UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cardiomyocyte differentiation with cyclic mechanical strain Zhao, Eric Jiahua 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2015_september_zhao_jiahua.pdf [ 7.31MB ]
JSON: 24-1.0166313.json
JSON-LD: 24-1.0166313-ld.json
RDF/XML (Pretty): 24-1.0166313-rdf.xml
RDF/JSON: 24-1.0166313-rdf.json
Turtle: 24-1.0166313-turtle.txt
N-Triples: 24-1.0166313-rdf-ntriples.txt
Original Record: 24-1.0166313-source.json
Full Text

Full Text

Cardiomyocyte Differentiation With Cyclic Mechanical StrainbyEric Jiahua ZhaoBSc., McGill University, 2012A THESIS SUBMITTED IN PARTIAL FULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OFMaster of Applied ScienceinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES(Biomedical Engineering)The University of British Columbia(Vancouver)June 2015c© Eric Jiahua Zhao, 2015AbstractThe creation of cardiomyocytes from pluripotent stem cells has the potential to revolutionize thetreatment of heart disease, the discovery and testing of drugs, and our understanding of humanphysiology. Thus far, differentiation protocols have mainly focused on recapitulating the biochemi-cal signalling events of cardiac organogenesis, and the resulting cardiomyocytes exhibit a fetal-likephenotype. We postulated that cyclic mechanical strain can mimic the mechanical environment ofthe developing heart, and, when applied in conjunction to biochemical differentiation protocols, canincrease the efficiency of differentiation and the maturity of the resulting cells. Using an inducedpluripotent stem cell line derived from human fibroblasts, we derived spontaneously contractingcardiomyocytes via treatment with activin A and bone morphogenetic protein 4 (BMP4). We ob-served that differentiation of induced pluripotent stem cells to cardiomyocytes is sensitive to cyclicstrain, with the application of 5% continuous cyclic strain at 1Hz having the effect of inhibitingspontaneous contractions and disrupting sarcomere formation.iiPrefaceThis thesis is original, unpublished, independent work by the author, E.J. Zhao.iiiTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiiGlossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xAcknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Cardiac regeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Cardiac development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 Gene regulatory networks for heart development . . . . . . . . . . . . . . . . . . . 61.3.1 The “kernel” of cardiac regulatory genes . . . . . . . . . . . . . . . . . . 61.3.2 The gene-to-structure relationship . . . . . . . . . . . . . . . . . . . . . . 71.3.3 Upstream control of the cardiac gene kernel . . . . . . . . . . . . . . . . . 71.4 Induced pluripotent stem cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.5 Transdifferentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.6 Defining characteristics of cardiomyocytes . . . . . . . . . . . . . . . . . . . . . . 111.7 Mechanical signalling in cell differentiation . . . . . . . . . . . . . . . . . . . . . 111.7.1 Mechanisms of mechanotransduction . . . . . . . . . . . . . . . . . . . . 111.7.2 Previous works involving mechanical strain . . . . . . . . . . . . . . . . . 121.8 Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14iv2 Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.1 Cell culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.1.1 Cell adhesion coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.1.2 BJ fibroblast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162.1.3 Celprogen “primary cardiomyocytes” . . . . . . . . . . . . . . . . . . . . 162.1.4 Induced pluripotent stem cells . . . . . . . . . . . . . . . . . . . . . . . . 162.1.5 Differentiation of iPSCs to cardiomyocytes . . . . . . . . . . . . . . . . . 172.2 Plasmids and transfection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2.1 Transfection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2.2 Reporter for cardiac-specific myosin heavy chain . . . . . . . . . . . . . . 182.2.3 Vehicle for introducing homo sapiens v-ets erythroblastosis virus E26 onco-gene homolog 2 (avian) (ETS2) and mesoderm posterior 1 (MESP1) . . . . 192.3 Lentivirus production and transduction . . . . . . . . . . . . . . . . . . . . . . . . 212.3.1 Lentivirus production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212.3.2 Lentivirus transduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.4 Immunocytochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.5 Western blots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232.6 Flow cytometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Cyclic strain device for tissue culture . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.1 Cyclic strain in the heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2 Previous devices and their limitations . . . . . . . . . . . . . . . . . . . . . . . . 243.3 Design of the HexCycler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.3.1 DC motor control with the Arduino . . . . . . . . . . . . . . . . . . . . . 283.4 Characterizing strain with finite-element analysis . . . . . . . . . . . . . . . . . . 293.4.1 The Ogden model for hyperelastic materials . . . . . . . . . . . . . . . . . 323.4.2 Model results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.5 Validating the effects of the HexCycler’s strain on cultured fibroblasts . . . . . . . 334 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384.1 Validating the α myosin heavy chain (αMHC) reporter . . . . . . . . . . . . . . . 384.1.1 Thyroid hormone fails to activate αMHC reporter transduced by lentivirus 384.1.2 Thyroid hormone fails to activate αMHC reporter transfected using nucle-oporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394.1.3 The αMHC reporter is activated by induced pluripotent stem cell (iPSC)-derived cardiomyocytes . . . . . . . . . . . . . . . . . . . . . . . . . . . 404.2 Primary cardiomyocytes lose core phenotypes in vitro . . . . . . . . . . . . . . . . 424.3 Cyclic strain affects iPSC differentiation to cardiomyocytes . . . . . . . . . . . . . 42v4.3.1 Cardiomyocyte differentiation is sensitive to cyclic strain parameters . . . . 455 Discussion and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485.1 Reporters of cardiac-like myocytes . . . . . . . . . . . . . . . . . . . . . . . . . . 485.2 The effects of cyclic strain on cardiomyocyte differentiation and maturation . . . . 495.3 Potential applications of in vitro cardiomyocytes and future directions . . . . . . . 505.3.1 Using iPSC-derived cardiomyocytes to model cardiomyopathies . . . . . . 505.3.2 Mature cardiomyocytes for drug toxicity testing . . . . . . . . . . . . . . . 505.3.3 Cellular therapy for heart failure . . . . . . . . . . . . . . . . . . . . . . . 515.3.4 Tissue engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515.3.5 Future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54A Design and fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64A.1 Drawings of HexCycler components . . . . . . . . . . . . . . . . . . . . . . . . . 64B Tissue culture protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74B.1 iPSC maintenance and differentiation protocols . . . . . . . . . . . . . . . . . . . 74B.1.1 Aliquoting Matrigel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74B.1.2 Plating Matrigel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75B.1.3 Preparing complete mTeSR1 medium . . . . . . . . . . . . . . . . . . . . 75B.1.4 Preparing rho-associated, coiled-coil containing protein kinase (ROCK) in-hibitor stock solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76B.1.5 Thawing iPSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76B.1.6 Feeding iPSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78B.1.7 Passaging iPSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78B.1.8 Freezing iPSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80B.1.9 Preparing Activin A and BMP4 stock solutions . . . . . . . . . . . . . . . 81B.1.10 Preparation of B-27–supplemented media . . . . . . . . . . . . . . . . . . 81viList of TablesTable 2.1 PCR for amplification and cloning of ETS2 and MESP1 . . . . . . . . . . . . . 20Table 3.1 Ogden hyperelastic model parameters for PDMS . . . . . . . . . . . . . . . . . 33viiList of FiguresFigure 1.1 From oocyte to morula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Figure 1.2 Embryonic disc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Figure 1.3 Gastrulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Figure 1.4 The murine blastocyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Figure 1.5 Formation of the egg cylinder . . . . . . . . . . . . . . . . . . . . . . . . . . 5Figure 1.6 Signalling in the mouse embryo . . . . . . . . . . . . . . . . . . . . . . . . . 6Figure 1.7 Cardiac regulatory genes and phylogeny . . . . . . . . . . . . . . . . . . . . . 8Figure 1.8 Mechanisms of mechanotransduction . . . . . . . . . . . . . . . . . . . . . . 13Figure 2.1 Map of the αMHC reporter plasmid . . . . . . . . . . . . . . . . . . . . . . . 20Figure 2.2 Map of the pIRES2 plasmid vector . . . . . . . . . . . . . . . . . . . . . . . . 21Figure 3.1 Cardiac coordinate system . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25Figure 3.2 The Flexcell R© FX-5000TM Tension System . . . . . . . . . . . . . . . . . . . 26Figure 3.3 The magnetically actuated cellular strain assessment tool (MACSAT) . . . . . 27Figure 3.4 The Scotch yoke mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Figure 3.5 The HexCycler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29Figure 3.6 HexCycler membrane motion . . . . . . . . . . . . . . . . . . . . . . . . . . 30Figure 3.7 HexCycler frequency response . . . . . . . . . . . . . . . . . . . . . . . . . . 31Figure 3.8 BioFlex membrane coordinate system . . . . . . . . . . . . . . . . . . . . . . 31Figure 3.9 FEA model of the BioFlex membrane . . . . . . . . . . . . . . . . . . . . . . 32Figure 3.10 Principal strains as a function of distance away from the centre of the culture well. 34Figure 3.11 Principal strain directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Figure 3.12 First principal strain across the BioFlex tissue culture membrane for piston dis-placements of 4mm to 10mm. . . . . . . . . . . . . . . . . . . . . . . . . . . 36Figure 3.13 Fibroblast alignment in response to cyclic strain . . . . . . . . . . . . . . . . . 36Figure 3.14 Fibroblast alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37Figure 4.1 T3 treatment of BJ fibroblasts transduced with the αMHC reporter . . . . . . . 39viiiFigure 4.2 T4 treatment of BJ fibroblasts transduced with the αMHC reporter . . . . . . . 40Figure 4.3 T4 treatment of BJ fibroblasts transfected with the αMHC reporter . . . . . . . 41Figure 4.4 iPSC-cardiomyocytes transfected with the αMHC reporter . . . . . . . . . . . 43Figure 4.5 Transfection of αMHC reporter in Celprogen cardiomyocytes. . . . . . . . . . 44Figure 4.6 Western blot assay for cardiac troponin T (cTnT) in Celprogen human car-diomyocytes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Figure 4.7 Cardiac differentiation protocol with strain. . . . . . . . . . . . . . . . . . . . 46Figure 4.8 A spontaneously beating cluster of cardiomyocytes. . . . . . . . . . . . . . . . 46Figure 4.9 Immunocytochemistry of iPSC-derived cardiomyocytes. . . . . . . . . . . . . 47ixGlossary5’-UTR 5’ untranslated region, a segment of DNA that is not used to code for amino acids,upstream of the actual amino acid sequenceαMHC α myosin heavy chainATP adenosine triphosphatebFGF basic fibroblast growth factorβMHC β myosin heavy chainBMP bone morphogenetic grotein, a group of growth factors that play roles in the patterning oftissues during developmentBMP4 bone morphogenetic protein 4BSA bovine serum albuminCAT chloramphenicol acetyltransferaseCER1 CerberuscTnT cardiac troponin TCMV human cytomegalovirusDAPI 4’,6-diamidino-2-phenylindole, a fluorescent stain that binds strongly to DNADC direct currentDIC differential interference contrastDKK1 Dickkopf homologue 1DMEM Dulbecco’s Modified Eagle’s MediumECM extracellular matrixxEDTA ethylenediaminetetraacetic acid, used as a chelating agent in tissue cultureEGF epidermal growth factorEOMES eomesodermin, also known as T-box brain protein 2 (TBR2)ESRRG estrogen-related receptor gammaETS2 homo sapiens v-ets erythroblastosis virus E26 oncogene homolog 2 (avian)EGFP enhanced green fluorescent proteinFBS fetal bovine serum, a common adjuvant to tissue culture mediaFEA finite element analysisFGF fibroblast growth factorFGF2 fibroblast growth factor 2 (basic)FITC fluoresceine isothiocyanate, a fluorescent molecule with emission and excitationwavelengths similar to EGFPGATA a family of transcription factors named for their ability to bind the DNA sequence“GATA”GATA4 GATA binding protein 4GFP green fluorescent proteinGMT GATA4, MEF2c, and TBX5, the three factors used by Ieda et al. [34] to transdifferentiatefibroblasts to cardiomyocyte-like cellsHAND heart- and neural crest derivatives-expressed, a family of genes that regulate ventriculardevelopmentHAND1 heart- and neural crest derivatives-expressed protein 1HAND2 heart- and neural crest derivatives-expressed protein 2HCN4 potassium/sodium hyperpolarization-activated cyclic nucleotide-gated channel 4hERG human ether-a`-go-go-related gene, a gene that encodes a subunit of the ion channel whoseinhibition can lead to long QT syndromehESC human embryonic stem cellxiHSVTK herpes simplex virus thymidine kinaseIGF-1 insulin-like growth factor 1iPSC induced pluripotent stem cellIRES internal ribosomal entry siteJAK Janus kinase, a family of proteins with the ability to phosphorylate and activatedownstream proteinsKanr kanamycin resistanceKLF4 Kruppel-like factor 4 (gut), one of the four Yamanaka factors used to induce pluripotencyin somatic cellsLEFTY1 left-right determination factor 1LIN28 Lin-28 homolog AMACSAT magnetically actuated cellular strain assessment toolMCS multiple cloning siteMEF2 myocyte enahncer factor-2, a family of transcription factors known to control cellulardifferentiationMEF2C myocyte enhancer factor 2CmESC mouse embryonic stem cellMESP mesoderm posterior, a family of transcription factors known to initiate the network ofcardiac regulatory network of transcription factorsMESP1 mesoderm posterior 1MESP2 mesoderm posterior 2MHC myosin heavy chainMOSFET metal-oxide-semiconductor field-effect transistorMYC v-myc avian myelocytomatosis viral oncogene homolog, one of the four Yamanakafactors used to induce pluripotency in somatic cells (also known as c-Myc)MYOCD myocardin, a nuclear protein known to play a role in cardiogenesis and differentiation ofthe smooth muscle cell lineagexiiMYOD1 myogenic differentiation 1, the master regulatory of skeletal muscle differentiationNANOG NANOG, a transcription factor expressed by pluripotent stem cellsNeor neomycin resistanceNK2 A family of transcription factors involved in organogenesisNKX2-5 NK2 homeobox 5, a transcription factor known to play a role in heart developmentOCT4 octamer-binding transcription factor 4, a transcription factor expressed by pluripotentstem cells and one of the four Yamanaka factors used to induce pluripotency in somaticcells. Also known as OCT3, OCT3/4, and POU5F1 (POU class 5 homeobox 1PBS phosphate-buffered salinePCR polymerase chain reactionPDGF platelet-derived growth factorPDMS polydimethylsiloxanePI propidium iodide, a fluorescent moleculePKA C-α protein kinase A, catalytic subunit αROCK rho-associated, coiled-coil containing protein kinaseSMARCD3 SWI/SNF Related, Matrix Associated, Actin Dependent Regulator Of Chromatin,Subfamily D, Member 3SOX2 SRY (sex determining region Y)-box 2, one of the four Yamanaka factors used to inducepluripotency in somatic cells, known to regulate embryonic development maintain stemcells in the central nervous systemSRF serum response factorSV40 simian virus 40, a polyoma virus capable of infecting monkeys and humansRT-PCR reverse-transcriptase polymerase chain reactionT brachyury homolog (mouse)TBX T-box, a family of genes that possess the T-box DNA-binding domainTBX5 T-box 5xiiiTGF-β transforming growth factor βTHY1 thymocyte differentiation antigen 1ZFPM2 zinc finger protein, FOG Family Member 2xivAcknowledgmentsThis project was made possible by the guidance and support of my supervisors, Dr. Mu Chiao andDr. Chinten James Lim. I will be forever grateful for their advice and encouragement in regards tomy professional development.Dr. Christopher Maxwell’s suggestions were crucial in inspiring new avenues of inquiry, cat-alyzing a collaboration with Sanam Shafaattalab and Dr. Glen Tibbits of Simon Fraser University,who were exceedingly generous in offering their expertise.My lab mates Farzad Khademolhosseini, Eva Yap, Dan He, Jack Liu, and Pascal Leclair, as wellas Helen Chen and Jihong Jiang of the Maxwell lab were instrumental in getting me up to speed inall manners of experimental techniques.I would like to thank everyone affiliated with UBC’s Engineers in Scrubs program, which pro-vided funding and invaluable first-hand experiences in biomedical device development. I also ac-knowledge the Natural Sciences and Engineering Research Council of Canada for providing fundingvia the Canada Graduate Scholarships-Master’s award.Special thanks are reserved for my parents and my partner, for moral support and understandingwhen feeding my cells took precedence over making dinner.xvChapter 1Introduction1.1 Cardiac regenerationWhy can some animals regenerate entire lost limbs while humans cannot hope to regrow even afingertip? Why do some human tissues – liver, for example – possess drastically higher regenerativecapacity than others? What if we could heal heart damage by growing new, functional heart tissue?If we could better model genetic cardiomyopathies by growing a patient’s cardiomyocytes in vitro,could we find new, more effective treatments? How much time and money could we save if we hadphenotypically mature cardiomyocytes on which to test candidate drugs for cardiotoxicity?These are the questions that drove me to complete the work presented herein.1.2 Cardiac developmentTo attempt cardiac regeneration, we must first have a thorough understanding of cardiac develop-ment during embryogenesis and fetal development. Let us embark upon a survey of embryology,with emphasis on the pre-cardiac mesoderm [53, 71].The fertilized egg, known as the zygote, is a single cell that undergoes mitotic division withoutcell growth to produce a ball of cells known as the morula. In a process known as compaction, theblastomeres form a spherical shell, known as the trophoblast, with a lump on the inside, known as theinner cell mass (Figure 1.1). The trophoblast then forms a single-layered epithelium and the spherefills with fluid. The trophoblast eventually develops into the fetal portion of the placenta, while theinner cell mass gives rise to the embryo. The embryo at this stage is known as the blastocyst. Anatural question emerges: how do the uniform clump of cells that comprise the morula know whichto become, the trophoblast or the inner cell mass? Experimenters have shown that transcriptionfactors play a role in the specification of the fate of blastomeres. In the uniform morula, all cellsexpress octamer-binding transcription factor 4 (OCT4) and NANOG, and indeed the expression ofthese transcription factors is used as a marker of the pluripotency of stem cells. However, OCT41A B CED FFigure 1.1: A schematic of human zygote development from ovulation to the hatched blasto-cyst. A: Oocyte with polar body surrounded by zona pellucida (grey border). B: Two-cellstage. C: Four-cell stage. D: Eight-cell stage. E: Sixteen-cell stage (morula). F: Blasto-cyst with inner cell mass (pink) surrounded by trophoblast (blue).and NANOG are downregulated in the cells that make up the trophoblast, while it is maintained inthe inner cell mass.In the next stage of development, the inner cell mass (also known as embryoblast) splits into2 layers: the epiblast and the hypoblast (also known as primitive endoderm). Together these twolayers comprise the bilaminar embryonic disc, which will later give rise to the germ layers of theembryo. Almost simultaneously, the amniotic cavity begins to form in the epiblast, lined with a layerof cells derived from the epiblast (Figure 1.2). Next, the two-layered embryonic disc gains a thirdlayer in the process of gastrulation, giving rise to the three germ layers – endoderm, mesoderm, andectoderm. Correct morphogenesis is achieved through tight spatiotemporal regulation of signallingmolecules such as bone morphogenetic grotein (BMP)s, fibroblast growth factor (FGF)s, and Wntproteins. Researchers have used these signalling proteins to great effect in the in vitro differentiation2Amniotic CavityBlastocoelAmniogenic Cells EpiblastHypoblastTrophoblastFigure 1.2: Soon after the blastocyst implants into the endometrium, the inner cell mass formsa circular, two-layered disc called the embryonic disc. The two types of cells that com-prise the two layers are known as epiblasts and hypoblasts.of pluripotent stem cells to cardiomyocytes [45].The morphological marker of the beginning of gastrulation is the appearance of the so-called“primitive streak,” which originates as a thickened ridge of cells in the epiblast (Figure 1.3). Thisridge forms with one end (the caudal end) abutting the edge of the embryonic disc and the other end(the cranial end) extending towards the centre of the embryonic disc. The appearance of the prim-itive streak breaks the symmetry of the embryonic disc and establishes the embryo’s craniocaudalaxis, its dorsal and ventral surfaces, and its right and left sides. The process of cardiomyogenesisfrom the mesoderm has been most thoroughly studied in the mouse, so we now move to the mousemodel for a closer view of the signalling events that direct tissue specification. Mouse embryologyuses the convention of prefixing the number of days since ovulation with “E” to indicate the age ofthe conceptus. Thus an oocyte at ovulation is at E0, and fertilization usually occurs at E0.5.At E4.5, the mouse embryo has entered the late blastocyst stage, but the spatial arrangement ofcell groups still resembles that of the human embryo. There is an outer layer of cells (the trophec-3Ectoderm Mesoderm EndodermFigure 1.3: Gastrulation, the specification of the three germ layers, begins with the formationof the primitive streak (shown as the divot in the green mesodermal cells in cross-sectionof the embryonic disc). Cells along the primitive streak migrate into the bilaminar discto give rise to the mesoderm.toderm) that surrounds an inner cell mass, comprised of the epiblast and the primitive ectoderm, asshown in Figure 1.4 [4]. After implantation, the mouse embryo undergoes a dramatic morpholog-ical folding to create the so-called “egg cylinder”. The cells of the primitive endoderm invaginateto form a double-walled cup, holding the cells of the epiblast at the bottom of the cup and the cellsof the extraembryonic ectoderm near the top of the cup. This break of symmetry establishes theproximodistal axis: the opening of the cup points to the proximal end of the embryo, while thebottom of the cup is distal end (Figure 1.5). At E5.0, cells in the proximal epiblast begin producingNODAL, a signalling peptide that exerts its effects by activating transcriptional modulators (Fig-ure 1.6) [4]. NODAL tells cells of the extraembryonic ectoderm nearest the epiblast to maintainexpression of bone morphogenetic protein 4 (BMP4). This, in turn, signals to the proximal epiblastcells to produce Wnt3.The anteroposterior symmetry is broken on E5.5, when what is now the anterior visceral en-doderm begins expressing the Wnt inhibitor Dickkopf homologue 1 (DKK1) and the NODAL in-hibitors left-right determination factor 1 (LEFTY1) and Cerberus (CER1). NODAL and Wnt sig-nalling continues in the posterior epiblast [10]. About 6 hours later, Wnt signalling tells the cellsof the posterior epiblast to produce the mesoendodermal markers brachyury (T) and eomesoder-min (EOMES). Together, T and EOMES activate mesoderm posterior 1 (MESP1), which begins thecardiac specification protocol in earnest [8].4Figure 1.4: The mouse blastocyst at 4.5 days after ovulation strongly resembles the humanblastocyst near implantation, with the inner cell mass divided into two layers of cells.Reprinted by permission from Macmillan Publishers Ltd: Nat Rev Mol Cell Biol, S. J.Arnold and E. J. Robertson. Making a commitment: cell lineage allocation and axispatterning in the early mouse embryo. Nat Rev Mol Cell Biol, 10(2):91-103, Feb. 2009,copyright 2009.A B C DFigure 1.5: The primitive endoderm of the murine blastocyst invaginates to form a double-walled cup. This forms the so-called egg cylinder stage of the mouse embryo. A: morula.B: Early blastocyst stage with inner cell mass (pink) surrounded by trophoblast (blue).C: Late blastocyst stage with primitive endoderm (orange). D: Egg cylinder.5Figure 1.6: Signalling in the mouse embryo is choreographed in time and localized to specificregions, resulting in the specification of the cardiac mesoderm. Figure reprinted fromCell Stem Cell; 10(1); Paul W. Burridge, Gordon Keller, Joseph D. Gold, and Joseph C.Wu; Production of De Novo Cardiomyocytes: Human Pluripotent Stem Cell Differenti-ation and Direct Reprogramming; Pages 16–28; Copyright 2012; with permission fromElsevier.1.3 Gene regulatory networks for heart development1.3.1 The “kernel” of cardiac regulatory genesStudies of heart development in a wide range of organisms, from the phylogenetically ancient jel-lyfish to more modern mammals, has revealed the existence of an evolutionarily conserved networkof transcription factors that control genes for muscle growth, patterning, and contractility [60]. Thegenes in this network are:• NK2• myocyte enahncer factor-2 (MEF2)• GATA• T-box (TBX)• heart- and neural crest derivatives-expressed (HAND)6Work in Drosophila has revealed that, once established by upstream signals, this regulatory networkis self-sustaining. Many other genes can be considered accessories to this core “kernel” of regulatorygenes; some of these accessory genes have been used successfully to transdifferentiation somaticcells into cardiomyocytes, as described in Section 1.5.As multicellular organisms increase in size, the need for a specialized vascular system to trans-port nutrients and wastes increases as well. The first structures that rose up for the task were mereperistaltic tubes lacking chambers and valves – they served merely to force fluid through the or-ganism, and could not pump blood or fluids unidirectionally [6]. But as more complex organismsevolved, so their hearts became commensurately more sophisticated, with chambers dedicated toreceiving and pumping blood, a conduction system ensuring synchronized contractions, valves andsepta to ensure efficient unidirectional flow, and seamless connections to a closed vascular system[54, 72].1.3.2 The gene-to-structure relationshipWith increase in heart complexity, investigators have found a concordant increase in the numberof paralogs for the genes of the cardiac regulatory network. The current theory states that geneduplication events during evolution increased the number of these regulatory transcription factors,increasing the possible combinations of gene expression and changing protein expression in differ-ent areas to create specialized structures [60]. Figure 1.7 shows the number of paralogs found in thecardiac tissue of representative animals, as well as their phylogenetic relationship.The gene-to-structure relationship is most clearly demonstrated with the role of the HAND genein ventricular development. There exists a one-to-one relationship between HAND genes and ven-tricles – amphibians and fish express only one HAND gene, and their hearts have only one ventricle.Zebrafish mutants without HAND expression do not form the ventricular chamber [87]. Mice andother mammals have two copies of HAND, with HAND1 expressed in the derivatives of the primaryheart field, and HAND2 expressed in the derivatives of the secondary heart field [9]. Mice mutantswithout HAND2 expression do not form a right ventricle [76], while mutant embryonic stem cellslacking HAND1 cannot contribute to the left ventricle [65]. Silencing of both HAND genes by delet-ing HAND2 and NK2 homeobox 5 (NKX2-5), which regulates HAND1 expression, results in a heartlacking ventricles altogether [9, 75].1.3.3 Upstream control of the cardiac gene kernelHow does the kernel of cardiac regulatory genes establish itself in the developing organism? Workin model organisms has revealed that other genes were recruited to act as upstream initiators of thecore kernel during evolution. For example, the MESP gene is required for the expression of NK2and HAND in cardiac progenitor cells of the tunicate Ciona intestinalis, a phylogenetically ancientchordate [18, 70]. In the mouse, MESP1 and mesoderm posterior 2 (MESP2) are required for7Figure 1.7: The number of cardiac regulatory genes expressed in cardiac tissue of representa-tive animals. Hearts with more structures and specialized tissues express a higher numberof gene paralogs. From E. N. Olson. Gene regulatory networks in the evolution and de-velopment of the heart. Science, 313(5795):1922-1927, 2006. Reprinted with permissionfrom AAAS.specification of the cardiac mesoderm and coordinate the migration of multipotent cardiovascularprogenitors [7].1.4 Induced pluripotent stem cellsIn 2006 and 2007, Takahashi and Yamanaka published the first reports of the creation of inducedpluripotent stem cell (iPSC)s in mice and human cells [77, 78]. With the induction of the fourtranscription factors, OCT4, SRY (sex determining region Y)-box 2 (SOX2), Kruppel-like factor 4(gut) (KLF4), and v-myc avian myelocytomatosis viral oncogene homolog (MYC), Takahashi andcolleagues transformed human dermal fibroblasts into embryonic stem cell-like cells capable forforming cells of all three germ layers in vitro and in teratomas. This seminal work catalyzed a flurryof efforts in both the differentiation of cardiomyocytes from iPSCs. Besides cardiac regeneration,researchers have been motivated by the need for a more accurate in vitro model of the heart for themodelling of disease and the testing of candidate drugs.The range of differentiation protocols share a common theme of attempting to recapitulate invivo cardiomyogenesis by recreating the signalling events of fetal development. The efficiency ofthese protocols is highly variable – the process is not robust and differences in the starting pluripo-tent cell population, the composition of the tissue culture medium, the concentration of growth8factors, the timing of growth factor application, and the morphology of the starting cell populationslead to different results [10]. Nonetheless, these methods can be classified into two main camps –the embryoid body method, in which pluripotent stem cells are grown in suspension and allowed toform spherical conglomerates of cells resembling the morula, and the monolayer method, in whichpluripotent stem cells are grown to a high density.The monolayer method is technically simpler and faster than the embryoid body method. It be-gins with 24 hours of treatment with Activin A, followed by 4 days of treatment with BMP4 withoutmedium change. Both Activin A and BMP4 belong to the transforming growth factor β (TGF-β)superfamily of signalling proteins. Since the first report of this method was published by Laflammeet al. [45], others have made incremental improvements. Paige et al. [61] added Wnt3a during thefirst 24 hours, and DKK1 between days 5 and 11. Uosaki et al. [82] used a Matrigel overlay 24 hoursbefore differentiation to enhance epithelial-mesenchymal transitions, used fibroblast growth factor2 (basic) (FGF2) and removed insulin during days 1–5, and used DKK1 from day 5–7. Hudsonet al. [33] saw improvements by adding the Wnt inhibitors IWR-1 and IWP-4 on day 3.A major limitation to the in vivo application of the above techniques is the risk of generatingteratomas via unspecific differentiation of iPSCs. Indeed, such risks have been verified in the mouseby Abad et al. [1]. Failure to engraft by the injected cells presents another challenge for clinicalapplications.1.5 TransdifferentiationIn 1987, Davis et al. [20] reported the discovery of the master regulatory gene of skeletal muscle,myogenic differentiation 1 (MYOD1). With the expression of this single gene, fibroblasts and othercell types can be directly reprogrammed to adopt a skeletal muscle phenotype [15]. Researchershave since searched extensively for a similar master regulator of cardiac muscle differentiation, butthe lack of success lead to a dwindling of efforts.The recent results of iPSC reprogramming rekindled interest in direct reprogramming, or trans-differentiation, of somatic cells into cardiomyocytes. If successful, such a method would eliminatethe risk of teratoma formation seen with the induction of pluripotency in vivo.Using a strategy similar to Takahashi and Yamanaka [77], Ieda et al. [34] managed to reprogrammurine cardiac and tail-tip fibroblasts into cells exhibiting the fetal cardiomyocyte phenotype. Start-ing from a pool of genes known to be highly expressed in embryonic cardiomyocytes and knownto cause severe cardiac developmental defects when mutated, they identified GATA4, MEF2c, andTBX5 (GMT) as the minimum essential transcription factors for cardiac transdifferentiation. Thisreport of successful transdifferentiation spawned a series of papers documenting improvements inreprogramming efficiency and the maturity of the resulting cardiomyocytes by adding various tran-scription factors, small molecules, and microRNAs.In mice and murine cells, Song et al. [73] added HAND2 to the 3-factor cocktail (GMT) intro-9duced by Ieda et al., transdifferentiating non-myocyte cardiac cells in the mouse heart to beatingmyocytes and improving heart function following induced myocardial infarction. Protze et al. [62]took a more rigorous approach in selecting the set of transcription factors with which to inducereprogramming – instead of using the Takahashi group’s strategy of removing factors one at a timeto find the minimum sufficient set, they tested all combinations of 3 factors from a pool of 10candidates. They found that set of T-box 5 (TBX5), myocyte enhancer factor 2C (MEF2C), and my-ocardin (MYOCD) upregulated a broader set of cardiac genes than the GMT cocktail, as determinedby quantitative PCR. Jayawardena et al. [40] also used a combinatorial approach to identify a set ofmicroRNAs (miR-1, miR-133, miR-208, miR-499) sufficient to transdifferentiate a significant por-tion of cardiac fibroblasts into cardiac-like cells. Interestingly, they found that treatment with Januskinase (JAK) inhibitor I increased reprogramming efficiency by ten-fold. Addis et al. [3] found thatthe combination of HAND2, NKX2-5, GATA binding protein 4 (GATA4), MEF2C, and TBX5 hada stronger reprogramming effect than GMT alone. Similarly, Christoforou et al. [17] found thatMYOCD, serum response factor (SRF), MESP1 and SWI/SNF Related, Matrix Associated, ActinDependent Regulator Of Chromatin, Subfamily D, Member 3 (SMARCD3) enhance the inductiveeffects of GMT.In human cell culture, Nam et al. [59] were the first to demonstrate the possibility of transd-ifferentiating fibroblasts to cardiomyocytes. They used a combination of transcription factors andmicroRNAs – GATA4, TBX5, HAND2, MYOCD, miR-1, and miR-133. This was followed closelyby work from Wada et al. [83], who added MESP1 and MYOCD to the GMT cocktail. Fu et al. [27]added estrogen-related receptor gamma (ESRRG), MESP1, MYOCD, and zinc finger protein, FOGFamily Member 2 (ZFPM2) to further enhance reprogramming.Islas et al. [35] took an alternate approach to selecting the genes for cardiac transdifferentiation.Whereas previous groups used a combinatorial approach or one-at-a-time removal approach to findthe optimal minimum subset of core cardiac regulatory genes to achieve reprogramming, Islas andcolleagues tried genes that were known to be involved in upstream cardiac specification in modelorganisms (Section 1.3.3). They found that homo sapiens v-ets erythroblastosis virus E26 oncogenehomolog 2 (avian) (ETS2) and MESP1 were sufficient activate the cardiac gene regulatory networkand convert human dermal fibroblasts to cardiac-like myocytes.Despite the extensive work in optimizing protocols, even the most effective reprogrammingprotocols struggle to attain efficiencies of 40% [74], and as yet unknown epigenetic states may pre-clude reprogramming, making some protocols difficult to reproduce [13]. In addition, all of thereprogramming protocols above are plagued by the same two problems: one, the cardiomyocyte-like cells generated are immature, resembling fetal cardiomyocytes in electrophysiology and cal-cium handling; two, the population of cells is heterogenous, with a variable mixture of atrial-,ventricular-, and nodal-type myocytes. Both problems must be overcome if we are to achieve areliable population of cells with which to model cardiac disease, test candidate drugs, and achieve10cardiac regeneration in vivo. Since some of the cells will be nodal-type pacemakers, the problem ofintroducing arrhythmogenic ectopic pacemakers looms large.Noticeably absent from the literature are works examining the effects of mechanical signallingon cardiomyocyte transdifferentiation and maturation. This is in spite of similar works in differen-tiation of pluripotent stem cells and knowledge of the role of tissue stiffness and cyclic mechanicalloads on stem cell differentiation [23], the spontaneous beating of primary cardiomyocytes in vitro[24], the TGF-β and BMP signalling pathways [84] in ventricular myocytes, and the structuralchanges in atrial myocytes [21].1.6 Defining characteristics of cardiomyocytesThe field of cardiac differentiation and reprogramming has not yet reached a consensus on where thedistinction between differentiating cardiac progenitor (or somatic cell) and de facto cardiomyocyteshould be drawn. The cells generated from differentiation and transdifferentiation experiments areheterogeneous and fall on a continuum of maturity from the starting cell population on one end andfully mature cardiomyocyte on the other. Typically, a large proportion of cells express baseline fea-tures (e.g., driving a cardiac-specific fluorescent reporter, such as α myosin heavy chain (αMHC)-enhanced green fluorescent protein (EGFP)), and ever-decreasing numbers of cells expressing morestringent criteria (e.g., calcium oscillations, action potentials, and spontaneous contraction).Addis and Epstein [2] proposed that three criteria be met before a cell is to be considered suc-cessfully reprogrammed to a cardiomyocyte:1. The cell should have a gene expression pattern (as determined by reverse-transcriptase poly-merase chain reaction (RT-PCR), microarray, or transcriptome sequencing) that more closelymatches those of cardiomyocytes than other cell types.2. The cell should express structural proteins (e.g., cardiac troponin T (cTnT), αMHC, and α-actinin) and organize these proteins into sarcomeres3. The cell should exhibit at least one functional attribute (e.g., action potentials, calcium oscil-lations, and spontaneous or induced beating)1.7 Mechanical signalling in cell differentiation1.7.1 Mechanisms of mechanotransductionHow do mechanical forces outside the cell lead to changes in gene expression in the nucleus? Ourcurrent understanding of intracellular force transmission mechanisms involves a handful of actors,some of which are depicted in Figure 1.8. For example, integrins span the cell membrane, binding11to extracellular matrix (ECM) proteins (e.g., collagen, fibronectin) on the outside and intracellularproteins on the inside. Talin and vinculin act as adapter proteins to bind the intracellular domain ofintegrin to actin filaments of the cytoskeleton. Near the nucleus, the nesprin proteins connect theactin filaments to nuclear membrane proteins such as SUN1 and SUN2. The intranuclear domainsof SUN1 and SUN2 interact with the nuclear envelope protein lamin, which can bind DNA andforms stable nuclear structures [37].Researchers have long suspected that mechanical signalling plays an important role in cell be-haviour [26], and indeed many working in the field of differentiation and transdifferentiation tocardiomyocytes have conjectured that the mechanical properties of the cardiac microenvironmentmay explain why in vivo experiments generally have better results than in vitro [63].Engler et al. [23] further confirmed the importance of the mechanical properties of the extra-cellular environment by showing that mesenchymal stem cells can be directed to differentiate byvarying only the substrate stiffness. When the non-muscle myosin IIs are blocked with blebbistatin,a selective inhibitor, the mesenchymal stem cells no longer respond to substrate stiffness. Macri-Pellizzeri et al. [48] reported similar findings with iPSCs.1.7.2 Previous works involving mechanical strainResearchers have long speculated that the cardiac microenvironment offered additional signallingcues to cells being differentiated or reprogrammed to cardiomyocytes [74]. This conjecture is basedon the results of in vivo reprogramming and differentiation experiments, which generally show muchhigher rates of conversion to cardiomyocytes than similar experiments in vitro [63, 73]. Possiblecandidates for these so-called ”non-soluble factors” that affect cell differentiation include tissuestiffness, extracellular matrix proteins, and mechanical strain.Engler and Discher were pioneers in the exploration of the effects of mechanical signalling onstem cell differentiation. They found that mesenchymal stem cells can be directed to differentiateto neuronal, myoblastic, or osteoblastic lineages by varying only the stiffness of the substrate onwhich they grew [23]. It is thought that cells can sense the stiffness of the extracellular matrix bypulling on it and sensing the force required to deform the matrix with force transduction proteins.Other workers have examined the effects of cyclic mechanical strain on stem cells and stem cell-derived cardiomyocytes. Horiuchi et al. [31] exposed mouse embryonic stem cell (mESC) to 10%at 0.17Hz (10 cycles per minute) and found that this stretching maintains expression of NANOG, amarker of pluripotency. Similarly, Saha et al. [68] and Saha et al. [67] stretched human embryonicstem cell (hESC) biaxially at 0.17Hz, 10% strain, and found that this inhibited the spontaneousdifferentiation of hESC and promoted their self-renewal.Interestingly, varying the parameters of cyclic strain (frequency, magnitude, and duration) canalso have pro-differentiation effects. Teramura et al. [79] stretched iPSCs at 0.2Hz, 15% strain, for12 hours. This cyclic strain regimen decreased the expression of pluripotency markers NANOG,12Figure 1.8: A schematic depicting some of the known actors in cellular mechanotransduction.Figure reprinted by permission from Macmillan Publishers Ltd: Nat Rev Mol Cell Biol,D. E. Jaalouk and J. Lammerding. Mechanotransduction gone awry. Nat Rev Mol CellBiol, 10(1):63-73, Jan. 2009, copyright 2009.13OCT4, and SOX2.Tulloch et al. [81] seeded hESC- and iPSC- derived cardiomyocytes into a three-dimensionalcollagen scaffold and subjected the scaffold to cyclic, unixaxial mechanical stress. This stretchingtreatment resulted in increased cardiomyocyte fibre alignment, myofibrillogenesis, and sarcomericbanding. The stretched cells showed greater hypertrophy and proliferation, and quantitative RT-PCRshowed increased mRNA transcript levels of cardiac-specific genes. The stretching regimen began16 days after the end of BMP4 treatment, after the start of spontaneous contractions. The scaffoldwas stretched at 1Hz, 5% strain, for 4 days.1.8 HypothesisThe aim of my thesis is to explore the phenomenological effects of cyclic strain on the efficiency ofdifferentiation and transdifferentiation to cardiomyocytes. We hypothesize that cyclic mechanicalstrain, in conjunction with transgene expression and treatment with soluble factors, will increasedifferentiation and transdifferentiation efficiency, as measured by the number of cells activatinga cardiac-specific fluorescent reporter (Section 2.2.2). We also hypothesize that cyclic mechanicalstrain can increase the maturity of generated cardiomyocytes, as measured by structural organizationof sarcomeres and calcium handling abilities.Chapter 2 specifies the details of my experimental procedures. In Chapter 3, I describe thedesign, fabrication, and characterization of a device used for the controlled application of strain tocultured cells. In Chapter 4, I report on the modification and testing of a cardiac-specific fluorescentpromoter, and I present the results of differentiating cardiomyocytes from iPSCs. In Chapter 5, Idiscuss the implications of my results and speculate on possible directions for future works.14Chapter 2Materials and methods2.1 Cell culture2.1.1 Cell adhesion coatingsMatrigel is a solubilized basement membrane protein mix extracted from the Engelbreth-Holm-Swarm mouse sarcoma. This tumour is rich in ECM proteins, including laminin, collagen IV,heparan sulfate proteoglycans, and entactin/nidogen. Growth Factor Reduced Matrigel coatingswere used in the culture of iPSCs prior to differentiation. Growth Factor Reduced Matrigel haslower levels of epidermal growth factor (EGF), insulin-like growth factor 1 (IGF-1), and platelet-derived growth factor (PDGF). Matrigel solution was kept on ice to remain liquid and aliquots werefrozen upon receipt. Coating of tissue culture wells was done by dissolving an aliquot of Matrigelin DMEM/F-12 medium (GE Healthcare Life Sciences Catalog #SH30023.FS), and immersing thewell surface in the Matrigel solution at room temperature for 1 hour. Each 6-well plate was coatedwith 0.5mg of Matrigel.Gelatin is a hydrolyzed mixture of collagens derived from animal connective tissue. It is oftenused to coat tissue culture surfaces to facilitate cell adhesion. Type B (lime-cured) gelatin pow-der (Fischer Scientific Catalog #G7-500) derived from bovine skin was dissolved in deionized andautoclaved water to make 0.1% (weight of solute in grams as a percentage of volume of solventin millilitres) gelatin solution. For each well of a 6-well plate, 2mL of 0.1% gelatin solution wasadded. The tissue culture vessels were incubated with the gelatin solution at 37◦C for 4 hours beforeuse.Fibronectin is another ECM protein commonly used to coat tissue culture surfaces to facilitatecell adhesion. Fibronectin also exists in a soluble form in blood plasma, which is the source ofthe fibronectin used in these experiments. Fibronectin was purified from human plasma by affinitychromatography using gelatin Sepharose 4B (GE Healthcare Life Sciences Catalog #17-0956-01).15Poly-L-lysine solution (0.01%, provider, catalog number) was used to pre-treat certain tissueculture wells prior to Matrigel coating in order to facilitate protein adhesion. The positively chargedside chain of the amino acid polymer increases the binding of negatively charged proteins and cells.Coating was done inside a biosafety cabinet using sterile technique at room temperature with 1mLof poly-L-lysine solution per 6-well plate well. After 5 minutes, the coating solution was aspiratedand the culture surface allowed to dry before introducing cells or additional coating.2.1.2 BJ fibroblastBJ fibroblasts were a gift from Dr. Christopher Maxwell of the Child and Family Research Insti-tute and the University of British Columbia. They were maintained in Dulbecco’s Modified Eagle’sMedium (DMEM)/Ham’s F-12 nutrient mix, 1:1 ratio, with 10% fetal bovine serum (FBS) and1% penicillin/streptomycin. Medium was changed every 4-5 days. Passaging was done when thecells reached 90% confluence (i.e., coverage of the available growth substrate by the cells reached90%). For dissociation, the cells were rinsed with phosphate-buffered saline (PBS) before incuba-tion for 5-8 minutes with trypsin-ethylenediaminetetraacetic acid (EDTA) solution (0.05% trypsin,GE Healthcare Catalog #SH30236.01).2.1.3 Celprogen “primary cardiomyocytes”Putative human cardiomyocytes derived from biopsy of adult cardiac tissue were purchased (Celpro-gen Catalog #36044-15). These cells were grown on 6-well plates with proprietary ECM adhesioncoatings (Celprogen Catalog #E36044-15) and in proprietary medium with serum (Celprogen Cat-alog #M36044-15S). The cells were maintained at 37◦C in an atmosphere of 5% carbon dioxide.Medium renewal was performed daily, and passaging was done at 80-90% confluence. Cells wereharvested for experiments at 90% confluence.2.1.4 Induced pluripotent stem cellsiPSC cells, iPS(IMR90)-4, from the WiCell Stem Cell Bank, were a gift from Sanam Shafaat Talaband Dr. Glen Tibbits of the Child and Family Research Institute and Simon Fraser University. TheiPS cells were transformed from the IMR-90 fibroblast cell line, which was itself derived from thelungs of a 16-week female human fetus. The fibroblasts were passaged 18 times prior to repro-gramming with OCT4, SOX2, NANOG, and Lin-28 homolog A (LIN28) [88]. Once transformed, thecells were passaged 41 times, the last 16 of which were cultured in mTeSR1 medium on a Matrigelcoating. Once received, the iPSCs were cultured on Corning R© Matrigel R© Growth Factor ReducedBasement Membrane Matrix (Corning Product #356230).The culture medium used was mTeSRTM1 (Stemcell Technologies Catalog #05850), contain-ing recombinant human TGF-β and recombinant human basic fibroblast growth factor (bFGF).16Complete mTeSR1 medium is made by adding the mTeSR1 5X Supplement to the mTeSR1 BasalMedium. The mTeSR1 5X supplement was thawed at room temperature or at 4◦C. Aliquots of thecomplete mTeSR1 medium were frozen at -20◦C until needed, at which point they were also thawedat 4◦C. Once seeded, the medium was renewed daily until passaging or initiation of differentiationprotocol. The optimal time for passaging depends on seeding density and cell aggregate size at thetime of seeding. Cell cultures were visually inspected and passaged when confluent islands of cellsreached at least 2 mm in size. The culture medium was aspirated and the cells were washed with1mL Versene R© (EDTA) 0.02%, (Lonza Catalog #17-711E). The wash solution was aspirated anda fresh 1mL of Versene was used to treat the cells for 8 minutes. The Versene was then carefullyaspirated without disturbing the cell layer. Next, 3mL of mTeSR1 was used to wash the cells offthe culture surface. For maintenance passage, a 5mL pipet and Pipet-Aid R© were used for gentledissociation forces. The larger cell aggregates created this way showed faster recovery after replat-ing. For seeding into wells prior to differentiation protocols, a manual pipettor was used to pipetup and down rigorously, resulting in almost fully dissociated cells that, once replated, grew into amore homogeneous monolayer. The split ratio varied from 1 to 12 for maintenance and expansionpassaging, to 1 to 3 for differentiation experiments. Immediately after replating, the small-moleculerho-associated, coiled-coil containing protein kinase (ROCK) inhibitor Y-27632 was added to theculture medium to a final concentration of 10µM. The use of ROCK inhibitor has been shown toincrease viability of stem cells in vitro by reducing dissociation-induced apoptosis (also known asanoikis) [85].Freezing the iPSCs was done by harvesting with Versene, as described above. After the 8-minuteVersene treatment, 3mL of mFreSR R©1 (Stemcell Technologies Catalog #05855) was used to washthe cells off the culture substrate. Multiple washes were done to ensure all cells were washed off.The cells from each 6-well plate were pooled in 10mL of mFreSR1, aliquoted into 10 cryovials,and put into an isopropanol freezing container to be frozen at -80◦C. The vials were transferred tovapour phase nitrogen the next day.2.1.5 Differentiation of iPSCs to cardiomyocytesWe used the monolayer method to differentiate iPSCs into cardiomyocytes [45]. iPSCs were thor-oughly dissociated and seeded onto Matrigel-coated culture plates at a density of 100,000 cells/cm2.After 3-4 days of daily medium renewal, the cells should reach 90% confluency. (If the cells do notgrow as expected, extending this culture period generally does not result in greater confluence. In-stead, non-adherent cells tend to proliferate and crowd out adherent cells, resulting in even lowerconfluence.) At this confluence, the mTeSR1 medium was replaced with RPMI 1640 basal mediumwith B-27 supplement (no insulin) (Life Technologies Catalog #A1895601). Recombinant humanActivin A (R&D Systems Catalog #338-AC-010) was added to this medium to a final concentrationof 100ng/mL. After 24 hours of Activin A treatment, the medium was refreshed and recombinant hu-17man BMP4 (R&D Systems Catalog #314-BP-010) was added to a final concentration of 10ng/mL.The cells were treated with BMP4 for 4 days without medium change so that secreted factors mayaccumulate and exert an autocrine or paracrine effect. After 4 days, the medium was changed toRPMI 1640 basal medium with B-27 supplement (complete) (Life Technologies Catalog #17504-044). This medium was renewed every 3 days. With traditional polystyrene cell culture substrates,spontaneous contractions can be observed 10 days after onset of Activin A treatment. With con-tinued maintenance, the spontaneously beating collections of cells remodel and recruit more cells,creating larger areas of contraction.2.2 Plasmids and transfection2.2.1 TransfectionTransfection, the introduction of foreign DNA to a cell, was done using electroporation. Electropo-ration uses electrostatic fields to increase the membrane permeability of cells to DNA. To performelectroporation, we used the NucleofectorTMdevice from Lonza, with programs optimized for thecell types under experimentation. Program A-024 was used for fibroblasts, and G-009 was used forcardiomyocytes or cardiac-like cells. In lieu of Amaza proprietary transfection reagents, nucleofec-tion was carried out in Opti-MEM R© media (Life Technologies Catalog #31985070). In most cases,nucleofection in Opti-MEM achieves comparable transfection efficiencies at much reduced costs[25].2.2.2 Reporter for cardiac-specific myosin heavy chainMyosin is a protein that generates force from the chemical potential energy stored in adenosinetriphosphate (ATP). The myosin present in cardiac muscles is a large protein made up of two heavysubunits (the myosin heavy chains), two light subunits (the myosin light chains), and two regulatorysubunits. The human genome encodes 17 different isoforms of myosin heavy chain (MHC)s, butonly αMHC and β myosin heavy chain (βMHC) are present in cardiac muscle [55].Researchers have taken advantage of the tissue specificity of αMHCs, as well as previous workcloning and characterizing its promoter, to create a reporter construct for cardiac-like cells. Specif-ically, the ∼5000 base pairs upstream of the transcription start site of of αMHC was used as apromoter driving fluorescent reporters (i.e., EGFP and mCherry) [28]. When the cells enter acardiac-like transcriptional state, it begins to transcribe αMHC mRNA, but this also activates thetranscription of the fluorescent proteins, which enables the quantification of differentiation effi-ciency with flow cytometry. Previous workers have used lentiviruses to create pluripotent stem cellsstably expressing bioengineered reporter constructs [44]. Despite the utility of these techniques,lentivirus transduction always carries the risk of random genomic integration and tumorigenesis.18We thus created a plasmid containing the same reporter construct that can be transiently expressedafter introduction to cells via nucleoporation.We began with the pEGFP-N1 plasmid backbone (Addgene plasmid #6085-1) and the αMHCreporter construct created by Kita-Matsuo et al. [44] (Addgene plasmid #21229). The αMHC pro-moter was cloned into the pEGFP-N1 multiple cloning site (MCS) with XhoI and AgeI. Then thehuman cytomegalovirus (CMV) promoter was excised with AseI and NheI, and a blank annealingoligo was used to close the plasmid. The resulting plasmid’s map is shown in Figure Vehicle for introducing ETS2 and MESP1We created a bicistronic plasmid vector capable of mammalian expression carrying the genes ETS2and MESP1. A map of the plasmid is shown in Figure 2.2.The CMV immediate early promoter and enhancer is used to express transgenes in a variety ofcell lines. Transcription factors bind to the enhancer region to facilitate transcription, while the pro-moter region contains DNA sequences for both transcription factors and RNA polymerase to bindand initiate transcription. The IRES allows 2 proteins to be translated from the same bicistronicmRNA [38, 39]. IRES originates from the 5’ untranslated region (5’-UTR) of the encephalomy-ocarditis virus, and folds into secondary hairpin structures that allows the eukaryotic ribosome tobind and initiate translation. The precise cellular mechanism for this ribosomal binding is still un-clear. Polyadenylation is the addition of several adenine bases to the 3’ end of an mRNA molecule,which allows the mRNA to remain stable while being transported from the nucleus and translated.The SV40 polyadenylation signal is an RNA element that promotes efficient polyadenylation [86].The f1 origin of replication is derived from the f1 phage. This enables a single-stranded copyof the plasmid to be made and packaged into a phage particle. The pIRES2 plasmid contains aneomycin/kanamycin-resistance cassette for bacterial and eukaryotic cell selection. The proteinencoded by the cassette hails from the Tn5 transposon of E. coli, and confers resistance to bothneomycin and kanamycin. Cells that have taken up the plasmid can thus be grown in media or onagar containing these antibiotics, allowing us to select for these cells. The expression of this geneis driven by an SV40 early promoter/enhancer sequence. The tail end of this cassette contains thepolyadenylation sequence from the herpes simplex virus thymidine kinase (HSVTK) gene. Eukary-otic cells stably transfected with a plasmid containing this cassette can thus be selected for using theG418 antibiotic. The pUC origin of replication allows the plasmid to be replicated in E. coli. Thisorigin of replication enables each cell to maintain a high copy number of plasmids (over 500).ETS2 and MESP1 were purchased from DNASU (clone ID HsCD00002188 and HsCD00080018).These genes were amplified using polymerase chain reaction (PCR) with custom-designed primersthat included flanking restriction enzyme sites (see Table 2.1). The PCR products were agarosegel-purified and ETS2 was then inserted upstream of the IRES. Downstream of IRES, the EGFPcassette was removed and replaced with MESP1.19Figure 2.1: Map of the αMHC reporter plasmid. The CMV promoter of the pEGFP-N1 plas-mid was replaced with the αMHC promoter.Table 2.1: PCR primers for the amplification and cloning of ETS2 and MESP1.Primer Name SequenceMESP1.FOR 5’-CTCGAGGAATTCAATGGCCCAGCCCCTGTGC-3’MESP1.REV 5’-AATCCCGGGCTACAACTTGGGCTCCTCAGGCAGC-3’ETS2.FOR 5’-GAATTCAGATCTCACCATGAATGATTTCGGAATC-3’EST2.REV 5’-GTCGACGGATCCATGGTCTAGAAAGCTTCCC-3’20Figure 2.2: Map of the pIRES2 plasmid vector used to deliver ETS2 and MESP1 to the cellsof interest. This plasmid features a CMV immediate early promoter and enhancer, anIRES sequence, an MCS upstream of the IRES sequence, a SV40 polyadenylation signaldownstream of the EGFP gene, an f1 single-strand DNA origin, a Kanr and Neor cassette,and a pUC origin of replication. Restriction sites are shown in bold with leader linesindicating their relative position on the plasmid.2.3 Lentivirus production and transductionLentivirus remains a popular choice for the introduction of transgenes to non-dividing cells for sta-ble expression. The technology was based on the human immunodeficiency virus and was developedin the late ’90s by the Trono lab.2.3.1 Lentivirus productionLentiviruses carrying the αMHC reporter construct were constructed with the help of Oksana Ne-mirovsky. We used a second-generation lentiviral packaging system, consisting of the psPAX221packaging plasmid, the pMD2.G envelop plasmid, and the transfer plasmid containing our gene ofinterest. The separation of the components of lentivirus (i.e., envelope proteins, packaging proteins)is a safety feature used to lower the risk of generating replication-competent viruses capable of in-fecting humans. The system also contains other variations (e.g., truncated 3’ long terminal repeat,deletion of auxiliary virulence genes) that increase safety [90, 91].The three plasmids were co-transfected into HEK-293FT cells, a cell line derived from humanembryonic kidney transformed with the SV40 large T antigen, in order to create lentiviruses carryingour gene of interest. Transfection was done using Lipofectamine 2000 R©. The large T antigenfunctions to increase DNA replication by suppressing tumour suppressor proteins and pushing thecell from G1 to S phase. This also increases virus genome replication, a desirable effect for lentiviralparticle production.After transfection, the culture medium containing viral particles was collected and filtered toremove cells. The virus-containing medium was then centrifuged to obtain a pellet of viral particles,the medium aspirated, and the viral particles resuspended in PBS.2.3.2 Lentivirus transductionBJ fibroblasts were cultured using methods described in Section 2.1.2. BJ fibroblasts were seededinto a 24-well tissue culture plate at a density of 4× 104 cells per well and allowed to attach for16 hours. The next day, the medium was changed to include the lentiviral particle suspension. Thecells were washed twice with PBS before the addition of the transduction medium (regular tissueculture medium and lentiviral suspension solution). A range of viral suspension volumes were used(2µL, 5µL, 10µL, and 15µL), with and without the transduction adjunct hexadimethrine bromide,to gauge transduction efficiency. Hexadimethrine bromide (Polybrene) is a cationic polymer thatincreases transduction efficiency by reducing the charge repulsion between the virus and the cellmembrane. Some cell types are sensitive to polybrene, so transduction was done with and withoutthe adjunct to gauge toxicity. Polybrene was added to a final concentration of 4µg/mL. After 24hours, the transduction medium was aspirated and the cells washed three times with PBS. From thispoint, cell culture and passaging proceeded as described in Section ImmunocytochemistryiPSC-derived cardiomyocytes were fixed in 3.7% formaldehyde fixation buffer and permeabilizedwith a buffer solution containing 0.1% TX-100. Anti-α-actinin antibody (Sigma-Aldrich #A7811)was applied at a 1:800 dilution. Anti-cTnT antibody (Abcam #ab10214) was applied at a 1:600dilution. Bound primary antibodies were visualized with goat anti-mouse IgG, DyLight 488 con-jugate (Life Technologies Catalog #35503). Specimens were mounted using ProLong Gold with4’,6-diamidino-2-phenylindole (DAPI) (Life Technologies Catalog #P-36931).222.5 Western blotsWestern blotting was conducted on Celprogen cardiomyocytes with anti-cTnT antibody (Abcam#ab10214) at a 1:2000 dilution. Mouse whole heart lysate and Jurkat cell lysate were used aspositive and negative controls, respectively. PKA C-α, a housekeeping gene, was used as a loadingcontrol.2.6 Flow cytometryFlow cytometry employs lasers to count and sort cells. Flow cytometers can separate individualcells on the basis of presence of fluorescent chromophores. Cell counting was performed at theChild & Family Research Institute (CFRI) Flow Core facility using the FACSCanto instrument (BDBiosciences). Post-acquisition analysis was done using FlowJo (Tree Star).23Chapter 3Cyclic strain device for tissue culture3.1 Cyclic strain in the heartModelling the ventricles of the heart as a hollow ovoid, we can define an orthogonal cardiac co-ordinate system as depicted in Figure 3.1. The radial axis is normal to the outer surface of theepicardium; the longitudinal axis is perpendicular to the radial axis, pointing from the apex of theventricle to the base; the circumferential axis is perpendicular to both the radial and longitudinalaxes and points counterclockwise around the ring of the epicardium, when looking from apex tobase. Moore et al. [52] characterized the strain evolution of the normal human left ventricle duringsystole using magnetic resonance imaging. They measured peak systolic strains of ∼45% in theradial axis, ∼–20% in the circumferential axis, and ∼–16% in the longitudinal axis.The goal of strain devices for tissue culture is to controllably mimic some aspect of this strainin vitro.3.2 Previous devices and their limitationsMany devices for probing cellular responses to cyclic strain have been previously described. Theyhave in common the use of an elastomeric membrane as a growth surface for cells and the means bywhich cells are strained. The membrane may be functionalized, for example with ECM proteins, tofacilitate cell adhesion and proliferation. The devices differ in the way the membrane is stretched.Flexcell R© International Corporation has long marketed a pneumatically-driven tissue strainsetup, complete with proprietary control software (Figure 3.2). A vacuum pump generates thepulling force that strains the membrane, parts of which are stretched against loading posts to gen-erate defined strains. Beyond the high cost associated with such systems, they are limited by theslow response times of the pneumatic system (max stimulation frequency of 5Hz) and their largesize (difficult to integrate with tissue culture incubators).Our lab has addressed some of these limitations with the magnetically actuated cellular strain24Figure 3.1: In order to conceptualize the strains generated and felt by cardiomyocytes, it helpsto establish a cardiac coordinate system, as used in echocardiography. Such a system hasa radial axis, a circumferential axis, and a longitudinal axis.assessment tool (MACSAT), which uses an electromagnet to pull on permanent magnets clampedon both sides of the membrane (Figure 3.3) [42]. However, scaling up the device to strain multiplewells simultaneously proved difficult, as each electromagnet required a signal generator and currentamplifier.3.3 Design of the HexCyclerTo address the above issues I designed the HexCycler, featuring 6 independently controlled directcurrent (DC) motors, each driving a Scotch yoke actuator, shown in Figure 3.4, that converts therotation of the motor to linear motion of a piston head that pushes the membrane from below. Thefinal design is the end result of many iterative prototypes, including rough functional prototypesmachined from polycarbonate and iterations with 3D-printed components. The supporting frameof the HexCycler is made from waterjet-cut aluminum sheet metal (6061 aluminum alloy) so as to becorrosion-resistant and autoclavable (Figure 3.5). A BioFlex 6-well plate from Flexcell International25Figure 3.2: The Flexcell R© FX-5000TM Tension System uses vacuum pressure to strain an elas-tomeric membrane on which cells are cultured. Image courtesy of Flexcell InternationalCorporation.26Figure 3.3: The magnetically actuated cellular strain assessment tool (MACSAT) uses an elec-tromagnet to exert force on permanent magnets attached to elastomeric membranes of theBioFlex tissue culture plate.27Figure 3.4: The motor assembly of the HexCycler features a Scotch yoke mechanism (alsoknown as slotted link mechanism) that turns the rotational motion of the DC motor intolinear motion.Corporation is clamped to the top stage of the device, and the motor assembly below pushes a pistonup and down to distend the membrane.The magnitude of membrane strain is controlled by raising and lowering the upper stage of theHexCycler. Figure 3.12 shows the range of estimated strains achievable with the HexCycler andBioFlex membrane for various max membrane displacements. Since the Scotch yoke mechanismcreates sinusoidal linear motion, the impingement of the piston on the membrane will be a ”half-wave rectified” form of this sinusoid, as shown in Figure 3.6. The maximum amplitude of oscillationcan be varied by changing the disc of the Scotch yoke mechanism.3.3.1 DC motor control with the ArduinoThe open-source electronics platform of the Arduino massively simplified the control of the DCmotors of the HexCycler. Coupled with the Adafruit motor shield, this off-the-shelf solution is anaffordable and easy-to-use alternative to specialized proprietary systems. The heart of the Adafruitmotor shield is the TB6612 metal-oxide-semiconductor field-effect transistor (MOSFET) driver,which uses the Arduino 8-bit digital output to control the large currents required to power the mo-tors. This gives the HexCycler 256 discrete speed settings for a given motor.28Figure 3.5: The HexCycler couples 6 independently-driven DC motors to achieve high paral-lelism. Its frame is made from waterjet-cut 6061 aluminum alloy, making it sterilizableand corrosion-resistant. The 6-well BioFlex plate is shown in magenta.Figure 3.7 shows the frequency response of the HexCycler under simulated experimental con-ditions. The line of best fit, which determines the digital input-to-frequency response relationship,isf = 0.0259d−0.4522[Hz] (3.1)where f is the frequency of oscillation and d is the digital input, an integer between 0 and 255. TheR2 value for this linear regression is 0.9995.3.4 Characterizing strain with finite-element analysisThe radial symmetry of the tissue culture well and piston creates 2 main strain axes – radial andcircumferential (Figure 3.8). With an upward deformation from the impinging piston, the membraneexperiences positive strain (i.e., elongates) in the radial axis, somewhat smaller positive strain in thecircumferential axis, and negative strain in the z axis.To better quantify this strain, I modelled the elastomeric membrane of the BioFlex 6-well plate290 1 2 3 4 5−10−50510Time [s]Displacement [mm]  Piston MotionMembrane MotionFigure 3.6: Depending on the height of the upper stage of the HexCycler, the sinusoidallyoscillating piston may only come into contact with the membrane for part of its cycle.This figure models a hypothetical setup with 5mm of piston travel that impinges on themembrane. The resulting motion of the membrane in direct contact of the piston is shownin blue.using COMSOL Multiphysics, a finite element analysis (FEA) software package. The geometricdimensions of the well are measured from the BioFlex plate. I exploit the radial symmetry of theculture well to simplify the model. Figure 3.9 shows the cross-section of the membrane along aradial strip, as well as the meshing of the model (i.e., the division of the model volume into smallelements on which to apply the constitutive equations). The centre of the culture well is r = 0, andthe outside edge is at r = 17.5, giving the membrane a diameter of 35mm. The thickness of themembrane is 0.5mm. The green section of Figure 3.9 denotes the section of the membrane that ispushed by the HexCycler’s piston.The model is given a fixed constraint on the outer edge, and a prescribed displacement is appliedto the bottom of the membrane with r ≤ 3 to simulate the piston pushing up on the membrane(Figure 3.9). This method of modelling avoids the complexity of modelling the interface betweenthe piston and membrane, but has the drawback of not accurately capturing strain over the areas ofthe membrane that overlay the piston, since it fails to account for membrane stretching and slidingin this area.300 50 100 150 200 250−10−50510Digital InputFrequency of Oscillation [Hz]Figure 3.7: Empirical determination of the HexCycler’s frequency response under operatingconditions. The frequency was measured using the time required for 30 oscillations. Theline of best fit, with equation f = 0.0259d−0.4522, has R2 = 0.9995.Radial AxisCircumferential AxisZ AxisFigure 3.8: The radial symmetry of the BioFlex tissue culture membrane gives rise to threeorthogonal axes: radial, circumferential, and z (axial).31Fixed ConstraintPrescribed DisplacementFigure 3.9: A radial cross-section of the BioFlex tissue culture membrane. The left-most edgeof the rectangle is the axis of radial symmetry, around which the rectangle is revolved togenerate a thin circular membrane. The green section indicates the area that directly con-tacts the HexCycler’s piston. The orange boundary is given a prescribed displacement,representing the pushing action of the HexCycler’s piston. The red boundary is a fixedconstraint, representing the clamping of the BioFlex membrane to the edge of the tissueculture well. The meshing of the membrane is also shown.3.4.1 The Ogden model for hyperelastic materialsSince the potential deformations of the membrane are large relative to the dimensions of the mem-brane, a hyperelastic material model is needed to capture the nonlinear stress-strain behaviour.Hyperelastic materials are elastic materials that possess a strain-energy function. Kim et al. [43]demonstrated that the second-order Ogden model best matched empirically measured stress-strainrelationships of various polydimethylsiloxane (PDMS) polymer mixtures.The Ogden model for incompressible materials begins with a strain energy density function W :W (λ1,λ2) =N∑p=1µpαp(λαp1 +λαp2 +λ−αp1 λ−αp2 −3)[MPa] (3.2)Here, N is the order of the model, λi are the principal stretches, and αp and µp are experimentallyderived parameters. The Ogden parameters determined by Kim et al. [43] are listed in Table 3.132Table 3.1: Empirically derived values of Ogden hyperelastic model parameters [43].Ogden Parameter Valueα1 63.4885 MPaα2 0.041103 MPaµ1 6.371×10−10µ2 3.81166The BioFlex plate uses a “polyorganosiloxane” elastomer as its flexible membrane [5]. Thematerial properties of this elastomer are taken from empirically derived values of a PDMS mixturethat best approximates its Young’s modulus. Most sources estimate the Poisson’s ratio of PDMS tobe ∼0.5, which I approximate in my simulations with 0.49999 in order to avoid the singularity inthe calculation of the bulk modulus [57].3.4.2 Model resultsFigure 3.10 shows the magnitude of the three principal strains when the membrane is stretched to5mm above resting position. The large peak at r = 3mm is due to the sharp discontinuity betweenthe loaded section of the model (i.e., the section given a prescribed displacement) and the rest ofthe membrane. Strain values near this area do not reflect reality, since the model does not accountfor membrane sliding over the piston. Figure 3.11 shows the directions of these principal strainsthroughout the radial cross-section of the membrane. As expected, we see positive stretch alongthe radial axis and compression in the z-axis. The second principal strain, corresponding to thecircumferential axis for most of the membrane, is perpendicular to the plane of the diagram andnot shown. However, in the region with r < 3mm, the model erroneously flips the second and firstprincipal strains.Figure 3.12 shows the range of achievable strains with membrane displacements of 4mm to10mm. The simulation suggests that, for moderate membrane displacements, there will be a plateauregion (4mm < r < 16mm) in which the strain is relatively homogeneous.3.5 Validating the effects of the HexCycler’s strain on culturedfibroblastsIn order to test that the HexCycler is capable of exerting biologically relevant strains to culturedcells, I recreated one of the known effects of cyclic strain on cultured fibroblasts – these spindle-shaped cells will align themselves such that their long axes are perpendicular to the axis of tensilestrain.Figure 3.13 shows differential interference contrast (DIC) images of BJ fibroblasts before andafter the application of cyclic strain. The cells begin oriented randomly, but after 24 hours of33Figure 3.10: Principal strains as a function of distance away from the centre of the culturewell.5% cyclic strain at 1Hz, they become aligned to a common orientation. Figure 3.14 shows thedistribution of fibroblast orientations before and after stretch, along with unstretched controls. Therange of orientations clearly narrows in the cells receiving the strain treatment. The orientationangle is defined to be the angle of the long axis of the cell, relative to the horizontal of the image.The average orientation of strained cells is subtracted from each orientation value to normalize theorientation across several images to zero degrees.34Figure 3.11: Vectors denoting the directions of principal strains. The first principal strain is anelongation along the radial axis, while the third principal strain is a contraction alongthe z-axis. The biologically relevant strain (i.e., the strain the cultured cells feel) is thefirst principal strain.350 5 10 1500. [mm]First Principal Strain  4mm5mm6mm7mm8mm9mm10mmFigure 3.12: First principal strain across the BioFlex tissue culture membrane for piston dis-placements of 4mm to 10mm.5% Cyclic Strain1Hz24 hoursFigure 3.13: DIC microscopy images of fibroblasts before (left) and after (right) 24 hrs of 5%cyclic strain at 1Hz. The cells reorient themselves such that their long axis is perpen-dicular to the axis of radial strain (white arrow).36-100 -50 0 50 100Unstrained 24hrUnstrained 0hrStrain 24hrStrain 0hrCell Orientation Angle [deg]Figure 3.14: Scatter box plot of fibroblast orientations, measured as the inclination from thehorizontal of the longest cell axis. Cells with no obvious longest axis were excludedfrom the analysis. Mean and standard deviations are indicated with red lines. Un-strained negative controls show no significant change in cell orientations, but strainedcells align to a common orientation (normalized to 0 degrees) after 24 hours of exposureto 5% cyclic strain.37Chapter 4Results4.1 Validating the αMHC reporterTo verify that the αMHC reporter is capable of driving expression of green fluorescent protein(GFP), we tested its response to thyroid hormone. Previous work characterizing the αMHC genehad shown the existence of thyroid response elements in its 5’ upstream region [80]. It is thought thatthese thyroid response elements mediate the hormonal regulation of MHC isoform switching duringdevelopment [49]. The thyroid hormones triiodothyronine (T3) and its prohormone, tetraiodothyro-nine (T4), have been used in vivo and in vitro to stimulate the expression of αMHC. More specifi-cally, Gustafson et al. [30] used T3 to drive the expression of a transfected fusion gene (the αMHCpromoter fused to chloramphenicol acetyltransferase (CAT)) in cultures of rat cardiomyocytes. Thesame group has shown that treatment of rat cardiomyocytes in vitro with T3 up-regulates αMHCand down-regulates βMHC [29]. Lompre´ et al. [46] showed that injection of T4 into rats has similareffects, increasing the transcription of αMHC mRNA and decreasing the transcription of βMHC.4.1.1 Thyroid hormone fails to activate αMHC reporter transduced by lentivirusWe tested the ability of T3 and T4 to induce expression of αMHC. BJ fibroblasts were stablytransduced with the αMHC reporter construct described in Section 2.2.2 (see Section 2.3 for detailedmethods of lentivirus production and transduction).Figure 4.1 and Figure 4.2 show the results of treatment of T3 and T4, respectively. The con-centrations of T3 and T4 were based on reported values in the literature. Cells transduced withEGFP act as positive controls for both flow cytometry and the lentiviral transduction process. Theemission from intracellular EGFP chromophores is captured using both the fluoresceine isothio-cyanate (FITC) and propidium iodide (PI) channels, which have bandpass filters around their emis-sion spectrum peaks of 519nm and 617nm, respectively. This allows for 2-dimensional plots ofemission intensity that better separate positive and negative populations. The emission profile of38Figure 4.1: Flow cytometric analysis of BJ fibroblasts stably transduced with the αMHC re-porter. Treatment with various concentrations of T3, the active form of thyroid hormone,does not result in increased expression of EGFP. FITC-A: fluorescein isothiocyanatechannel emission intensity, arbitrary units. PI-A: propidium iodide channel emissionintensity, arbitrary units.cells transduced with EGFP (positive controls) was compared with that of untreated cells trans-duced with αMHC only (negative controls). The difference in these emission profiles determinesthe gated area that contains cells considered true positives.Despite achieving a viral transduction rate of ∼72% and ∼43% of the BJ fibroblasts (EGFPcontrol group), negligible αMHC reporter driven fluorescence was observed from either the T3 orT4 treated groups. The transduced αMHC reporter is not inducible by thyroid hormone in the BJfibroblast.4.1.2 Thyroid hormone fails to activate αMHC reporter transfected usingnucleoporationWe also examined the response of the the αMHC reporter plasmid to thyroid hormone. Figure 4.3shows the result of treatment of transfected BJ fibroblasts with T4. Cells were treated for 4 days be-39Figure 4.2: Flow cytometric analysis of BJ fibroblasts stably transduced with the αMHC re-porter. Treatment with various concentrations of T4, the circulating prohormone of thy-roid hormone, does not result in increased expression of EGFP.fore nucleoporation, then replated and treated for 2 more days before harvesting for flow cytometry.Despite achieving a nucleofection rate of 30–40% of the BJ fibroblasts (EGFP control group),negligible αMHC reporter driven fluorescence was observed from the T4 treated group. The trans-fected αMHC reporter is not inducible by T4 in the BJ fibroblast. The lack of thyroid hormoneinducibility in these constructs may be due to the difference in transcription factor profiles of BJfibroblasts compared to primary cardiomyocytes or differentiated cardiomyocytes. Tsika et al. [80]speculated that the transcription factors required for T3 inducibility are tissue-specific.4.1.3 The αMHC reporter is activated by iPSC-derived cardiomyocytesSpontaneously beating cardiomyocytes were derived from iPSCs using the techniques describedin Section 2.1.5. On day 31 after the beginning of the differentiation protocol, the cells weretrypsinized, transfected via nucleoporation with the αMHC reporter, and replated in tissue cul-ture wells coated with gelatin and fibronectin. Three days after replating, the cells were trypsinized40Figure 4.3: Flow cytometric analysis of BJ fibroblasts transiently transfected with the αMHCreporter. Treatment with various concentrations of T4, the circulating prohormone ofthyroid hormone, does not result in increased expression of EGFP.again for flow cytometry.Figure 4.4 shows the expression of EGFP as measured by flow cytometry. Plotting the cellsalong the dimensions of forward scatter (FSC, proportional to cell size) and side scatter (SSC, pro-portional to cell granularity) reveals two distinct populations – low side scatter and high side scatter.The high side scatter population exhibits low transfectability (panel B), as only ∼7% of EGFP-transfected cells were fluorescent. The low side scatter population exhibits high transfectability(panel C), with ∼48% of EGFP-transfected cells fluorescing. The difference in transfectabilitymay be partially due to the nucleofection program used, which was optimized for transfectingcardiomyocytes (see Section 2.2.1). The high side scatter population showed negligible αMHCreporter driven fluorescence, while a small proportion of low side scatter cells, ∼4%, were defini-tively positive expressers of αMHC-driven EGFP. The high side scatter population’s low activationof the αMHC, as well as its low transfectability using the nucelofection program optimized for41cardiomyocytes, suggests that it is likely a fibroblast population. Assuming ∼48% of the low sidescatter cells were successfully transfected with the αMHC reporter, then one may estimate that∼8.8% of this population (i.e., 4.23% divided by 48.5%) is positive for αMHC as an indication of acardiomyocyte-like cell type.4.2 Primary cardiomyocytes lose core phenotypes in vitroWe wanted to test the αMHC promoter in a known cardiomyocyte to establish a true positive control.We purchased a cardiomyocyte cell line (Celprogen Catalog #36044-15) and transfected the αMHCreporter using nucleofection. The results of this test are shown in Figure 4.5.Though these cells showed a small amount of αMHC-driven fluorescence, it was not the defini-tive result we expected for a supposed cardiomyocyte cell line. Given the transfectability rates of∼30% and ∼50% for the two populations (demarcated by forward and side scatter profile), thesecells only expressed αMHC-driven fluorescence at rates of 3.6% and 5.0%. To verify the identityof the cell line we tested for the expression of cardiac-specific proteins. The protein of interest wascTnT (also known as cardiac troponin T2, or TNNT2), a core part of the contractile apparatus ofcardiomyocytes that also contains troponin I, troponin C, and tropomyosin. Together, these proteinsregulate myosin and its ability to generate force by ratcheting against actin.Figure 4.6 shows the lack of expression of this protein in Celprogen cardiomyocytes. It is typicalof cultures of primary human cardiomyocytes to rapidly lose cardiac phenotypes in vitro, and thesecells, though marketed as possessing cardiac-specific proteins, are no exception. Both early (p2)and late passage (p6) cells fail to express cTnT. To rule out overgrowth by rapidly proliferatingfibroblasts, we cultured the same cells in medium containing cytosine arabinoside to suppress thegrowth of rapidly dividing cells. However, these cells cultured with this drug also failed to expresscTnT. The lack of this core element of the cardiac contractile apparatus indicates that these cellsare not reliable models of human cardiomyocytes.4.3 Cyclic strain affects iPSC differentiation to cardiomyocytesInspired by the high efficiencies of in vivo transdifferentiation, we asked if we can increase invitro cardiac differentiation efficiency by recreating some of the mechanical aspects of the cardiacmicroenvironment. To this end, we modified the monolayer differentiation method of Laflammeet al. [45], adding cyclic mechanical strain to explore whether this could speed up the differentia-tion process, convert more of the starting cell population to cardiomyocytes, generate more maturecardiomyocytes with more functional characteristics, or generate a more uniform population of car-diomyocytes.42Figure 4.4: Flow cytometric analysis of iPSC-derived cardiomyocytes transiently transfectedwith the αMHC reporter and EGFP as a positive control. A: Forward scatter (FSC) vsside scatter (SSC) separates the cells into two distinct populations – high and low sidescatter. B: Cells with high side scatter do not exhibit good transfectability and do notactivate the αMHC reporter. C: Cells with low side scatter have high transfectability. Asmall proportion of cells definitively express the αMHC-promoter driven EGFP.43Figure 4.5: Celprogen cardiomyocytes were nucleofected with the plasmid αMHC reporter totest promoter-driven EGFP activity. Untransfected cells were used as negative controls,while cells transfected with EGFP were used as positive controls. A: Forward and sidescatter profile of the cells reveal two distinct populations, dubbed “singles” and “dou-bles”. B: The “singles” population was less transfectable, as shown by ∼29% of cellsexpressing EGFP, and did not show αMHC-driven fluorescence to a significant degree.C: The ”doubles” population was more transfectable, as shown by ∼50% of cell express-ing EGFP, but did not show any more αMHC-driven fluorescence.44Figure 4.6: Western blot assay for presence of cTnT in Celprogen human cardiomyocytes.Jurkat: a human T lymphocyte cell line used as a negative control. p6: Celprogen car-diomyocytes at passage 6. p2 (Ara-C): Celprogen cardiomyocytes at passage 2, main-tained in medium containing cytosine arabinoside to suppress fibroblast growth. p2: Cel-progen cardiomyocytes at passage 2. Mouse heart lysate was used as a positive control.PKA C-α, a housekeeping gene, was used as a loading control.4.3.1 Cardiomyocyte differentiation is sensitive to cyclic strain parametersFigure 4.7 shows the modified cardiac differentiation protocol. The iPSCs are cultured to 90% con-fluence in a well of the BioFlex plate in anticipation of the application of cyclic strain. Strainingof the cells begins after the 4 days of BMP4 treatment (i.e., the end of the biochemical differentia-tion). The cyclic strain was applied continuously at 1Hz, with a peak piston displacement of 5mm,translating to an average strain of 5% in the plateau region of the membrane (Figure 3.12). A non-strained control well in the BioFlex plate was used, as well as a control well seeded on traditionaltissue culture polystyrene.Figure 4.8 shows a representative spontaneously beating cluster in the non-strained control wellof the BioFlex plate. Whereas the non-strained control and the polystyrene control exhibited sponta-neous beating 10 days after the start of Activin A treatment, the strained well never exhibited spon-taneous beating while strain was applied. However, after cyclic strain was stopped, 19 days after thestart of Activin A treatment, spontaneous beating was observed. This suggests that the strain exertedby the HexCycler was in a physiologically appropriate range to interfere with excitation-contractioncoupling. Cyclic strain also had an effect on the formation of the cardiomyocyte contractile appara-45Day 0 Day 4 Day 5 Day 9 Day ??Activin A BMP4 Begin cyclic strainFigure 4.7: A schematic of the differentiation protocol used to generate cardiomyocytes fromiPSCs. Cells are seeded on Matrigel on Day 0 and reach 90% confluence by Day 4.Activin A (100ng/mL) is applied for 24 hours, followed by 4 days of BMP4 treatment(10ng/mL) without medium change. After BMP4 treatment, 5% cyclic strain is appliedcontinously at 1Hz using the HexCycler.Figure 4.8: A: A spontaneously beating cluster of cardiomyocytes. The yellow line indicatesthe region used for kymograph analysis. B: Kymograph analysis gives a beat frequencyof 0.35Hz.tus, as shown by immunocytochemistry. Figure 4.9 shows staining of iPSC-derived cardiomyocytes.Cell nuclei are stained blue by DAPI, which binds to regions of DNA rich in adenine and thymine.The proteins of interest are stained green. Those nuclei not surrounded by green belong to non-cardiomyocytes, most likely cells that have differentiated to a cardiac fibroblast lineage. Both theunstrained and strained cells stained positively for cTnT, but the strained cells exhibited punctatestaining while the unstrained cells had more filamentous staining. Sarcomeric striations revealedby α-actinin staining were common in the unstrained specimen, but much rarer in the unstrainedspecimen. This suggests that the strain regimen hindered the maturation of these cardiomyocytes.46Figure 4.9: Cyclic strain affects the formation of sarcomeric striations. Anti-α-actinin anti-body was used to stain for sarcomeric striations. Striations are common in the unstrainedspecimen (inset) but much rarer in the strained specimen. DAPI stains cell nuclei blue,while green indicates either cTnT or α-actinin.47Chapter 5Discussion and conclusion5.1 Reporters of cardiac-like myocytesBased on previous work characterizing the thyroid response elements present in the αMHC pro-moter [80], we expected thyroid hormone to be able to activate the αMHC reporter, even in non-cardiac cells. We wanted to activate the αMHC-EGFP fluorescence reporter with thyroid hormoneto establish a baseline reading as a positive control. However, we determined that the αMHC re-porter is not inducible by thyroid hormone in non-cardiac cells. This suggests that, in addition tothyroid hormone, other unknown cardiac-specific transcription factors are required to mediate theresponse of the αMHC promoter to thyroid hormone.We delivered the αMHC reporter into BJ fibroblasts using both lentiviral transduction and trans-fection using nucleoporation. The lentiviral transduction system co-opts the gene delivery mecha-nisms of the human immunodeficiency virus to create stably transduced cells. Both gene transfertechniques are capable of generating cells that stably express the transgene, though lentiviral trans-duction can achieve higher efficiency than nucleofection [11]. The strong electric fields used topermeabilize cell membranes in nucleofection also result in a significant amount of cell lysis, andnot all permeabilized cells successfully take up plasmid DNA. Both these factors contribute to thelower transfection efficiency of nucleofection compared to lentiviral transduction.When transfected, the αMHC fluorescence reporter is not significantly activated by iPSC-derived cardiomyocytes, with only 8.8% of cells fluorescing. Previous reports have used lentiviraltransduction to stably integrate this reporter into pluripotent stem cells prior to differentiation tocardiac lineages [27, 35, 44]. For example, Fu et al. [27] report that 18.1% ± 11.2% of humanfibroblasts transduced with cardiac-transdifferentiating factors activated a similar αMHC fluores-cence reporter. Future work with this reporter may require such lentiviral methods.Laflamme et al. [45] were able to achieve over 50% βMHC-positive cells using the same mono-layer differentiation protocol used in our experiments, but they began with human embryonic stem48cells. This is likely an over-estimate of the differentiation efficiency, as it was determined by visualinspection of fixed and stained specimens that obscures the underlying support layer of fibroblast-like cells. βMHC is the isoform of MHC predominantly expressed in the human ventricle, whileαMHC is predominantly expressed in the atrium. Thus the low activation of our αMHC reportermay be a reflection that the cardiac differentiation protocol generates predominantly ventricularmyocytes [64]. Much of the difference in differentiation efficiency could also be accounted for byepigenetic variability between starting cells. Kattman et al. [41] demonstrated that human embry-onic stem cells lines have different cardiac differentiation capacities, and it has been speculated thatiPSCs have similar if not greater variability [10]. iPSC cell line variability may be due to the cellsused to create the cell line [50], as well as the passage history [61].That said, activation of the reporter only represents a single aspect of the cardiac phenotype. Amore rigorous profiling of gene expression, and functional testing of electrophysiology and calciumhandling characteristics will be required to fully assess the maturity of differentiated cells.5.2 The effects of cyclic strain on cardiomyocyte differentiation andmaturationOrganogenesis is a delicately choreographed event that is sensitive to changes in timing. We showedthat the application of ∼5% cyclic strain at 1Hz at the end of biochemical differentiation in the mono-layer method of cardiac differentiation inhibits spontaneous contractions and disrupts the formationof sarcomeres. It remains to be seen whether altering the timing, frequency, and/or magnitude ofstrain may speed up or improve differentiation. The HexCycler, with its 6 independently drivenactuators, is well-positioned to facilitate the optimization of these parameters.The presence of spontaneously contracting clusters of cardiomyocytes indicates the presence ofpacemaker cells (also known as nodal cells, for the sinoatrial node). These pacemaker cells entrainthe nearby myocytes to contract in synchrony. Though not definitively shown with immunoflu-orescence staining, the presence of these multicellular synchronously beating islands implies thesuccessful formation of intercalated discs between cardiomyocyte. These cell interfaces consist ofadherens junctions, desmosomes, and gap junctions and enable the cells to form a electrochemicalsyncytium.Immunocytochemistry and bright-field microscopy reveals that the cardiomyocytes are sup-ported by a layer of non-myocytes, likely cells that have differentiated to the cardiac fibroblast lin-eage. These cells, along with the pacemaker cells that drive the spontaneously beating cell clusters,are indicative of the heterogeneity of the cell population. Each tissue culture well likely containscardiac fibroblasts, nodal cells, atrial myocytes, and ventricular myocytes.Though the inhibition of spontaneous contraction coincided with lack of sarcomeres in our ex-periments, the lack of spontaneous beating may counterintuitively indicate greater maturity – it hasbeen shown that primary myocytes from embryonic hearts beat spontaneously in vitro [22], while49those isolated from adult hearts do not beat without stimulation [2, 58]. This is attributed to thegreater expression of certain ion channels as cardiomyocytes mature, which alters the electrophysi-ology and thus the excitation-contraction threshold of cells [19, 69].5.3 Potential applications of in vitro cardiomyocytes and futuredirectionsA systematic review of the global burden of disease revealed that cardiomyopathy and myocarditiskilled an estimated 443,300 people in 2013, an increase of 51% over the number of deaths in 1990.The same study puts the number of deaths attributable to all heart disease, including rheumatic heartdisease, ischaemic heart disease, hypertensive heart disease, atrial fibrillation, and endocarditis, atover 10.1 million [56]. The uses of a reliable source of in vitro cardiomyocytes to understand andtreat these diseases are numerous. I discuss some of these applications below.5.3.1 Using iPSC-derived cardiomyocytes to model cardiomyopathiesGreat strides have been made in using iPSCs from patients with known cardiomyopathies to modeldisease and gain insights into possible treatments. For example, Itzhaki et al. [36] used these tech-niques to study dermal fibroblasts from a patient with long QT syndrome to generate iPSCs, thendifferentiated these iPSCs to cardiomyocytes. (Long QT syndrome is named for the extension of thetime interval between the Q and T waveforms of the electrocardiogram.) With this high-fidelity invitro model of patient-specific disease, they were then able to test existing and novel pharmacologi-cal agents to see which had positive effects on the disease phenotype. Perhaps even more exciting isthe prospect of modelling cardiac diseases without a known genetic component, and the modellingof the response of cardiac tissues to infectious agents.5.3.2 Mature cardiomyocytes for drug toxicity testingIn the 1990s, several drugs were withdrawn from the market after it was revealed that they wereresponsible for multiple unexpected cardiac deaths [14]. The pharmaceutical industry later learnedthat these drugs adversely affected ion channels that altered the electrophysiology of cardiomy-ocytes, which prolonged the QT interval and put the patient at risk of fatal arrhythmias, much likepatients with hereditary long QT syndrome. In response to these deaths and the regulations thatfollowed, the pharmaceutical industry has spent hundreds of millions of dollars testing candidatedrugs for cardiotoxicity. Lacking a reliable in vitro model, the current standard of testing involvesa “thorough QT study” in which healthy human volunteers are exposed to the drug to see if it pro-longs their QT interval. This testing procedure is arduous, expensive, and worst of all, puts healthyvolunteers at risk.iPSC-derived and transdifferentiated cardiomyocytes have the potential to dramatically increase50the fidelity of these tests while decreasing costs. While such in vivo models of the heart may notbe able to recreate a full electrocardiogram, one may use patch-clamp techniques to measure theeffect of drugs on a cardiomyocyte’s ion channel conductances. It is known, for example, thatinhibition of the human ether-a`-go-go-related gene (hERG) potassium ion channel can lead to longQT syndrome [14]. However, the accuracy of these models depends on the ability to generate maturecardiomyocytes, as many of the ion channels targeted by cardiotoxic drugs are only expressed inadult cells.5.3.3 Cellular therapy for heart failureOvercoming the poor regenerative potential of cardiomyocytes could revolutionize the treatment ofheart failure. Robey et al. [66] estimated that 1 billion new cardiomyocytes are needed after my-ocardial infarction to restore function . The source of these cells could be cardiomyocytes generatedfrom directed differentiation of iPSCs or cardiomyocytes transdifferentiated from native cardiacfibroblasts. Several groups showed that in situ transdifferentiation of cardiac fibroblasts to car-diomyocytes can modestly improve cardiac function in mice after induced myocardial infarction[40, 63, 73]. Such in situ techniques do not have to contend with the problem of cell engraftment,but those protocols that involve reprogramming somatic cells to iPSCs followed by directed differ-entiation to cardiomyocytes run the risk of generating teratomas [1].Alternatively, cardiomyocytes may be produced in vitro and implanted to a failing heart to re-store function. Such a technique may use also the patient’s own cells as starting materials to avoidimmunological rejection. Much work remains to be done to ensure that the transplanted cells engraftto the existing myocardium, and that homogeneous populations of cells are transplanted to matchtheir targeted repair site and avoid ectopic pacemakers that may be arrhythmogenic.5.3.4 Tissue engineeringTissue engineering is defined as the growth of three-dimensional tissues in vitro, with the ultimategoal of creating a fully functional organ by directing the proliferation, differentiation, and remod-elling of cells. The current state of the art of directed cardiac differentiation remains far removedfrom realizing this goal, though engineered cardiac constructs (loops of cardiac tissue) have beenemployed in animal models to improve cardiac function after infarction [89]. Fine control over theresults of cardiac differentiation, in terms of the mixture of cell types and maturity, will be neededfor the success of such a construct in humans.5.3.5 Future directionsProblems of iPSC variability leading to poor reproducibility of differentiation have plagued our ex-periments, as well as those of our collaborators. The mTeSR1 medium used in the culture of our51iPSCs contain bovine serum albumin, which is known to exhibit considerable variability from lotto lot [51]. Indeed, the creators of the mTeSR1 medium acknowledge this variability and empha-size rigorous quality control [47]. More recently, Chen et al. [12] developed a chemically defined,albumin-free medium called E8 that has been shown to able to sustain iPSC pluripotency in culture.Use of E8, in conjunction with a recombinant truncated version of vitronectin (an ECM glycopro-tein) as an alternative to Matrigel, may reduce some of the variability seen in our experiments.The applications in Section 5.3 could all benefit from a reliable method to separate and quan-tify the cardiac cell populations (atrial myocytes, ventricular myocytes, nodal cells, and cardiacfibroblasts). Such a method would also be crucial for the optimization of differentiation protocols,to determine the effects of novel stimuli. Such quantification and separation of cell populationsmay be done with flow cytometry, employing a panel of cell-type–specific antibodies. For exam-ple, αMHC and βMHC may be used as a rough marker of atrial and ventricular myocyte pop-ulations, respectively [64], while potassium/sodium hyperpolarization-activated cyclic nucleotide-gated channel 4 (HCN4) may be used as a marker of nodal cells [16], and thymocyte differentiationantigen 1 (THY1) may be used as a marker of cardiac fibroblasts [32]. Such a high-throughputquantitative method could be complemented with more detailed imaging and functional measures,such as immunofluorescence imaging, calcium flux imaging and patch-clamping. The presence ofcalcium-handing proteins, such as the ryanodine receptor, calsequestrin, and phosopholamban, canbe verified with RT-PCR and immunofluorescence imaging. The use of cell-type specific markerscan also verify whether the high side-scatter population of Figure 4.4 is of a fibroblast, endothelial,or some other lineage.We have observed improvements in cardiomyocyte beating synchrony and increases in the sizeof spontaneously beating clusters over the period of several weeks in culture. We postulate thatthis may be due to the electromechanical influence of more mature cardiomyocytes on less maturecells. As the more mature cardiomyocytes contract, they pull on adjacent cells and depolarize theirmembranes (if the cells have formed electrochemical couplings via gap junctions). This may beenough to tip the immature cell towards a cardiac lineage. The combination of electrical signallingand mechanical stretch presents another dimension for bioengineers to recreate the cardiac microen-vironment. More specifically, the tissue culture well can be outfitted with electrodes to create anelectrochemical cell. The experimenter can then bias the two electrodes to create an electric fieldand force the cells to depolarize.In both cyclic mechanical stretch and entrained electrical depolarization, the nascent cardiomy-ocyte will no doubt be sensitive to parameters of frequency and magnitude. The cell can be thoughtof as a simple harmonic oscillator with a time-varying natural frequency. If we perturb the system atthe correct frequency, with a gentle enough magnitude not to obliterate delicate internal structuresand balances, the cell may respond by developing the phenotypes of a mature cardiomyocyte. Theaim of future work will be to find these natural frequencies.525.4 ConclusionThe field of cardiac engineering has advanced considerably over the last decade, and exciting ad-vances in bioreactor design are imminent. The problems to be solved at this juncture of engineeringand cell biology will no doubt be of interest to interdisciplinary scholars of the future.53Bibliography[1] M. Abad, L. Mosteiro, C. Pantoja, M. Canamero, T. Rayon, I. Ors, O. Grana, D. Megias,O. Dominguez, D. Martinez, M. Manzanares, S. Ortega, and M. Serrano. Reprogramming invivo produces teratomas and ips cells with totipotency features. Nature, 502(7471):340–+,Oct. 2013. doi:10.1038/nature12586. → pages 9, 51[2] R. C. Addis and J. A. Epstein. Induced regeneration–the progress and promise of directreprogramming for heart repair. Nat Med, 19(7):829–836, Jul 2013. doi:10.1038/nm.3225.URL http://dx.doi.org/10.1038/nm.3225. → pages 11, 50[3] R. C. Addis, J. L. Ifkovits, F. Pinto, L. D. Kellam, P. Esteso, S. Rentschler, N. Christoforou,J. A. Epstein, and J. D. Gearhart. Optimization of direct fibroblast reprogramming tocardiomyocytes using calcium activity as a functional measure of success. J Mol CellCardiol, 60:97–106, Jul 2013. doi:10.1016/j.yjmcc.2013.04.004. URLhttp://dx.doi.org/10.1016/j.yjmcc.2013.04.004. → pages 10[4] S. J. Arnold and E. J. Robertson. Making a commitment: cell lineage allocation and axispatterning in the early mouse embryo. Nat Rev Mol Cell Biol, 10(2):91–103, Feb. 2009.ISSN 1471-0072. URL http://dx.doi.org/10.1038/nrm2618. → pages 4[5] A. Banes. Biocompatible polyorganosiloxane composition for cell culture apparatus, Dec. 61988. URL https://www.google.ca/patents/US4789601. US Patent 4,789,601. → pages 33[6] N. H. Bishopric. Evolution of the heart from bacteria to man. Ann N Y Acad Sci, 1047:13–29,Jun 2005. doi:10.1196/annals.1341.002. URL http://dx.doi.org/10.1196/annals.1341.002.→ pages 7[7] A. Bondue and C. Blanpain. Mesp1: a key regulator of cardiovascular lineage commitment.Circ Res, 107(12):1414–1427, Dec 2010. doi:10.1161/CIRCRESAHA.110.227058. URLhttp://dx.doi.org/10.1161/CIRCRESAHA.110.227058. → pages 8[8] A. Bondue, G. Lapouge, C. Paulissen, C. Semeraro, M. Iacovino, M. Kyba, and C. Blanpain.Mesp1 acts as a master regulator of multipotent cardiovascular progenitor specification. CellStem Cell, 3(1):69–84, Jul 2008. doi:10.1016/j.stem.2008.06.009. URLhttp://dx.doi.org/10.1016/j.stem.2008.06.009. → pages 4[9] M. Buckingham, S. Meilhac, and S. Zaffran. Building the mammalian heart from two sourcesof myocardial cells. Nat Rev Genet, 6(11):826–835, Nov 2005. doi:10.1038/nrg1710. URLhttp://dx.doi.org/10.1038/nrg1710. → pages 754[10] P. Burridge, G. Keller, J. Gold, and J. Wu. Production of de novo cardiomyocytes: Humanpluripotent stem cell differentiation and direct reprogramming. Cell Stem Cell, 10(1):16 – 28,2012. ISSN 1934-5909. doi:http://dx.doi.org/10.1016/j.stem.2011.12.013. URLhttp://www.sciencedirect.com/science/article/pii/S1934590911005947. → pages 4, 9, 49[11] F. Cao, X. Xie, T. Gollan, L. Zhao, K. Narsinh, R. J. Lee, and J. C. Wu. Comparison ofgene-transfer efficiency in human embryonic stem cells. Molecular Imaging and Biology, 12(1):15–24, Mar. 2009. ISSN 1860-2002. URLhttp://www.ncbi.nlm.nih.gov/pmc/articles/PMC2803751/. → pages 48[12] G. Chen, D. R. Gulbranson, Z. Hou, J. M. Bolin, V. Ruotti, M. D. Probasco, K. Smuga-Otto,S. E. Howden, N. R. Diol, N. E. Propson, R. Wagner, G. O. Lee, J. Antosiewicz-Bourget,J. M. C. Teng, and J. A. Thomson. Chemically defined conditions for human ipsc derivationand culture. Nat Methods, 8(5):424–429, May 2011. doi:10.1038/nmeth.1593. URLhttp://dx.doi.org/10.1038/nmeth.1593. → pages 52[13] J. X. Chen, M. Krane, M.-A. Deutsch, L. Wang, M. Rav-Acha, S. Gregoire, M. C. Engels,K. Rajarajan, R. Karra, E. D. Abel, J. C. Wu, D. Milan, and S. M. Wu. Inefficientreprogramming of fibroblasts into cardiomyocytes using gata4, mef2c, and tbx5. Circ Res,111(1):50–55, Jun 2012. doi:10.1161/CIRCRESAHA.112.270264. URLhttp://dx.doi.org/10.1161/CIRCRESAHA.112.270264. → pages 10[14] K. R. Chi. Revolution dawning in cardiotoxicity testing. Nat Rev Drug Discov, 12(8):565–567, Aug. 2013. ISSN 1474-1776. URL http://dx.doi.org/10.1038/nrd4083. → pages50, 51[15] J. Choi, M. L. Costa, C. S. Mermelstein, C. Chagas, S. Holtzer, and H. Holtzer. Myodconverts primary dermal fibroblasts, chondroblasts, smooth muscle, and retinal pigmentedepithelial cells into striated mononucleated myoblasts and multinucleated myotubes. ProcNatl Acad Sci U S A, 87(20):7988–7992, Oct 1990. → pages 9[16] V. M. Christoffels, G. J. Smits, A. Kispert, and A. F. M. Moorman. Development of thepacemaker tissues of the heart. Circulation Research, 106(2):240–254, 2010.doi:10.1161/CIRCRESAHA.109.205419. URLhttp://circres.ahajournals.org/content/106/2/240.abstract. → pages 52[17] N. Christoforou, M. Chellappan, A. F. Adler, R. D. Kirkton, T. Wu, R. C. Addis, N. Bursac,and K. W. Leong. Transcription factors myocd, srf, mesp1 and smarcd3 enhance thecardio-inducing effect of gata4, tbx5, and mef2c during direct cellular reprogramming. PLoSONE, 8(5):e63577–, May 2013. doi:10.1371/journal.pone.0063577. URLhttp://dx.doi.org/10.1371%2Fjournal.pone.0063577. → pages 10[18] M. M. Davidson, C. Nesti, L. Palenzuela, W. F. Walker, E. Hernandez, L. Protas, M. Hirano,and N. D. Isaac. Novel cell lines derived from adult human ventricular cardiomyocytes. JMol Cell Cardiol, 39(1):133–147, Jul 2005. doi:10.1016/j.yjmcc.2005.03.003. URLhttp://dx.doi.org/10.1016/j.yjmcc.2005.03.003. → pages 755[19] M. P. Davies, R. H. An, P. Doevendans, S. Kubalak, K. R. Chien, and R. S. Kass.Developmental changes in ionic channel activity in the embryonic murine heart. Circ Res, 78(1):15–25, Jan 1996. → pages 50[20] R. L. Davis, H. Weintraub, and A. B. Lassar. Expression of a single transfected cdna convertsfibroblasts to myoblasts. Cell, 51(6):987–1000, Dec 1987. → pages 9[21] A. M. De Jong, A. H. Maass, S. U. Oberdorf-Maass, R. A. De Boer, W. H. Van Gilst, andI. C. Van Gelder. Cyclical stretch induces structural changes in atrial myocytes. J Cell MolMed, 17(6):743–753, Jun 2013. doi:10.1111/jcmm.12064. URLhttp://dx.doi.org/10.1111/jcmm.12064. → pages 11[22] R. L. DeHaan and S. H. Gottlieb. The electrical activity of embryonic chick heart cellsisolated in tissue culture singly or in interconnected cell sheets. J Gen Physiol, 52(4):643–665, Oct 1968. → pages 49[23] A. J. Engler, S. Sen, H. L. Sweeney, and D. E. Discher. Matrix elasticity directs stem celllineage specification. Cell, 126(4):677–689, Aug. 2006. ISSN 0092-8674. URLhttp://www.sciencedirect.com/science/article/pii/S0092867406009615. → pages 11, 12[24] A. J. Engler, C. Carag-Krieger, C. P. Johnson, M. Raab, H. Y. Tang, D. W. Speicher, J. W.Sanger, J. M. Sanger, and D. E. Discher. Embryonic cardiomyocytes beat best on a matrixwith heart-like elasticity: scar-like rigidity inhibits beating. Journal of Cell Science, 121(22):3794–3802, Nov. 2008. doi:10.1242/jcs.029678. → pages 11[25] M. Flanagan, J. M. Gimble, G. Yu, X. Wu, and S. Li. Competitive electroporationformulation for cell therapy. Cancer gene therapy, 18(8):579–586, June 2011. ISSN1476-5500. URL http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3238913/. → pages 18[26] J. Folkman and A. Moscona. Role of cell shape in growth control. Nature, 273(5661):345–349, June 1978. URL http://dx.doi.org/10.1038/273345a0. → pages 12[27] J.-D. Fu, N. Stone, L. Liu, C. Spencer, L. Qian, Y. Hayashi, P. Delgado-Olguin, S. Ding,B. Bruneau, and D. Srivastava. Direct reprogramming of human fibroblasts toward acardiomyocyte-like state. Stem Cell Reports, 1(3):235 – 247, 2013. ISSN 2213-6711.doi:http://dx.doi.org/10.1016/j.stemcr.2013.07.005. URLhttp://www.sciencedirect.com/science/article/pii/S2213671113000544. → pages 10, 48[28] J. Gulick, A. Subramaniam, J. Neumann, and J. Robbins. Isolation and characterization ofthe mouse cardiac myosin heavy chain genes. Journal of Biological Chemistry, 266(14):9180–9185, 1991. URL http://www.jbc.org/content/266/14/9180.abstract. → pages 18[29] T. A. Gustafson, J. J. Bahl, B. E. Markham, W. R. Roeske, and E. Morkin. Hormonalregulation of myosin heavy chain and alpha-actin gene expression in cultured fetal rat heartmyocytes. Journal of Biological Chemistry, 262(27):13316–13322, Sept. 1987. URLhttp://www.jbc.org/content/262/27/13316. → pages 38[30] T. A. Gustafson, B. E. Markham, J. J. Bahl, and E. Morkin. Thyroid hormone regulatesexpression of a transfected alpha-myosin heavy-chain fusion gene in fetal heart cells. ProcNatl Acad Sci U S A, 84(10):3122–3126, May 1987. → pages 3856[31] R. Horiuchi, T. Akimoto, Z. Hong, and T. Ushida. Cyclic mechanical strain maintains nanogexpression through pi3k/akt signaling in mouse embryonic stem cells. Exp Cell Res, 318(14):1726–1732, Aug 2012. doi:10.1016/j.yexcr.2012.05.021. URLhttp://dx.doi.org/10.1016/j.yexcr.2012.05.021. → pages 12[32] F. Hudon-David, F. Bouzeghrane, P. Couture, and G. Thibault. Thy-1 expression by cardiacfibroblasts: lack of association with myofibroblast contractile markers. J Mol Cell Cardiol,42(5):991–1000, May 2007. doi:10.1016/j.yjmcc.2007.02.009. URLhttp://dx.doi.org/10.1016/j.yjmcc.2007.02.009. → pages 52[33] J. Hudson, D. Titmarsh, A. Hidalgo, E. Wolvetang, and J. Cooper-White. Primitive cardiaccells from human embryonic stem cells. Stem Cells Dev, 21(9):1513–1523, Jun 2012.doi:10.1089/scd.2011.0254. URL http://dx.doi.org/10.1089/scd.2011.0254. → pages 9[34] M. Ieda, J.-D. Fu, P. Delgado-Olguin, V. Vedantham, Y. Hayashi, B. G. Bruneau, andD. Srivastava. Direct reprogramming of fibroblasts into functional cardiomyocytes by definedfactors. Cell, 142(3):375 – 386, 2010. ISSN 0092-8674.doi:http://dx.doi.org/10.1016/j.cell.2010.07.002. URLhttp://www.sciencedirect.com/science/article/pii/S0092867410007713. → pages xi, 9[35] J. F. Islas, Y. Liu, K.-C. Weng, M. J. Robertson, S. Zhang, A. Prejusa, J. Harger,D. Tikhomirova, M. Chopra, D. Iyer, M. Mercola, R. G. Oshima, J. T. Willerson, V. N.Potaman, and R. J. Schwartz. Transcription factors ets2 and mesp1 transdifferentiate humandermal fibroblasts into cardiac progenitors. Proc Natl Acad Sci U S A, 109(32):13016–13021,Aug 2012. doi:10.1073/pnas.1120299109. URLhttp://dx.doi.org/10.1073/pnas.1120299109. → pages 10, 48[36] I. Itzhaki, L. Maizels, I. Huber, L. Zwi-Dantsis, O. Caspi, A. Winterstern, O. Feldman,A. Gepstein, G. Arbel, H. Hammerman, M. Boulos, and L. Gepstein. Modelling the long qtsyndrome with induced pluripotent stem cells. Nature, 471(7337):225–229, Mar. 2011. ISSN0028-0836. URL http://dx.doi.org/10.1038/nature09747. → pages 50[37] D. E. Jaalouk and J. Lammerding. Mechanotransduction gone awry. Nat Rev Mol Cell Biol,10(1):63–73, Jan. 2009. ISSN 1471-0072. URL http://dx.doi.org/10.1038/nrm2597.mechanotransduction mechanobiology review. → pages 12[38] R. J. Jackson, M. T. Howell, and A. Kaminski. The novel mechanism of initiation ofpicornavirus rna translation. Trends Biochem Sci, 15(12):477–483, Dec 1990. → pages 19[39] S. Jang, H. Krusslich, M. Nicklin, G. Duke, A. Palmenberg, and E. Wimmer. A segment ofthe 5’ nontranslated region of encephalomyocarditis virus rna directs internal entry ofribosomes during in vitro translation. J Virol, 62(8):2636–2643, 1988. → pages 19[40] T. M. Jayawardena, B. Egemnazarov, E. A. Finch, L. Zhang, J. A. Payne, K. Pandya,Z. Zhang, P. Rosenberg, M. Mirotsou, and V. J. Dzau. Microrna-mediated in vitro and in vivodirect reprogramming of cardiac fibroblasts to cardiomyocytes. Circ Res, 110(11):1465–1473, May 2012. doi:10.1161/CIRCRESAHA.112.269035. URLhttp://dx.doi.org/10.1161/CIRCRESAHA.112.269035. → pages 10, 5157[41] S. J. Kattman, A. D. Witty, M. Gagliardi, N. C. Dubois, M. Niapour, A. Hotta, J. Ellis, andG. Keller. Stage-specific optimization of activin/nodal and bmp signaling promotes cardiacdifferentiation of mouse and human pluripotent stem cell lines. Cell Stem Cell, 8(2):228–240,Feb 2011. doi:10.1016/j.stem.2010.12.008. URLhttp://dx.doi.org/10.1016/j.stem.2010.12.008. → pages 49[42] F. Khademolhosseini and M. Chiao. Magnetically actuated cellular strain assessment tool; astudy on the strain induced directional migration of human endothelial cells. In Solid-StateSensors, Actuators and Microsystems Conference (TRANSDUCERS), 2011 16thInternational, pages 2275–2278, 2011. doi:10.1109/TRANSDUCERS.2011.5969360. →pages 25[43] T. K. Kim, J. K. Kim, and O. C. Jeong. Measurement of nonlinear mechanical properties ofpdms elastomer. Microelectronic Engineering, 88(8):1982–1985, Aug. 2011. ISSN0167-9317. URL http://www.sciencedirect.com/science/article/pii/S0167931710005964. →pages 32, 33[44] H. Kita-Matsuo, M. Barcova, N. Prigozhina, N. Salomonis, K. Wei, J. G. Jacot, B. Nelson,S. Spiering, R. Haverslag, C. Kim, M. Talantova, R. Bajpai, D. Calzolari, A. Terskikh, A. D.McCulloch, J. H. Price, B. R. Conklin, H. S. V. Chen, and M. Mercola. Lentiviral vectors andprotocols for creation of stable hesc lines for fluorescent tracking and drug resistanceselection of cardiomyocytes. PLoS ONE, 4(4):e5046, 04 2009.doi:10.1371/journal.pone.0005046. URLhttp://dx.doi.org/10.1371%2Fjournal.pone.0005046. → pages 18, 19, 48[45] M. A. Laflamme, K. Y. Chen, A. V. Naumova, V. Muskheli, J. A. Fugate, S. K. Dupras,H. Reinecke, C. Xu, M. Hassanipour, S. Police, C. O’Sullivan, L. Collins, Y. Chen,E. Minami, E. A. Gill, S. Ueno, C. Yuan, J. Gold, and C. E. Murry. Cardiomyocytes derivedfrom human embryonic stem cells in pro-survival factors enhance function of infarcted rathearts. Nat Biotech, 25(9):1015–1024, Sept. 2007. ISSN 1087-0156. URLhttp://dx.doi.org/10.1038/nbt1327. → pages 3, 9, 17, 42, 48[46] A. M. Lompre´, B. Nadal-Ginard, and V. Mahdavi. Expression of the cardiac ventricularalpha- and beta-myosin heavy chain genes is developmentally and hormonally regulated.Journal of Biological Chemistry, 259(10):6437–6446, May 1984. URLhttp://www.jbc.org/content/259/10/6437.abstract. → pages 38[47] T. E. Ludwig, V. Bergendahl, M. E. Levenstein, J. Yu, M. D. Probasco, and J. A. Thomson.Feeder-independent culture of human embryonic stem cells. Nat Methods, 3(8):637–646,Aug 2006. doi:10.1038/nmeth902. URL http://dx.doi.org/10.1038/nmeth902. → pages 52[48] L. Macri-Pellizzeri, B. Pelacho, A. Sancho, O. Iglesias-Garcia, A. M. Simon-Yarza,M. Soriano-Navarro, S. Gonzalez-Granero, J. M. Garcia-Verdugo, E. M. De-Juan-Pardo, andF. Prosper. Substrate stiffness and composition specifically direct differentiation of inducedpluripotent stem cells. Tissue Engineering Part A, 21(9-10):1633–1641, May 2015.doi:10.1089/ten.tea.2014.0251. → pages 1258[49] V. Mahdavi, E. E. Strehler, M. Periasamy, D. F. Wieczorek, S. Izumo, and B. Nadal-Ginard.Sarcomeric myosin heavy chain gene family: organization and pattern of expression. Med SciSports Exerc, 18(3):299–308, Jun 1986. → pages 38[50] A. Martinez-Fernandez, T. J. Nelson, and A. Terzic. Nuclear reprogramming strategymodulates differentiation potential of induced pluripotent stem cells. Journal ofCardiovascular Translational Research, 4(2):131–137, Nov. 2010. ISSN 1937-5395. URLhttp://www.ncbi.nlm.nih.gov/pmc/articles/PMC3047690/. → pages 49[51] S. H. McKiernan and B. D. Bavister. Different lots of bovine serum albumin inhibit orstimulate in vitro development of hamster embryos. In Vitro Cell Dev Biol, 28A(3 Pt 1):154–156, Mar 1992. → pages 52[52] C. C. Moore, C. H. Lugo-Olivieri, E. R. McVeigh, and E. A. Zerhouni. Three-dimensionalsystolic strain patterns in the normal human left ventricle: characterization with tagged mrimaging. Radiology, 214(2):453–466, Feb 2000. doi:10.1148/radiology.214.2.r00fe17453.URL http://dx.doi.org/10.1148/radiology.214.2.r00fe17453. → pages 24[53] K. L. Moore, T. V. N. Persaud, and M. G. Torchia. The developing human : clinically orientedembryology, 2013. ISSN 9781455707492 145570749X. URL http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=458794. → pages 1[54] A. F. M. Moorman and V. M. Christoffels. Cardiac chamber formation: development, genes,and evolution. Physiol Rev, 83(4):1223–1267, Oct 2003. doi:10.1152/physrev.00006.2003.URL http://dx.doi.org/10.1152/physrev.00006.2003. → pages 7[55] E. Morkin. Regulation of myosin heavy chain genes in the heart. Circulation, 87(5):1451–1460, May 1993. → pages 18[56] G. . Mortality and C. of Death Collaborators. Global, regional, and national age-sex specificall-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematicanalysis for the global burden of disease study 2013. The Lancet, 385(9963):117 – 171, 2015.ISSN 0140-6736. doi:http://dx.doi.org/10.1016/S0140-6736(14)61682-2. URLhttp://www.sciencedirect.com/science/article/pii/S0140673614616822. → pages 50[57] P. Mott, J. Dorgan, and C. Roland. The bulk modulus and poisson’s ratio of incompressiblematerials. Journal of Sound and Vibration, 312(45):572–575, May 2008. ISSN 0022-460X.URL http://www.sciencedirect.com/science/article/pii/S0022460X08000771. → pages 33[58] C. L. Mummery, J. H. Zhang, E. S. Ng, D. A. Elliott, A. G. Elefanty, and T. J. Kamp.Differentiation of human embryonic stem cells and induced pluripotent stem cells tocardiomyocytes a methods overview. Circulation Research, 111(3):344–358, July 2012.doi:10.1161/CIRCRESAHA.110.227512. → pages 50[59] Y.-J. Nam, K. Song, X. Luo, E. Daniel, K. Lambeth, K. West, J. A. Hill, J. M. DiMaio, L. A.Baker, R. Bassel-Duby, and E. N. Olson. Reprogramming of human fibroblasts toward acardiac fate. Proc Natl Acad Sci U S A, 110(14):5588–5593, Apr. 2013.doi:10.1073/pnas.1301019110. → pages 1059[60] E. N. Olson. Gene regulatory networks in the evolution and development of the heart.Science, 313(5795):1922–1927, 2006. doi:10.1126/science.1132292. URLhttp://www.sciencemag.org/content/313/5795/1922.abstract. → pages 6, 7[61] S. L. Paige, T. Osugi, O. K. Afanasiev, L. Pabon, H. Reinecke, and C. E. Murry. Endogenouswnt\β-catenin signaling is required for cardiac differentiation in human embryonic stemcells. PLoS ONE, 5(6):e11134, 06 2010. doi:10.1371/journal.pone.0011134. URLhttp://dx.doi.org/10.1371%2Fjournal.pone.0011134. → pages 9, 49[62] S. Protze, S. Khattak, C. Poulet, D. Lindemann, E. M. Tanaka, and U. Ravens. A newapproach to transcription factor screening for reprogramming of fibroblasts tocardiomyocyte-like cells. J Mol Cell Cardiol, 53(3):323 – 332, 2012. ISSN 0022-2828.doi:http://dx.doi.org/10.1016/j.yjmcc.2012.04.010. URLhttp://www.sciencedirect.com/science/article/pii/S0022282812001575. → pages 10[63] L. Qian, Y. Huang, C. I. Spencer, A. Foley, V. Vedantham, L. Liu, S. J. Conway, J.-d. Fu, andD. Srivastava. In vivo reprogramming of murine cardiac fibroblasts into inducedcardiomyocytes. Nature, 485(7400):593–598, May 2012. ISSN 0028-0836. URLhttp://dx.doi.org/10.1038/nature11044. → pages 12, 51[64] P. J. Reiser, M. A. Portman, X. H. Ning, and C. Schomisch Moravec. Human cardiac myosinheavy chain isoforms in fetal and failing adult atria and ventricles. Am J Physiol Heart CircPhysiol, 280(4):H1814–H1820, Apr 2001. → pages 49, 52[65] P. R. Riley, M. Gertsenstein, K. Dawson, and J. C. Cross. Early exclusion of hand1-deficientcells from distinct regions of the left ventricular myocardium in chimeric mouse embryos.Dev Biol, 227(1):156–168, Nov 2000. doi:10.1006/dbio.2000.9864. URLhttp://dx.doi.org/10.1006/dbio.2000.9864. → pages 7[66] T. E. Robey, M. K. Saiget, H. Reinecke, and C. E. Murry. Systems approaches to preventingtransplanted cell death in cardiac repair. Journal of Molecular and Cellular Cardiology, 45(4):567 – 581, 2008. ISSN 0022-2828. doi:http://dx.doi.org/10.1016/j.yjmcc.2008.03.009.URL http://www.sciencedirect.com/science/article/pii/S0022282808003532. Special Issue:Stem Cells And Regenerative Medicine. → pages 51[67] S. Saha, L. Ji, J. J. de Pablo, and S. P. Palecek. Inhibition of human embryonic stem celldifferentiation by mechanical strain. J Cell Physiol, 206(1):126–137, Jan 2006.doi:10.1002/jcp.20441. URL http://dx.doi.org/10.1002/jcp.20441. → pages 12[68] S. Saha, L. Ji, J. J. de Pablo, and S. P. Palecek. Tgf beta/activin/nodal pathway in inhibitionof human embryonic stem cell differentiation by mechanical strain. Biophysical Journal, 94(10):4123–4133, May 2008. doi:10.1529/biophysj.107.119891. → pages 12[69] J. Satin, I. Kehat, O. Caspi, I. Huber, G. Arbel, I. Itzhaki, J. Magyar, E. A. Schroder,I. Perlman, and L. Gepstein. Mechanism of spontaneous excitability in human embryonicstem cell derived cardiomyocytes. J Physiol, 559(Pt 2):479–496, Sep 2004.doi:10.1113/jphysiol.2004.068213. URL http://dx.doi.org/10.1113/jphysiol.2004.068213. →pages 5060[70] Y. Satou, K. S. Imai, and N. Satoh. The ascidian mesp gene specifies heart precursor cells.Development, 131(11):2533–2541, Jun 2004. doi:10.1242/dev.01145. URLhttp://dx.doi.org/10.1242/dev.01145. → pages 7[71] G. C. Schoenwolf, S. B. Bleyl, P. R. Brauer, P. H. Francis-West, and W. J. Larsen. Larsen’shuman embryology, 2015. ISSN 9781455706846 1455706841. URLhttps://www.clinicalkey.com/dura/browse/bookChapter/3-s2.0-C20100689383. → pages 1[72] M. S. Simo˜es-Costa, M. Vasconcelos, A. C. Sampaio, R. M. Cravo, V. L. Linhares,T. Hochgreb, C. Y. I. Yan, B. Davidson, and J. Xavier-Neto. The evolutionary origin ofcardiac chambers. Dev Biol, 277(1):1–15, Jan 2005. doi:10.1016/j.ydbio.2004.09.026. URLhttp://dx.doi.org/10.1016/j.ydbio.2004.09.026. → pages 7[73] K. Song, Y.-J. Nam, X. Luo, X. Qi, W. Tan, G. N. Huang, A. Acharya, C. L. Smith, M. D.Tallquist, E. G. Neilson, J. A. Hill, R. Bassel-Duby, and E. N. Olson. Heart repair byreprogramming non-myocytes with cardiac transcription factors. Nature, 485(7400):599–604,May 2012. ISSN 0028-0836. URL http://dx.doi.org/10.1038/nature11139. → pages 9, 12, 51[74] D. Srivastava and E. C. Berry. Cardiac reprogramming: from mouse toward man. CurrentOpinion In Genetics & Development, 23(5):574–578, Oct. 2013.doi:10.1016/j.gde.2013.07.007. → pages 10, 12[75] D. Srivastava and E. N. Olson. A genetic blueprint for cardiac development. Nature, 407(6801):221–226, Sep 2000. doi:10.1038/35025190. URLhttp://dx.doi.org/10.1038/35025190. → pages 7[76] D. Srivastava, T. Thomas, Q. Lin, M. L. Kirby, D. Brown, and E. N. Olson. Regulation ofcardiac mesodermal and neural crest development by the bhlh transcription factor, dhand. NatGenet, 16(2):154–160, Jun 1997. doi:10.1038/ng0697-154. URLhttp://dx.doi.org/10.1038/ng0697-154. → pages 7[77] K. Takahashi and S. Yamanaka. Induction of pluripotent stem cells from mouse embryonicand adult fibroblast cultures by defined factors. Cell, 126(4):663–676, Aug 2006.doi:10.1016/j.cell.2006.07.024. URL http://dx.doi.org/10.1016/j.cell.2006.07.024. → pages8, 9[78] K. Takahashi, K. Tanabe, M. Ohnuki, M. Narita, T. Ichisaka, K. Tomoda, and S. Yamanaka.Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131(5):861–872, Nov. 2007. ISSN 0092-8674. URLhttp://www.sciencedirect.com/science/article/pii/S0092867407014717. → pages 8[79] T. Teramura, T. Takehara, Y. Onodera, K. Nakagawa, C. Hamanishi, and K. Fukuda.Mechanical stimulation of cyclic tensile strain induces reduction of pluripotent related geneexpressions via activation of rho/rock and subsequent decreasing of akt phosphorylation inhuman induced pluripotent stem cells. Biochemical and Biophysical ResearchCommunications, 417(2):836–841, Jan. 2012. doi:10.1016/j.bbrc.2011.12.052. → pages 12[80] R. W. Tsika, J. J. Bahl, L. A. Leinwand, and E. Morkin. Thyroid hormone regulatesexpression of a transfected human alpha-myosin heavy-chain fusion gene in fetal rat heartcells. Proc Natl Acad Sci U S A, 87(1):379–383, Jan 1990. → pages 38, 40, 4861[81] N. L. Tulloch, V. Muskheli, M. V. Razumova, F. S. Korte, M. Regnier, K. D. Hauch,L. Pabon, H. Reinecke, and C. E. Murry. Growth of engineered human myocardium withmechanical loading and vascular coculture. Circ Res, 109(1):47–59, Jun 2011.doi:10.1161/CIRCRESAHA.110.237206. URLhttp://dx.doi.org/10.1161/CIRCRESAHA.110.237206. → pages 14[82] H. Uosaki, H. Fukushima, A. Takeuchi, S. Matsuoka, N. Nakatsuji, S. Yamanaka, and J. K.Yamashita. Efficient and scalable purification of cardiomyocytes from human embryonic andinduced pluripotent stem cells by vcam1 surface expression. PLoS ONE, 6(8):e23657, 082011. doi:10.1371/journal.pone.0023657. URLhttp://dx.doi.org/10.1371%2Fjournal.pone.0023657. → pages 9[83] R. Wada, N. Muraoka, K. Inagawa, H. Yamakawa, K. Miyamoto, T. Sadahiro, T. Umei,R. Kaneda, T. Suzuki, K. Kamiya, S. Tohyama, S. Yuasa, K. Kokaji, R. Aeba, R. Yozu,H. Yamagishi, T. Kitamura, K. Fukuda, and M. Ieda. Induction of human cardiomyocyte-likecells from fibroblasts by defined factors. Proc Natl Acad Sci U S A, 2013.doi:10.1073/pnas.1304053110. URLhttp://www.pnas.org/content/early/2013/07/16/1304053110.abstract. → pages 10[84] S. Wang, A. Sun, L. Li, G. Zhao, J. Jia, K. Wang, J. Ge, and Y. Zou. Up-regulation of bmp-2antagonizes tgf-β1/rock-enhanced cardiac fibrotic signalling through activation ofsmurf1/smad6 complex. J Cell Mol Med, 16(10):2301–2310, Oct 2012.doi:10.1111/j.1582-4934.2012.01538.x. URLhttp://dx.doi.org/10.1111/j.1582-4934.2012.01538.x. → pages 11[85] K. Watanabe, M. Ueno, D. Kamiya, A. Nishiyama, M. Matsumura, T. Wataya, J. B.Takahashi, S. Nishikawa, S.-i. Nishikawa, K. Muguruma, and Y. Sasai. A rock inhibitorpermits survival of dissociated human embryonic stem cells. Nat Biotechnol, 25(6):681–686,Jun 2007. doi:10.1038/nbt1310. URL http://dx.doi.org/10.1038/nbt1310. → pages 17[86] C. Wu and J. C. Alwine. Secondary structure as a functional feature in the downstream regionof mammalian polyadenylation signals. Molecular and Cellular Biology, 24(7):2789–2796,2004. doi:10.1128/MCB.24.7.2789-2796.2004. URLhttp://mcb.asm.org/content/24/7/2789.abstract. → pages 19[87] D. Yelon, B. Ticho, M. E. Halpern, I. Ruvinsky, R. K. Ho, L. M. Silver, and D. Y. Stainier.The bhlh transcription factor hand2 plays parallel roles in zebrafish heart and pectoral findevelopment. Development, 127(12):2573–2582, Jun 2000. → pages 7[88] J. Yu, M. A. Vodyanik, K. Smuga-Otto, J. Antosiewicz-Bourget, J. L. Frane, S. Tian, J. Nie,G. A. Jonsdottir, V. Ruotti, R. Stewart, I. I. Slukvin, and J. A. Thomson. Induced pluripotentstem cell lines derived from human somatic cells. Science, 318(5858):1917–1920, 2007.doi:10.1126/science.1151526. URLhttp://www.sciencemag.org/content/318/5858/1917.abstract. → pages 16[89] W.-H. Zimmermann, I. Melnychenko, G. Wasmeier, M. Didie, H. Naito, U. Nixdorff,A. Hess, L. Budinsky, K. Brune, B. Michaelis, S. Dhein, A. Schwoerer, H. Ehmke, andT. Eschenhagen. Engineered heart tissue grafts improve systolic and diastolic function in62infarcted rat hearts. Nat Med, 12(4):452–458, Apr. 2006. ISSN 1078-8956. URLhttp://dx.doi.org/10.1038/nm1394. → pages 51[90] R. Zufferey, D. Nagy, R. J. Mandel, L. Naldini, and D. Trono. Multiply attenuated lentiviralvector achieves efficient gene delivery in vivo. Nat Biotechnol, 15(9):871–875, Sep 1997.doi:10.1038/nbt0997-871. URL http://dx.doi.org/10.1038/nbt0997-871. → pages 22[91] R. Zufferey, T. Dull, R. J. Mandel, A. Bukovsky, D. Quiroz, L. Naldini, and D. Trono.Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol, 72(12):9873–9880, Dec 1998. → pages 2263Appendix ADesign and fabricationComputer aided design (CAD) was done with NX (Siemens). Prototypes were fabricated in the UBCDepartment of Mechanical Engineering Student Machine Shop. An iterative design and prototypingprocess was followed, with rapid prototyping facilitated by the use of the Asiga Pico 3D printer andthe OMAX R© MAXIEM abrasive waterjet cutter. The NX modelling module was used to createparts for 3D printing. Files were exported to a stereolithography format (STL) compatible with theAsiga Pico 3D printer. The NX sheet metal module was used to create parts for waterjet cutting.Sheet metal flat pattern files were exported to a CAD format compatible with the Intelli-MAXLayout software (DXF). The DXF files were used to create tool paths for the waterjet cutter inOMAX Layout. Cut quality was set to 3 (medium) and tabs were created after using the auto-pathtool to hold work pieces in place. The tool paths were saved as ORD files and transferred to theOMAX MAXIEM waterjet cutter. Device components were waterjet cut out of gauge 14 and gauge6 aluminum sheet metal (6061-T6). Bending of the sheet metal was done with a box-and-pan brake.The toggle clamp of the HexCycler was purchased from McMaster-Carr (Catalog #5128A39).Open-source CAD files for the design of the HexCycler are available at https://github.com/paradeofwolves/HexCycler. Drawings for the components of the HexCycler are included in Sec-tion A.1.A.1 Drawings of HexCycler components64656667686970717273Appendix BTissue culture protocolsB.1 iPSC maintenance and differentiation protocolsWe used a protocol for culturing pluripotent stem cells that does not rely on a supporting layer ofnon-pluripotent feeder cells. The lack of a feeder layer necessitates a layer of ECM proteins tofacilitate the attachment of the iPSCs. We used a thin coating of Matrigel as this protein coating.Matrigel is frozen below -20◦C, liquid at 4◦C, and gels rapidly at room temperature. Thus Matrigelshould be kept frozen until ready for aliquoting and should be kept on ice during aliquoting.B.1.1 Aliquoting Matrigel1. One day before aliquoting, place 4 centrifuge racks and 1 sterile container of 1.5mL tubes inthe -80◦C freezer.2. Calculate the volume of Matrigel needed per tube. The recommended coating density is0.5mg per 6 well plate, so aliquots of 0.25mg, 0.5mg, and 1mg should be made. Matrigelconcentration is measured in mg/mL and varies by lot.3. Based on the volumes calculated above, place unopened box(es) of the appropriate sizedpipette tips in the -20◦C freezer.4. Place the bottle of Matrigel in a covered container filled with ice in a 4◦C refrigerator overnight.Make sure that the neck of the bottle is not submerged in ice.5. Fill two small containers with ice and place both in the sterile biosafety cabinet. One containerwill hold the Matrigel bottle, while the other will hold the box of pipette tips.6. Open the metal seal on the Matrigel bottle using the sharp end of a scoopula. Carefullyremove the rubber cover of the bottle and place the bottle on ice.747. Retrieve one centrifuge tube rack from the freezer, place in the biosafety cabinet, and fill withmicrotubes.8. Aliquot Matrigel into each tube according to the volumes calculated above. Switch pipettetips every 5 tubes to keep the Matrigel cold and ensure sterility.9. When the centrifuge tube rack is filled, label and transfer tubes to the -20◦C freezer andretrieve a new rack. Work quickly to prevent the Matrigel from congealing.B.1.2 Plating MatrigelMatrigel cannot be thawed and refrozen, so each aliquot must be used at one time. However, excessMatrigel may be plated and used within 7 days.1. Place a sterile 15mL conical tube, one P1000 pipetman, and a cold, sterile bottle of DMEM/F-12 media into the sterile biosafety cabinet.2. Remove one aliquot of Matrigel from the freezer and add 1mL of DMEM/F-12 to the tube.3. Gently pipette up and down to thaw and dissolve the Matrigel. Transfer to 15mL conical tube.4. Add additional DMEM/F-12 to bring the total volume to 1mL for each well to be coated.Plate 1mL per well.5. Tilt the well in orthogonal directions to fully cover the surface and allow to set for 1 hour atroom temperature in the biosafety cabinet.6. Wells may be used after the 1-hour coating, or stored for later use. Wells may be stored byadding 1mL of DMEM/F-12 to each well and placing in a 37◦C incubator, or wrapping inparafilm and storing in the 4◦C refrigerator. Stored wells must be used within 7 days.7. If using plates that have been stored at 4◦C, warm to room temperature in the biosafety cabinetfor 1 hour before introducing cells.B.1.3 Preparing complete mTeSR1 mediumThe mTeSR1 medium is shipped in two components: basal medium and 5X supplement. Thecomplete medium is made by adding the 5X supplement to the basal medium. The 5X supplementcontains contains recombinant proteins that degrade rapidly at high temperatures and with repeatedfreeze-thaw cycles. For this reason, the supplement and the complete medium are kept frozen untiluse, and should not be warmed in a 37◦C water bath. Avoid rough agitation of the medium orsupplement that may form bubbles. The complete medium may be stored in the 4◦C refrigerator forup to 2 weeks, or frozen at -20◦C for up to 6 months.75Figure B.1: Chemical structure of the ROCK inhibitor Y-27632.1. Thaw the 5X supplement overnight in the 4◦C refrigerator and mix thoroughly by gentlyinverting.2. In a biosafety cabinet, add 100mL of the 5X supplement to 400mL of the basal medium. Mixthoroughly by gently inverting.3. Aliquot the complete medium into sterile 50mL conical tubes and store in the -20◦C freezer.4. When needed, thaw an aliquot of the complete medium at room temperature or overnight at4◦C.5. Unused medium can be stored at 4◦C for up to 2 weeks, but should not be refrozen.B.1.4 Preparing ROCK inhibitor stock solutionWe used the small-molecule ROCK inhibitor Y-27632 (Figure B.1) to decrease anoikis and increasecell viability after dissociation.1. Spin down the vial of Y-27632 powder at 10,000 rpm for 1 minute.2. Place a P1000 pipetman in the biosafety cabinet.3. Dissolve the Y-27632 powder in 624µL of PBS to create a 5mmol L−1 stock solution.4. Aliquot the stock solution to 3 microtubes of 208µL each.5. Store the aliquots at -20◦C. The solution is stable for up to 6 months.B.1.5 Thawing iPSCsWear eye protection when handling vials that may have been submerged in liquid nitrogen, as thesevials may explode when warmed.761. Place a sterile 15mL conical tube, one 6-well plate coated with Matrigel, one P1000 pipetman,an aliquot of ROCK inhibitor, and mTeSR1 medium warmed to room temperature in thebiosafety cabinet.2. Add 1.5mL of mTeSR1 medium to each well to be plated with cells.3. Replace the aspiration Pasteur pipette with a new sterile pipette.4. Remove the vial of iPSCs from the liquid nitrogen dewar and roll it between your glovedhands for 15 seconds to remove frost.5. Immerse the vial in a 37◦C water bath without submerging the cap, swirling gently.6. When only an ice crystal remains, remove the vial from the bath.7. Ensure the cap is tight and immerse the vial into a 95% ethanol bath to sterilize the outside ofthe tube.8. Air-dry the tube in the biosafety cabinet for 30 seconds.9. Transfer the cells to the sterile 15mL conical tube using the P1000 pipetman.10. Add 9mL of complete mTeSR1 medium drop-wise to the cells, gently rocking the tube whiledoing so to mix the cells and reduce osmotic shock.11. Centrifuge the cells at 200 x g for 5 minutes.12. Aspirate and discard the supernatant.13. Using a 5mL pipette, re-suspend the cell pellet in 0.5mL mTeSR1 medium for every well thatwill receive cells.14. Gently pipette cells up and down in the tube 3 times.15. Slowly add 0.5mL of the cell suspension drop-wise into each well.16. Add 4µL of 5mmol L−1 ROCK inhibitor stock solution to each well to reach a final concen-tration of 10µmol L−1.17. Label the plate with the cell line, the passage number from the vial, the date, and your initials.18. Place the plate in the incubator and gently rock the plate in orthogonal directions to evenlydistribute the cells. Avoid circular motions to prevent the cells from pooling in the centre ofthe well.77Figure B.2: The appearance of iPSC colonies 1 day after plating (left) and immediately priorto passaging (right).B.1.6 Feeding iPSCsiPSC medium must be renewed daily until the cells are ready for freezing, passaging, or experimen-tation.1. Observe the pluripotent stem cells using a microscope. If they require passaging, follow theprocedures in Section B.1.7.2. Warm a tube of mTeSR1 medium to room temperature in the biosafety cabinet for 15 minutes.3. Replace the aspiration Pasteur pipette with a new sterile pipette.4. Aspirate the spent medium with the sterile Pasteur pipette and a P200 pipette tip. Use adifferent pipette tip for each well to reduce the risk of contamination.5. Add 2mL of fresh mTeSR1 medium to each well. Pipette against the side wall of the culturewell to reduce fluid shear forces on the cells.6. Return the plate to the 37◦C incubator.B.1.7 Passaging iPSCsiPSCs are ready for passaging when continguous colonies of cells reach a size of greater than 1mmin its widest dimension, as shown in Figure B.2. Passaging is also required when cells show vis-ible morphological changes indicative of differentiation. We use EDTA (Versene, 0.02%) for forpassaging, which is a gentle method that gives clusters of cell aggregate instead of fully dissociatedcells. The passage ratio can vary between 1:8 and 1:20, depending on the density of cell coloniesafter plating.1. Warm a tube of mTeSR1 medium to room temperature in the biosafety cabinet for 15 minutes.782. Place a P10 pipetman, EDTA solution, and an aliquot of ROCK inhibitor in the biosafetycabinet. A P200 pipetman and 15mL sterile conical tube may also be needed if the split ratiois high.3. Replace the aspiration Pasteur pipette with a new sterile pipette.4. Aspirate the Matrigel plating medium from each well to be seeded with new cells5. Add 2mL of mTeSR1 medium to each well.6. Label the new plate with the cell line name, the new passage number, the date, split ratio, andyour initials.7. If harvesting from more than one well, stagger treatment with EDTA to avoid overexposure.8. Aspirate the spent medium with the sterile Pasteur pipette and a P200 pipette tip. Use adifferent pipette tip for each well to reduce the risk of contamination.9. Rinse each well with 1mL room temperature EDTA and aspirate.10. Treat each well with 1mL room temperature EDTA for 7–9 minutes.11. Aspirate the EDTA carefully, without disturbing the attached cell layer. If cells become free-floating, collect and spin down, remove EDTA, and resuspend with mTeSR1.12. Using 3mL of mTeSR1 medium per well, hold a 5mL pipette perpendicular to the plate andgently wash the cells off the plate. Repeat if necessary, but do not use more than 3mL ofmedium per well to avoid contamination. Pipette gently to avoid bubbles, and avoid touchingthe cells with the tip of the pipette.13. If harvesting from more than one well, use the same volume of medium to remove cells.14. If the split ratio is high and the volume of cell suspension is small, pool the cell suspension ina sterile conical plate and use a pipetman for the next step. Otherwise, use a 5mL pipette.15. Add the appropriate volume of cell suspension drop-wise to each well of the new plate.16. Add ROCK inhibitor to each well to reach a final concentration of 10µmol L−1.17. Place the plate in the incubator and gently rock the plate in orthogonal directions to evenlydistribute the cells. Avoid circular motions to prevent the cells from pooling in the centre ofthe well.18. While cells are attaching, limit opening and closing of the incubator. If access is required,open and close the door gently.79B.1.8 Freezing iPSCs1. Obtain a room-temperature isopropanol freezing container; the isopropanol must be replacedevery 5 uses.2. In the biosafety cabinet, label cryovials with the cell line, passage number (1 more than thenumber on the plate), freeze date, and your initials. Use a pen or labels that resist liquidnitrogen and ethanol.3. Thaw the mFreSR1 cryopreservation medium in the biosafety cabinet.4. Place a P1000 pipetman, 50mL sterile conical tube, and EDTA solution in the biosafety cabi-net.5. Replace the aspiration Pasteur pipette with a new sterile pipette.6. If freezing more than one plate, stagger EDTA treatment to avoid overexposure.7. Aspirate the spent medium with the sterile Pasteur pipette and a P200 pipette tip. Use adifferent pipette tip for each well to reduce the risk of contamination.8. Rinse each well with 1mL room temperature EDTA and aspirate.9. Treat each well with 1mL room temperature EDTA for 7–9 minutes.10. Aspirate the EDTA carefully, without disturbing the attached cell layer.11. Gently wash cells off using 3mL of mFreSR1 medium for each plate, transferring the mediumfrom well to well.12. Pool the cell suspension in the 50mL sterile conical tube.13. Repeat harvest for any remaining plates and pool all the cells to create a uniform lot.14. Add mFreSR1 medium to reach desired freezing density. One 6-well plate can be frozen into10 cryovials, using a total of 10mL of mFreSR1 medium.15. Use a 5mL pipette to gently mix the cell suspension.16. Using the same pipette, aliquot 1mL of cell suspension to each prepared cryovial. Mix thepooled cells every 5 vials to maintain even distribution.17. Place the cryovials into the isopropanol freezing container and place the container in the-80◦C freezer overnight.18. The next day, transfer the cryovials to liquid nitrogen storage.80B.1.9 Preparing Activin A and BMP4 stock solutionsRecombinant human Activin A and BMP4 were purchased from R&D Systems and are shipped inlyophilized powder form.1. Create 5mL of sterile HCl solution and 5mL of sterile HCl solution containing 0.1% bovineserum albumin (BSA). Sterilize the solutions using a 0.2µm filter.2. Dissolve 10µg of lyophilized Activin A in 100µL of sterile 4mmol L−1 HCl solution to createa 100µg mL−1 stock solution.3. Make 10µL aliquots of the Activin A stock solution. This solution is stable for up to 3 monthsat -20◦C.4. Dissolve 10µg of lyophilized BMP4 in 200µL of sterile 4mmol L−1 HCl solution containing0.1% BSA to create a 50µg mL−1 stock solution.5. Make 5µL aliquots of the BMP4 stock solution. This solution is stable for up to 3 months at-20◦C.6. Dilute the 50µg mL−1 stock BMP4 solution using sterile 4mmol L−1 HCl solution containing0.1% BSA to create a 5µg mL−1 working solution.B.1.10 Preparation of B-27–supplemented mediaThe differentiation of iPSCs is conducted in RMPI-1640 medium containing B-27 complete sup-plement, and B-27 supplement without insulin. Section 2.1.5 has details of when each supplementis required. B-27 is shipped frozen at a 50X concentration and should protected from light.1. Thaw B-27 supplements at 4◦C.2. Make 400µL and 100µL aliquots of B-27.3. Label the microtubes with the type of supplement (B-27 with or without insulin), the volume,and the date.4. When needed, thaw an aliquot of B-27 in the biosafety cabinet. The 100µL aliquot makes5mL of complete medium, while the 400µL aliquot makes 20mL of complete medium.5. B-27 supplement should not be freeze-thawed more than twice.6. Cover a sterile conical tube with aluminum foil and label with the medium and supplementname, the date, and your initials. Place the tube in the biosafety cabinet.7. Pipette 5mL or 20mL of RPMI-1640 into the covered conical tube.818. Using a P1000 or P200 pipetman, transfer the B-27 supplement to the RPMI-1640 medium.9. Gently invert the tube to mix, taking care not to form bubbles.10. Complete RPMI-1640 medium with B-27 supplements is stable at 4◦C for up to 7 days.11. Prior to addition to cells, the complete RPMI-1640 medium with B-27 supplements shouldbe warmed to room temperature, not in a 37◦C water bath.82


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items