transplantation of human embryonic stem cell-derived cardiomyocytes improves myocardial performance in infarcted rat hearts.J Am Coll Cardiol 2007;50:1884–1893.
7. Laflamme MA, Chen KY, Naumova AV, Muskheli V, Fugate JA, Dupras SK,Reinecke H, Xu C, Hassanipour M, Police S, O’Sullivan C, Collins L, Chen Y,Minami E, Gill EA, Ueno S, Yuan C, Gold J, Murry CE. Cardiomyocytes derivedfrom human embryonic stem cells in pro-survival factors enhance function ofinfarcted rat hearts. Nat Biotechnol 2007;25:1015–1024.
Cardiomyocyte differentiation of iPSC Page 11 of 12
8. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, Yamanaka S.Induction of pluripotent stem cells from adult human fibroblasts by definedfactors. Cell 2007;131:861–872.
9. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embry-onic and adult fibroblast cultures by defined factors. Cell 2006;126:663–676.
10. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, Frane JL, Tian S, Nie J,Jonsdottir GA, Ruotti V, Stewart R, Slukvin II, Thomson JA. Induced pluripotentstem cell lines derived from human somatic cells. Science 2007;318:1917–1920.
11. Haase A, Olmer R, Schwanke K, Wunderlich S, Merkert S, Hess C, Zweigerdt R,Gruh I, Meyer J, Wagner S, Maier LS, Han DW, Glage S, Miller K, Fischer P,Scholer HR, Martin U. Generation of induced pluripotent stem cells fromhuman cord blood. Cell Stem Cell 2009;5:434–441.
12. Zhang J, Wilson GF, Soerens AG, Koonce CH, Yu J, Palecek SP, Thomson JA,Kamp TJ. Functional cardiomyocytes derived from human induced pluripotentstem cells. Circ Res 2009;104:e30–e41.
13. Zwi L, Caspi O, Arbel G, Huber I, Gepstein A, Park IH, Gepstein L. Cardiomyo-cyte differentiation of human induced pluripotent stem cells. Circulation 2009;120:1513–1523.
14. Itzhaki I, Maizels L, Huber I, Zwi-Dantsis L, Caspi O, Winterstern A, Feldman O,Gepstein A, Arbel G, Hammerman H, Boulos M, Gepstein L. Modelling the longQT syndrome with induced pluripotent stem cells. Nature 2011;471:225–229.
15. Matsa E, Rajamohan D, Dick E, Young L, Mellor I, Staniforth A, Denning C. Drugevaluation in cardiomyocytes derived from human induced pluripotent stemcells carrying a long QT syndrome type 2 mutation. Eur Heart J 2011;32:952–962.
16. Moretti A, Bellin M, Welling A, Jung CB, Lam JT, Bott-Flugel L, Dorn T, Goedel A,Hohnke C, Hofmann F, Seyfarth M, Sinnecker D, Schomig A, Laugwitz KL. Patient-specific induced pluripotent stem-cell models for long-QT syndrome. N Engl JMed 2010;363:1397–1409.
17. Sommer CA, Sommer AG, Longmire TA, Christodoulou C, Thomas DD,Gostissa M, Alt FW, Murphy GJ, Kotton DN, Mostoslavsky G. Excision of repro-gramming transgenes improves the differentiation potential of iPS cells generatedwith a single excisable vector. Stem Cells 2010;28:64–74.
18. Kehat I, Khimovich L, Caspi O, Gepstein A, Shofti R, Arbel G, Huber I, Satin J,Itskovitz-Eldor J, Gepstein L. Electromechanical integration of cardiomyocytesderived from human embryonic stem cells. Nat Biotechnol 2004;22:1282–1289.
19. Nakagawa M, Koyanagi M, Tanabe K, Takahashi K, Ichisaka T, Aoi T, Okita K,Mochiduki Y, Takizawa N, Yamanaka S. Generation of induced pluripotent stemcells without Myc from mouse and human fibroblasts. Nat Biotechnol 2008;26:101–106.
20. Huangfu D, Maehr R, Guo W, Eijkelenboom A, Snitow M, Chen AE, Melton DA.Induction of pluripotent stem cells by defined factors is greatly improved by small-molecule compounds. Nat Biotechnol 2008;26:795–797.
21. Jalving M, Schepers H. Induced pluripotent stem cells: will they be safe? Curr OpinMol Ther 2009;11:383–393.
22. Okita K, Ichisaka T, Yamanaka S. Generation of germline-competent inducedpluripotent stem cells. Nature 2007;448:313–317.
23. Stadtfeld M, Nagaya M, Utikal J, Weir G, Hochedlinger K. Induced pluripotentstem cells generated without viral integration. Science 2008;322:945–949.
24. Okita K, Nakagawa M, Hyenjong H, Ichisaka T, Yamanaka S. Generation ofmouse induced pluripotent stem cells without viral vectors. Science 2008;322:949–953.
25. Warren L, Manos PD, Ahfeldt T, Loh YH, Li H, Lau F, Ebina W, Mandal PK,Smith ZD, Meissner A, Daley GQ, Brack AS, Collins JJ, Cowan C,Schlaeger TM, Rossi DJ. Highly efficient reprogramming to pluripotency and direc-ted differentiation of human cells with synthetic modified mRNA. Cell Stem Cell2010;7:618–630.
26. Zhou H, Wu S, Joo JY, Zhu S, Han DW, Lin T, Trauger S, Bien G, Yao S, Zhu Y,Siuzdak G, Scholer HR, Duan L, Ding S. Generation of induced pluripotent stemcells using recombinant proteins. Cell Stem Cell 2009;4:381–384.
27. Woltjen K, Michael IP, Mohseni P, Desai R, Mileikovsky M, Hamalainen R,Cowling R, Wang W, Liu P, Gertsenstein M, Kaji K, Sung HK, Nagy A. piggyBactransposition reprograms fibroblasts to induced pluripotent stem cells. Nature2009;458:766–770.
28. Gepstein L, Ding C, Rehemedula D, Wilson EE, Yankelson L, Caspi O,Gepstein A, Huber I, Olgin JE. In vivo assessment of the electrophysiological inte-gration and arrhythmogenic risk of myocardial cell transplantation strategies. StemCells 2010;28:2151–2161.
29. Halbach M, Pfannkuche K, Pillekamp F, Ziomka A, Hannes T, Reppel M,Hescheler J, Muller-Ehmsen J. Electrophysiological maturation and integrationof murine fetal cardiomyocytes after transplantation. Circ Res 2007;101:484–492.
30. Rubart M, Pasumarthi KB, Nakajima H, Soonpaa MH, Nakajima HO, Field LJ.Physiological coupling of donor and host cardiomyocytes after cellular transplant-ation. Circ Res 2003;92:1217–1224.
31. Kirouac DC, Zandstra PW. The systematic production of cells for cell therapies.Cell Stem Cell 2008;3:369–381.
32. Yang L, Soonpaa MH, Adler ED, Roepke TK, Kattman SJ, Kennedy M,Henckaerts E, Bonham K, Abbott GW, Linden RM, Field LJ, Keller GM. Humancardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population. Nature 2008;453:524–528.
