engineering three-dimensional stem cell morphogenesis for the development of tissue models and...

Engineering Three-Dimensional Stem Cell Morphogenesis

for the Development of Tissue Models and Scalable

Regenerative Therapeutics

MELISSA A. KINNEY,1 TRACY A. HOOKWAY,1 YUN WANG,1 and TODD C. MCDEVITT1,2

1The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology & Emory University, 313Ferst Drive, Atlanta, GA 30332-0532, USA; and 2The Parker H. Petit Institute for Bioengineering and Bioscience, Georgia

Institute of Technology, Atlanta, GA, USA

(Received 30 September 2013; accepted 21 November 2013)

Associate Editor Robert Nerem oversaw the review of this article.

Abstract—The physiochemical stem cell microenvironmentregulates the delicate balance between self-renewal anddifferentiation. The three-dimensional assembly of stem cellsfacilitates cellular interactions that promote morphogenesis,analogous to the multicellular, heterotypic tissue organiza-tion that accompanies embryogenesis. Therefore, expansionand differentiation of stem cells as multicellular aggregatesprovides a controlled platform for studying the biologicaland engineering principles underlying spatiotemporal mor-phogenesis and tissue patterning. Moreover, three-dimen-sional stem cell cultures are amenable to translationalscreening applications and therapies, which underscores thebroad utility of scalable suspension cultures across labora-tory and clinical scales. In this review, we discuss stem cellmorphogenesis in the context of fundamental biophysicalprinciples, including the three-dimensional modulation ofadhesions, mechanics, and molecular transport and highlightthe opportunities to employ stem cell spheroids for tissuemodeling, bioprocessing, and regenerative therapies.

Keywords—Stem cells, Organoid, Intercellular adhesions,

Biophysical, Molecular transport, Regenerative medicine,

Tissue engineering.

INTRODUCTION

The balance between stem cell proliferation and dif-ferentiation is tightly controlled by local cues present inthe stem cell niche microenvironment.111,137 In responseto chemical or physical perturbations, cells exit the nicheand undergo differentiation processes,102 often to medi-ate regenerationor repair in pathological contexts suchas

hemogenic repopulation92 or wound healing.156 Oneparticularly dynamic example of stem cell microenvi-ronment regulation occurs within the blastocyst-stageembryo, whereby a compact cluster of cells, known as theinner cellmass (ICM),develop intoall somatic tissues andorgans.61 During the early stages of pre-implantationdevelopment, the cells of the ICM undergo sequentialspecification, through which cells commit along thethree germ lineages—endoderm, ectoderm, and meso-derm—and continue to make cell fate decisions in aspatially and temporally controlled manner, therebyproviding a robust model bywhich to study cell plasticityand tissue formation. The patterning of cell fates ismediated by physical processes, such as proliferation62

and migration,56 which occur concomitant with bio-chemical gradients,47 thereby highlighting the need fornovel technologies to recapitulate the multiparametricstimuli present within the tissue microenvironment. Forexample, during gastrulation, the prospective mesodermcells undergo a dynamic epithelial-to-mesenchymaltransition (EMT) and migrate through the primitivestreak.18,31Similarly, collective cellmigrationof epithelialsheets has been implicated in processes such as branchingmorphogenesis.50 Biophysical signals mediating thespatiotemporal dynamics of cell migration mediate theformation of functionally and structurally distinct, yetadjacent, tissue structures, such as heart, lungs and kid-ney, each of which is defined by precisely controlled,heterotypic multicellular organization. The precise pre-sentation of biochemical and biophysical cues in vivomotivates the development of engineering approachesthat recapitulate the stem cell niche in order to createfunctional heterotypic multicellular structures in vitrowhich are amenable to the replacement of damaged or

Address correspondence to Todd C. McDevitt, The Wallace H.

Coulter Department of Biomedical Engineering, Georgia Institute of

Technology&EmoryUniversity, 313FerstDrive,Atlanta,GA30332-

0532, USA. Electronic mail: [email protected]

Annals of Biomedical Engineering (� 2013)

DOI: 10.1007/s10439-013-0953-9

� 2013 Biomedical Engineering Society

diseased tissue through scalable bioprocessing and tissueengineering approaches, and offer new cellular platformsfor high-throughput pharmaceutical screening and drugdevelopment.

In order to emulate in vivo tissue-scale morphogenicprocesses, in vitro platforms have been developed topresent chemical and physical cues in three-dimensionalconfigurations, analogous to themulticellular structureofnative tissues. Early studies of pluripotent embryonalcarcinoma cells created high-density cellular environ-ments in vitro112 byE-cadherin-mediated self-assembly ofcells,98 which alters the microenvironment through local3D presentation of adhesive and biochemical signals,analogous to aspects of embryonic development. Similarapproaches have been adapted to initiate differentiationof pluripotent stem cells (PSCs) and induced pluripotentstem cells (iPS) via 3D aggregation as embryoid bodies(EBs), either through forced aggregation in small mediavolumes (hanging drop, microwell centrifugation) orthrough self-assembly via random association in bulksuspension cultures.94 More recently, the aggregation ofmultipotent stem and progenitor cells, such as mesen-chymal stemcells (MSCs)7 andneural stemcells (NSCs),14

has alsobeen explored, inorder to provide physiologicallyrelevant cell–cell and cell–ECM adhesions and to induceparacrine factor secretion, which together are hypothe-sized to support tissue maturation. Together, engineeringapproaches to systematically control88 and perturb three-dimensional scaffold-free stem cell-derived microtissuesenable new routes to study the expansion and differenti-ation of stem cells and have increasingly become moreaccessible and broadly applicable in the fields of stem cellbiology and tissue engineering.

Pluripotent stem cells, in particular, exhibit strikingexamples of morphogenesis upon three-dimensionalassembly and differentiation,134 thereby providing anintriguing model system amenable to studying andperturbing physiochemical elements mediating cellfate. For example, induction of Rx + neuroepitheliumin 3D PSC spheroids resulted in spatially distinctpatterns of differentiation, with the resulting neo-tis-sues exhibiting phenotypic markers and architecturereminiscent of the native optic cup.46,115 Interestingly,the dynamic structural changes, including evaginationof epithelial vesicles and subsequent invagination,yielded distinct layers of neural retinal and retinalpigment epithelial (RPE) cells that were directed solelyby cell-intrinsic morphogenic processes and self-orga-nization of cells in 3D, thereby highlighting the abilityof stem cell-derived microtissues to recapitulate aspectsof tissue development in vivo.46 Similarly, simple bio-chemical induction of differentiation directed themorphogenesis of 3D microtissues comprised of or-ganized enterocytes, goblet, Paneth and enteroendo-crine cells, thereby establishing an in vitro organoid

model of intestinal structure and function.149 Anothermodel exhibiting self-formation of complex cerebralstructures97 was developed to study the pathogenesis ofhuman microcephaly using iPS cells. Moreover, similarapproaches have yielded functional anterior pitui-tary,151 thyroid,4 and hepatic,154 structures which ex-hibit secretory functions when transplanted in vivo.Together, 3D aggregation of stem cells providesopportunities to study tissue morphogenesis (Fig. 1)across a wide array of cell phenotypes and microtissuestructures. In this review, we discuss the biophysicalbasis and engineering opportunities to control andperturb 3D PSC morphogenesis, as well as thepotential applications of PSC-derived microtissues asvaluable tools for developmental biology, drug dis-covery and regenerative medicine.

3D STEM CELL MORPHOGENIC PROCESSES

Intercellular Adhesion Organization in 3D

The formation and morphogenesis of three-dimen-sional cellular aggregates is a dynamic process regu-lated by differential cellular adhesions, matrixsynthesis, and remodeling. In contrast to the cell–ma-trix adhesions that mediate cell seeding on scaffolds orwithin hydrogels, the rapid formation of 3D aggregates(12–24 h) is primarily directed by cellular adhesionmolecules (CAMs) such as cadherins and connexins.6

CAMs are also key regulators of self-renewal and theinduction or maintenance of pluripotency27,148 withpluripotent cells being sensitive to disruption of cell–cell contacts and dependent on CAM-mediated inter-actions for survival. Therefore, the central role ofCAMs in both regulating pluripotency and 3D aggre-gate formation highlights opportunities to study thekinetics and mechanisms of CAM biology in regulatingstem cell morphogenesis.

Owing to the central role of CAMs in mediatingthe rapid transition from single cells to neo-tissues,spheroid cultures offer unique opportunities to studythe adhesive signatures characteristic of the earliestcell fate decisions and developmental morphogenesis.CAM expression within pluripotent aggregates isaltered temporally during differentiation, withundifferentiated PSCs highly expressing E-cadherin.In contrast, E-cadherin expression is often decreasedupon differentiation, concomitant with differentialexpression of other CAMs during lineage commit-ment. For example, Stankovich et al.150 identified theCAM profile (i.e., E-cadherin, Cldn4, ZO-1, ZO-2,Cx43) related to hematopoietic and endothelial cellsubpopulations. Further, modulation of CAMs, viaknock-down of E-cadherin and connexin 43 within

KINNEY et al.

PSCs led to decreased efficiency of hematopoieticdifferentiation,150 indicating that intercellular adhe-sions play a critical role in regulating stem cell dif-ferentiation.

In addition to physically connecting adjacent cells,intercellular adhesions also influence local cell signal-ing. Consistent with the loss of E-cadherin duringprimitive streak migration in embryonic develop-ment,18 cadherin-mediated assembly of PSCs alters thetemporal expression of E-cadherin, and influences thedownstream signaling within the associated Wnt/b-catenin pathway.90 Simply altering the kinetics ofaggregate assembly, therefore, results in modulation ofWnt-dependent mesoderm differentiation, with preciseregulation of Wnt signaling required for cardiac dif-ferentiation.162 Similarly, hanging drop formation ofstem cell aggregates also modulates the extent of car-diomyocyte differentiation, likely due to similarmechanisms downstream of intercellular adhe-sions.132,178 In addition, controlling the cellular com-position (cells per aggregate) also modulates cardiacdifferentiation efficacy, which is thought to arise due tochanges in the surface area-to-volume ratio, which

alters the cell polarity and distribution of CAMs.12

Together, these examples illustrate the interrelation-ship between CAMs and developmentally-relevantsignaling, which can be manipulated to engineer stemcell microenvironments by controlling cell aggregationand spheroid formation.

Extracellular Matrix Deposition and Remodelingin 3D Stem Cell Spheroids

In addition to intercellular adhesions, stem cells arealso in contact with the extracellular matrix (ECM), acomplex three-dimensional network of macromole-cules that provides distinct biomechanical, biophysical,and biochemical cues.36 Production of the ECM is adynamic cyclical process comprised of matrix synthe-sis, deposition, organization, degradation and remod-eling39; if not properly regulated, alterations in ECMprocesses contribute to disease or pathologicalstates.34,120 ECM synthesis and assembly are regulatedthrough bi-directional signaling between the cells andthe underlying matrix via integrins.173 Concomitantly,physical signals transmitted via integrins affect the

FIGURE 1. Physical parameters influenced by 3D assembly of stem cells. Multicellular assembly of stem cells alters the cell–celland cell–matrix associations, the mechanical tension resulting from cytoskeletal rearrangements, and the distribution of exoge-nous and endogenous biochemical factors.


intercellular actomyosin network, which inducesROCK-mediated proliferation, ultimately leading tofibronectin fiber assembly and cleft stabilization.38

These examples demonstrate how extracellular cuesinitiated through cell–ECM interactions influenceintracellular signaling that mediate morphogenic pro-cesses.

During early stages of development, ECM is spa-tiotemporally regulated to ensure proper induction ofdevelopmental processes. For example, branchingmorphogenesis of the submandibular gland requiresthe presence of laminin a5 to stimulate integrin b1-mediated MAPK signaling induced proliferation andbud formation.126 Similarly, as stem cells differentiate,the ECM composition changes,113 leading to distinctprofiles throughout different stages of tissue morpho-genesis. The loss of polarity and cell–cell adhesionsmediates cell migration73 characteristic of EMT viaECM molecules such as glycosaminoglycans(GAGs)158 and proteoglycans106 and remodeling of thenative ECM by matrix metalloproteinases (MMPs).119

Three-dimensional culture of stem cells as EBs in vitrorecapitulates aspects of EMT,25 including alterations inECM composition and cellular organization as afunction of differentiation. For example, GAGs suchas hyaluronan and versican are increasingly synthe-sized with EB differentiation and co-localize withinmesenchymal regions ofv the EBs.143 GAGs are knownto sequester and bind growth factors within the ECMto facilitate the local presentation to cells,179 whichreflects the ability of ECM to regulate biochemicalsignals in addition to providing physical cues. Inaddition to GAGs, other fibrillar ECM molecules suchas collagen I and IV, fibronectin, and laminin areobserved throughout EBs63,114,128; while generally inlower abundance within pluripotent aggregates com-pared to mature tissues in vivo, ECM synthesis anddeposition may play an important role in early stemcell morphogenesis. While three-dimensional culture of

PSCs recapitulates many early developmental events,the specific role of ECM in PSC morphogenic pro-cesses remains largely unknown due to the limitedtechniques for achieving spatial and temporal precisionsimilar to developmental processes, as well as thecomplexity associated with studying such multivariateprocesses in three-dimensional multicellular models.

Cellular Mechanotransduction and Stem Cell Phenotype

Three-dimensional remodeling of intercellularadhesions and ECM modulates the intracellulararchitecture, which is responsible for transmittingforces within and between cells.79 The mechanicalstructure of individual cells is often described by the‘‘tensegrity’’ model,78 in which interior compressedelements (i.e., microtubules, microfilaments) are bal-anced by opposing elements (i.e., contractile actincytoskeleton) in tension.52 Cellular forces are alsoopposed through extracellular factors,144 such as ECMmolecules, which impart dynamic feedback to mediatereorganization of the cytoskeleton108 and associatedcell–cell and cell–matrix adhesions.105

Approaches to manipulate cell shape have beenextensively explored in monolayer cultures, as theregulation of inter- and intra-cellular tension mediatesstem cell responses, including proliferation116 and dif-ferentiation.109 Engineering approaches aim to controlstem cell mechanics (Table 1) by changing adhesiveligand coating density49,159 or through manipulation ofsubstrate cross-linking,48 independent of ECM con-centration.58 Downstream signaling of the Rho asso-ciated protein kinase (ROCK) pathway is oftenimplicated in mediating cytoskeletal reorganizationand stiffness-mediated phenotypic changes.109 Forexample, EMT occurs within regions of high cellulartension, and is abrogated by perturbation of theROCK pathway.64 Similarly, MSC differentiationcan also be patterned by controlling the geometry of

TABLE 1. Stem cell response to mechanical cues.

Mechanical signal Method of force delivery

Force sensing

mechanisms Cell fate response(s) Reference(s)

Ligand presentation Ligand density CAMs Liited spreading and increased

differentiation on substrates with

decreased ligand densities

49,159

Substrate stiffness Cross linking density, elastomeric

micropost geometry

CAMs Increased osteogenesis

on stiff substrates, adipogenesis

on soft substrates

48,58

Cellular tension Manipulation of cell shape via

patterning cellular/multicellular

geometry

Cytoskeleton Increased proliferation, EMT and

osteogenesis in response to

increased cellular tension

and spreading

64,109,116,127

Cellular

compression

Surrounding cells Cytoskeleton Decreased proliferation in regions

of increased cellular compression

1,75

KINNEY et al.

multicellular sheets, with increased osteogenesis inregions of high tension.127 In addition, defectivemechanotransduction has been linked to decreases inthe expression of nuclear proteins, such as nuclearlamins,96 which have also been implicated as markersof hPSC differentiation.32 Therefore, in addition toadhesive and cytoskeletal changes, nuclear shape mayalso play a key role in mediating and responding tocellular tension and directing phenotypic changes,37

thereby demonstrating multiparametric effects of theextracellular environment on cellular tension and stemcell fate.

The role of mechanics in mediating three-dimen-sional stem cell differentiation, however, remains lesswell defined. Given the dramatic changes in cellularorganization and adhesive profiles when stem cells areassembled as spheroids,36 the cytoskeletal organiza-tion, and thereby biomechanics, are fundamentallydistinct from properties measured in monolayer cul-tures (Fig. 2). While integrin adhesions and ECMimpart the primary forces opposing cytoskeletal ten-sion in 2D, the cells themselves take on load bearingprocesses in 3D, with the network of cell–cell cadherinadhesions and cytoskeletal associations regulating

cellular tension. Moreover, the orientation and com-position of extracellular and pericellular matrices varyspatially and temporally during differentiation,113 andplay an active mechanical role in mediating changes inadhesive signature or clustering.122 Therefore, due tothe established role of biomechanics in mediating stemcell phenotype, opportunities remain to developmethods to monitor and perturb local tension in 3Daggregates as a novel approach to direct spatial pat-terns of differentiation and morphogenesis.

Multicellular Mechanics of 3D Tissues and Spheroids

Three-dimensional morphogenesis, often associatedwith tumorigenesis and embryonic development, is afundamentally biophysical process,175 as cellularmigration and proliferation occur in conjunction withdynamic rearrangements of adhesions and the cyto-skeleton. Mechanics have been directly implicated inmediating aspects of embryonic gastrulation,82 elon-gation,182 and dorsal and neural tube closure.54

Moreover, the actomyosin cytoskeleton exhibits dis-tinct dynamics during mesoderm migration accompa-nying gastrulation.84

FIGURE 2. Cytoskeletal structure in monolayer and spheroid formats. Pluripotent (mESC, hESC) and multipotent (hMSC) stemcell populations exhibit distinct morphologies when expanded in adherent monolayer cultures, with PSCs exhibiting tightlypacked, rounded colonies and cortical actin structures, in contrast to the prominent stress fibers exhibited by spread, fibroblast-like MSCs. When assembled in 3D, however, MSCs exhibit similar actin structures compared to PSCs, thereby demonstrating thefundamental structural changes upon multicellular assembly. Adherent MSCs in monolayer exhibit isolated spread morphologieswith pronounced stress fibers, whereas high-density 3D aggregates of ESCs and MSCs exhibit cortical actin distributions.Red 5 F-actin staining via phalloidin (AlexaFluor 546; 1:40). Blue 5 Hoechst nuclear counterstain (10 lg/mL). Scale bar 5 100 lm.


The association between morphogenic processesand cell plasticity suggests an active role for biome-chanics in 3D stem cell morphogenesis. The stiffness oftissues has been related to force generation,167 withmesodermal structures, which actively migrate throughthe primitive streak, exhibiting a 10-fold increase instiffness compared to endoderm in embryonicexplants.182 While direct experimental measurementsof local 3D cellular tension are difficult to obtain withcurrent approaches, computational simulations basedupon mechanical models suggest that the homeostaticstate of epithelial tissue structure is maintained bycellular tension, and accurately recapitulate experi-mentally observed cellular rearrangements upon laserablation of individual cells.51 Similar methods haveelucidated the biomechanical basis underlying theregulation of wing size in drosophila embryos1,75;despite the presence of biochemical signals, cell growthis inhibited by cell compression at the center of thedisk, which ultimately dictates the final wing size.Similarly, non-uniform proliferation has been linked togeometrical regulation of mechanical tension incells,142 with proliferation occurring in regions ofincreased local tension.116 Moreover, Rho-mediatedcontractility has also been implicated as a mechanismunderlying the dissociation of adherens junctions,130

thereby demonstrating the interdependence of bio-chemical and biophysical processes in morphogenesis.Together, the mechanical mechanisms mediating pro-liferation and differentiation enable the formation ofdistinct shapes and cellular patterning to direct diver-gent lineages and form complex tissues.

In addition to the synergistic effects mediating cellphenotype, mechanical and biochemical signals aretransmitted across different length scales in tissues.44

For example, in spheroids, tension is transmittedbetween cells via the actomyosin–CAM network.Moreover, the tension on cells at the exterior of tissuesor spheroids differs compared to those residing insidethe mass of cells.107 Such transmission of force acrossand within spheroid structures may account, in part,for some of the differences in cell fate trajectoriesobserved in aggregates of different sizes.11,12 In con-trast, tissues comprised of large quantities of ECM orcells encapsulated in hydrogel materials restrict tensiontransmission, which potentially limits the spatialtransduction of mechanical forces. Therefore, thedynamics of tissue remodeling during morphogenesisalso dictates the local mechanical cues. While PSCs, inparticular, mimic aspects of embryonic development onthe biochemical level, opportunities remain to elucidatethe biomechanical morphogenic processes in 3D stemcell spheroids, in order to draw from analogous in vivoevents to direct cell fate by recapitulating physical cuespresent in embryonic microenvironments.

Diffusive Mass Transport of Exogenous Factors Within3D Aggregates

Inductive cues, including oxygen, nutrients andmorphogens, are important factors governing stem cellfate. Precise control of biochemical delivery withinstem cell cultures is valuable for understanding cellularresponses and critical for establishing reproduciblestem cell ex vivo expansion and differentiation. Intraditional 2D monolayer cultures, exogenous solublefactors or cell-secreted endogenous factors diffusefreely throughout the medium, and thereby reach anequilibrium in which all cells are exposed to similarbiochemical environments.5 In contrast, in 3D aggre-gate cultures, a concentration gradient is establishedbetween the surrounding culture environment and theinterior of the spheroids.165 The distinct cellulardynamics in 2D and 3D stem cell culture136,170 as wellas the heterogeneity within individual EBs most likelyarise, at least in part, due to the aforementioned dis-parity in mass transport between the culture systems128

The mass transport within EBs has been measuredexperimentally128 and modeled mathematically165 as afunction of the EB size (radius), ECM composition,cell packing density and molecular uptake rate.Changes in EB size impact the surface area-to-volumeratio, which is inversely related to the aggregate radius;therefore, EB size is a primary critical determinant ofmass transfer in and out of EBs. The partial pressure ofoxygen in mouse PSC aggregates has been theoreti-cally165 and experimentally170 studied, and confirmthat the concentration is inversely related to aggregatesize, with decreasing concentration from the surfacetoward the center of aggregates. The varying oxygenlevels throughout PSC aggregates may favor differen-tiation toward specific lineages. For example, hypoxicconditions within aggregates enhances VEGF secre-tion, which influences the differentiation of PSCs to-wards hematopoietic lineages.123 Mathematicalmodeling also indicated that hematopoietic capacity isdependent on the size of EBs, with a 6-fold increase inCD45+ cells from larger size of EBs compared tosmaller EBs under the same soluble differentiationcondition.69 Additionally, oxygen gradients may bespatiotemporally modulated during stem cell differen-tiation, as the oxygen uptake rate varies depending onthe cell lineages.2 The concentration gradients of glu-cose and cytokines have also been calculated using theThiele modulus.165 Within hPSC aggregates, the criti-cal limit for glucose concentration was mathematicallydetermined to occur after 74 h in culture,165 which islonger than the conventional time between mediaexchanges (24–48 h). The cytokine concentration waslargely dependent on its depletion rate, with EB size-dependent cytokine gradients occurring at increased

KINNEY et al.

depletion rates.165 In addition, the cellular composition(packing density), also modulates the mass transport,as the cells at the outer edge of EB consume nutrientsor bind growth factors faster than the characteristicdiffusion time, which can significantly reduce the masstransport in EBs with high cell packing density165 orincreased size during long-term culture.147

In addition to EB size, other structural parameters,such as the ECM composition, may also influence themass transport and ultimately affect the differentiationpropensity or efficiency.11,12,19,110,117 A diffusive barriercomprised by the outer layer of differentiating day 7 EBswas first reported by Sachlos and Auguste128; the 20 lmtri-layer shell structure consists of a superficial collagentype I outer layer, a squamous epithelial cell layer and acollagen type IV basement membrane.128 The physicaltransport barrier posed by this structure significantly re-duces molecular diffusion, and is likely more pronouncedfor growth factors of varying molecular weights, therebycreating disparate concentration gradients for differentmolecules.128 Manipulation of the exterior spheroidstructure using enzymes, such as collagenase, has beenexplored as an approach to promote increased transportand demonstrated enhanced efficacy for retinoic acid-mediated neural differentiation.128 Additionally, bioma-terial principles have been employed to successfully con-trol the release kinetics and spatial delivery of exogenousfactors (i.e., retinoic acid, BMP4) within PSC aggregatesin order to overcome diffusion limitations.21,22,123

Endogenous Factor Transport

Pluripotent stem cells endogenously produce manyof the morphogenic growth factors, such as BMPs,Activin and Wnts, that can control primitive cell fatedecisions.41,80 Similar to gradients present duringembryonic development,141 transport limitations canlead to non-uniform distributions of endogenous fac-tors in stem cell aggregates that contribute to mor-phogenesis and spatial patterning of distinctphenotypes.11,86 The size of PSC aggregates has beenlinked to specific differentiation trajectories, as lowergene expression of Wnt5A and enhanced cardiogenesiswas observed in larger aggregates.76 Similarly, a radialgradient of Oct4, Nanog and pSTAT3 was observedwithin PSC colonies in the absence of exogenous ofLIF, suggesting that spatially organized self-renewal iscontrolled by gradients of endogenous Gp130ligands.41 The cell density and spheroid size also affectthe interplay between endogenous stimulatory orinhibitory signals and exogenous factors,123 as localcell density influences the relative concentration ofendogenous factors, leading to enhanced blood pro-genitor differentiation efficiency in 100-cell aggregatescompared to 10-cell aggregates.123

While the aforementioned parameters affect theconcentration profile of endogenous factorsthroughout PSC aggregates, the ECM can alsosequester growth factors, in order to regulate thebioavailability and local presentation to cells, andtherefore contribute to the endogenous moleculargradients within 3D aggregates. ECM proteins suchas fibronectin, vitronectin, collagens, proteogly-cans, and heparin can actively bind many growthfactors, such as BMPs, TGF-bs, FGFs, HGF, andVEGFs.66,70,121,169,183 In contrast to soluble delivery,morphogens retained in ECM are presented to cellsin a localized manner, which can regulate the spatialpatterning of cell fate.77 As reported during the gas-trulation of Xenopus embryos, PDGF-AA secretedfrom the ectoderm binds to heparin sulfate proteo-glycan (HPSG)-modified, fibronectin-rich ECM andgenerates a gradient to direct mesendoderm migra-tion.153 Engineering strategies inspired by such bio-logical phenomena have been developed to modulatethe spatial presentation of extracellular environmen-tal cues by incorporation of ECM-based microparti-cles into PSC aggregates.15

NOVEL APPLICATIONS ENABLED BY 3D STEM

CELL-DERIVED MICROTISSUES

Tissue Modeling

One significant advancement enabled by three-dimensional stem cell aggregation is the creation ofin vitro microtissues for developmental studies,157

pharmacological screening,166 and disease modeling(Fig. 3).129 Several methods have been explored togenerate tissues via aggregation of differentiated,83,174

progenitor,103 multipotent,7,57 and pluripotent cells.43

The simplest approach is to aggregate homotypic cellpopulations (primary somatic cells, stem cells, or dif-ferentiated progeny) to yield microtissues such as EBs,mesenspheres, cardiospheres, or neurospheres. Whilefunctional microtissues have been achieved by homo-typic seeding approaches, many researchers are alsoinvestigating heterotypic cultures as a means to moreclosely recapitulate the multicellular environmentfound in functional tissues. To achieve formation of3D heterotypic structures, cells can be assembled to-gether within a single aggregate42,45,131 or individualaggregates comprised of distinct homotypic cell pop-ulations can be merged, which enables spatial controland the formation of higher order structures.125 Forexample, merging of aggregates has been accomplishedby a side-by-side seeding approach within micro-wells,17,125 via manipulation using magnetic micro-particles within PSC aggregates,16 or using microfluidic


devices to force aggregates into close proximity andfacilitate merger of distinct spheroids.152

Parallel studies of embryonic development and stemcell differentiation have enabled increased unders-tanding and control of the molecular mechanisms gov-erning cell fate. For example, human PSCs have beensuccessfully differentiated into neurons24,81 cardiomyo-cytes,95,104 hepatocytes,146 and pancreatic endocrinecells,180 largely via sequential administration of bio-chemical cues implicated in various stages of embryo-genesis. While similar approaches to promotedifferentiation in 3D remain an active area of research,the multicellular assembly, including the formation ofadhesions and matrix synthesis, are thought to supporttissue maturation, and therefore are expected to enablethe establishment of complex models of tissue structurein vitro. While complex morphogenesis occurs throughself-organization and differentiation in 3D stemcell-derived organoid models,4,46,97,115,149,151,154 opportu-nities also exist to engineer heterogeneity into the systems,in order to inform aspects of mammalian embryonicdynamics, particularly during some of the earliest cell fatedecisions of the pre-implantation embryo.

Due to the aforementioned progress toward gener-ation of in vitro tissue models, aggregate cultures offeropportunities to model disease progression and screenpharmacological targets. iPS enable patient-specificstudies of disease states and population heterogene-ity.65 The combination of iPS technology with 3Daggregate culture will enable the generation of tissuemodels that manifest cellular disease phenotypes andcan be used for drug screens even if the geneticmechanisms are unknown. To further advance thedrug screening potential, developments in microfluidicsand high-throughput screening have resulted in meth-ods to subject an array of microtissues to varyingmolecule types and gradients in order to define optimaltreatments.30,71

Bioprocessing

The expansion and differentiation of stem cells asmulticellular spheroids is also directly compatible withsuspension culture platforms, which enables the simpleintegration of three-dimensional microtissues withinthe pipeline of bioprocess development. Scalable

FIGURE 3. Applications enabled by 3D stem cell expansion and differentiation. The three-dimensional assembly of stem cellsenables scalable bioprocessing approaches, including microliter scale upstream processing via microfluidics, as well as milliliter/liter-scale manufacturing. In addition, the self-organization and developmentally relevant signaling in microtissues and heterotypiccellular assemblies enables the study of tissue models for pharmacological applications or to model disease pathologies. Themodular nature of stem cell spheroids also highlights the utility for therapeutics, as materials amenable to direct injection fordelivery in vivo.

KINNEY et al.

bioprocessing platforms have been implemented forboth expansion and differentiation of PSC spher-oids,20,53,138,184 and are amenable to scalable culture ofmultipotent stem and progenitor cells.7 As spheroidsmaintained in suspension are not constrained by sur-face area limitations posed by monolayer or micro-carrier cultures,89 it is possible to produce large cellyields by increasing the throughput and batch size,which is advantageous for limiting batch-to-batchvariability. Moreover, an important benefit of scalablebioprocesses is the capacity for automation, control,and monitoring of physiochemical environmentalparameters, in order to precisely and reproduciblycontrol concentrations of oxygen, nutrients, morpho-gens and metabolites.10

While scaffold-based platforms also benefit fromdynamic and controlled bioprocess culture configura-tions,181 the differentiation of stem cells as three-dimensional aggregates, in particular, is ideal forscalable tissue production because it enables parallelprocessing of many individual multicellular structureswithin a single bioreactor volume. Microscale tissuesare easily maintained within the size limitations of non-vascularized tissues,165 in order to provide the ade-quate molecular transport required to maintain cellviability and limit morphogen gradients.124 The agita-tion and physiochemical conditions in mixed bioreac-tor environments are easily tunable and can be refinedin the context of different stem cell applications tocontrol phenotypic parameters, such as prolifera-tion.33,177 For example, the mode and frequency ofmedia replenishment can be monitored and adjusted inreal-time, based upon culture requirements.23,35,85

Moreover, in addition to the precisely maintainedenvironmental parameters, microtissue control isafforded through manipulation of individual aggregateenvironments using microencapsulation.40,172 Together,control of stem cell macro- and microenvironmentsenables maintenance of homogeneity between individ-ual stem cell aggregates,88 which is important fromthe standpoint of quality assurance for industrialapplications.

Although many current bioprocess configurationshave been developed based on modifications fromestablished bioprocessing industries, the distinctrequirements and applications of stem cell aggregatesprovides opportunities to develop new platforms. Inparticular, bioprocesses can be developed in conjunc-tion with additional technologies for upstream moni-toring, processing, and sorting, analogous to industrialmanufacturing pipelines, which require minimal man-ual manipulation. For instance, in addition to standardbiochemical analytics, integration of bulk metricsbased on cell dynamics,164 physical characteristics145 ormetabolite/morphogen profile87 will enable non-destruc-

tive monitoring and processing with high temporalresolution. Interestingly, many analytical approachesare being developed through microfluidic technologies,which provide opportunities to pair platforms acrossdifferent volumetric scales, including the application ofmicrofluidics upstream or downstream of bioreactorcultures.55

In anticipation of scalable stem cell expansion anddifferentiation, laboratory-scale culture techniquesneed to incorporate bioprocess-relevant environmentalparameters. For example, even relatively subtle chan-ges in mixing dynamics can dictate changes in differ-entiated phenotypes.88,133 As multiple parameters (i.e.,mechanical cues, molecular transport, cell adhesionkinetics) are simultaneously modulated in response toenvironmental perturbations, it is traditionally difficultto dissect or systematically perturb the relative influ-ence on stem cell fate.89 In addition to environmentalchanges, the precision of physiochemical control inbioprocess platforms may alter the temporal kineticsor concentration of morphogens required to achievethe same results compared to a batch-fed, static culturesystem.35 Therefore, due to the promise of three-dimensional stem cell aggregates as scalable thera-peutics, bioprocess design should be a consideration,even in laboratory-scale experiments, when developingapproaches intended to control stem cell expansionand differentiation.

In vivo Delivery

Stem cell delivery in vivo is a critical step to translatethe regenerative potential toward clinical applications.Several factors should be taken into considerationprior to delivery and post-delivery, including thera-peutic dose (i.e., cell number), delivery format, and cellsurvival, retention and engraftment. The unique fea-tures of stem cell aggregates provide several advanta-ges as an in vivo delivery platform compared toconventional delivery of cell-populated 3D scaffolds orsingle cell suspensions.

Cell transplantation for regenerative therapies typ-ically requires on the order of several million cells perkilogram of the patient.99 Delivery of therapeuticallyrelevant cell doses in small volumes is a key constraintfor clinical practice, especially for direct intramyocar-dial implantation with typical injection volumesbetween 0.1 mL and 0.5 mL.3,161 High cell densitiesare difficult to achieve using 3D scaffolds, which areoften on the scale of 105 cells per cubic mm.155 Incontrast, cell aggregation in a scaffold-free format canconcentrate cells on the order of 106 cells/mm3 uponaggregation.9,88,101,140 Additionally, by varying for-mation conditions, the diameter and cell compositionof stem cells aggregates can be highly controlled,76,88


which enables customization for specific clinicalapplications.

Implantation of 3D cell/scaffold constructs viainvasive surgical procedures is an unavoidable caveatof most current tissue engineering approaches.13 Incontrast, scaffold-free cell aggregates are readilyinjectable and therefore compatible with minimallyinvasive delivery techniques. Unlike single cell sus-pensions, which can also be easily injected, enzymaticdissociation is not necessary for cell aggregates, whichreduces cell handling and preparation time, and avoidsassociated decreases in cell viability,68 ultimately sup-porting enhanced cell viability and engraftment in vivo.Cell aggregates ranging from 100 to 1000 lm indiameter have been injected intramyocardially, subcu-taneously, or intramuscularly to induce myocar-dial,9,101 adipose,29 or bone140 tissue regeneration,respectively. Cell aggregates can be passed throughstandard needles without clogging and retain theirnative spheroid morphology.9,29 While it has beenreported that single cells suspensions exhibit more than30% acute cell death due to the mechanical disruptioncaused by extensional flow, the viability of aggregatesremained unchanged after passing through a 30Gneedle.10

The functional benefits of current stem cell-basedtherapies have been limited due to the low survival,retention and engraftment of implanted cells. Uponimplantation, donor cells are immediately exposed tothe microenvironment of the injury site, which is oftennot favorable for donor cell survival and induces celldeath.160,176 Various cytoprotective efforts have beenexplored for pre-conditioning stem cells beforeimplantation.67 Alternatively, multicellular aggrega-tion creates a transferable microenvironment amenableto enhancing donor cell survival and engraftment post-implantation. Implantation of multicellular aggregateshas demonstrated significantly higher cell survival(80%) in comparison to single cell suspension (40%)4 days post-implantation following cardiac injury.3

Similarly, intramyocardial injection of human amni-otic-fluid stem-cell aggregates resulted in 50-fold and20-fold increases in cell retention compared to singlecell treatments after 24 h (50%) and 4 weeks (20%)post-implantation, respectively.101 The improvementof short-term cell survival and cell engraftment is aprerequisite for long-term functional enhancement.72

The increased retention and engraftment of stem cellaggregates can be attributed to cell–cell interactionsand enriched ECM. Cell–cell interaction is essential formediating many cellular processes, including cell sur-vival,8 proliferation26 and differentiation.91 In the caseof single cell suspension, this cellular interaction iscompletely abrogated by dissociation, whereas it ispreserved in the cell aggregates.8,9,67,72 As previously

discussed, cell aggregates also retain endogenous ECM(collagen III, fibronectin and laminin),29,100 which istransplanted as part of the spheroid microenvironmentin vivo and supports cell survival, cell retention andfunctional enhancement after transplantation.74

Finally, the composition of cell aggregates may shieldthe interior cell populations from the damage due tothe influence of shear forces upon injection andenable increased retention due to existing cell–cellinteractions.

FUTURE PERSPECTIVES

To date, many approaches have been developed orproposed as a means to precisely control the bio-physical characteristics of stem cell aggregates,including high throughput platforms to control thespheroid size and homogeneity,60,135,139,163 microflu-idic approaches to increase the precision and profile ofmolecular transport,28,59,93,118,168 and biomaterialtechniques to deliver, present, and/or sequester induc-tive cues (physical, chemical) in a highly efficient andspatiotemporally controlled manner.22,123 However, asdiscussed throughout, the impact of biophysical cuesare manifested through multiparametric, synergisticresponses to simple perturbations. Therefore, oppor-tunities remain to develop experimental and mathe-matical approaches to understand the complexinterrelationship between cellular composition (adhe-sions, remodeling) and physical characteristics(mechanics, transport), particularly as a function ofspace and time during morphogenesis. For example,ongoing efforts should focus on establishing advancedmathematical models to simulate the dynamic changesin mass transport within EB culture as the function oftime (differentiation state, phenotype) and developingsensitive experimental tools to sample and control themicroenvironment within aggregates. Such advanceswill elucidate the relative influences of various bio-physical factors in mediating developmental dynam-ics171 and ultimately inform engineering approaches toperturb and control stem cell morphogenesis to guidethe formation of tissue-specific microtissues amenableto screening and therapeutic applications.

Due to the recent increased interest and rapid ad-vances in self-organization and organoid formationfrom PSCs, several questions remain regarding prac-tical translation of such approaches for pharmaceuticaland therapeutic applications. For example, the extentto which microtissues in vitro accurately recapitulatenative tissue function and drug responses in vivo re-mains unclear. As high throughput testing technologiesare developed, the minimal functional units necessaryto recapitulate tissue-scale behaviors will need to be

KINNEY et al.

determined in parallel to establish reproducible, phys-iologically relevant screening platforms. In addition,modular tissue engineering approaches may enable thedevelopment of heterogeneous tissue structures com-prised of heterotypic cell types and/or aggregates forboth ex vivo tissue modeling and for in vivo regenera-tion; however, there is an inherent tradeoff betweenengineering and cell-mediated self-organization.Therefore, parallel studies will elucidate opportunitiesto guide self-organization via systematic perturbationsto stem cell-derived microtissues. Ultimately, modulartissue engineering approaches will enable increasedorganization and complexity within single heteroge-neous tissue units, which will fundamentally impact thefuture of pharmaceutical screening and biomedicaltherapeutics.

ACKNOWLEDGMENTS

The authors are supported by funding from theNational Institute of Health (NIH) (EB010061,GM088291, AR062006) and National Science Foun-dation (NSF) (CBET 0939511). M.A.K. is currentlysupported by an American Heart Association (AHA)Pre-Doctoral Fellowship and previously by an NSFGraduate Research Fellowship.

REFERENCES

1Aegerter-Wilmsen, T., C. M. Aegerter, E. Hafen, and K.Basler. Model for the regulation of size in the wingimaginal disc of drosophila. Mech. Dev. 124:318–326,2007.2Allen, J. W., S. R. Khetani, and S. N. Bhatia. In vitrozonation and toxicity in a hepatocyte bioreactor. Toxicol.Sci. 84:110–119, 2005.3Amado, L. C., et al. Cardiac repair with intramyocardialinjection of allogeneic mesenchymal stem cells aftermyocardial infarction. Proc. Natl Acad. Sci. U. S. A.102:11474–11479, 2005.4Antonica, F., et al. Generation of functional thyroid fromembryonic stem cells. Nature 491:66–71, 2012.5Baker, B. M., and C. S. Chen. Deconstructing the thirddimension: how 3D culture microenvironments alter cel-lular cues. J. Cell Sci. 125:3015–3024, 2012.6Bao, B., J. Jiang, T. Yanase, Y. Nishi, and J. R. Morgan.Connexon-mediated cell adhesion drives microtissue self-assembly. FASEB J. 25:255–264, 2011.7Baraniak, P. R., and T. C. Mcdevitt. Scaffold-free cultureof mesenchymal stem cell spheroids in suspension pre-serves multilineage potential. Cell Tissue Res. 347:701–711, 2012.8Bartosh, T. J., Z. Wang, A. A. Rosales, S. D. Dimitrije-vich, and R. S. Roque. 3D-model of adult cardiac stem

cells promotes cardiac differentiation and resistance tooxidative stress. J. Cell. Biochem. 105:612–623, 2008.9Bauer, M., et al. Adult cardiac progenitor cell aggregatesexhibit survival benefit both in vitro and in vivo. PLoSONE 7:e50491, 2012.

10Bauwens, C., T. Yin, S. Dang, R. Peerani, and P. W.Zandstra. Development of a perfusion fed bioreactor forembryonic stem cell-derived cardiomyocyte generation:oxygen-mediated enhancement of cardiomyocyte output.Biotechnol. Bioeng. 90:452–461, 2005.

11Bauwens, C. L., et al. Control of human embryonic stemcell colony and aggregate size heterogeneity influencesdifferentiation trajectories. Stem Cells 26:2300–2310,2008.

12Bauwens, C. L., et al. Geometric control of cardiomyo-genic induction in human pluripotent stem cells. TissueEng. Part A 17:1901–1909, 2011.

13Bencherif, S. A., et al. Injectable preformed scaffolds withshape-memory properties. Proc. Natl Acad. Sci. U. S. A.109:19590–19595, 2012.

14Bez, A., et al. Neurosphere and neurosphere-forming cells:morphological and ultrastructural characterization. BrainRes. 993:18–29, 2003.

15Bratt-Leal, A. M., R. L. Carpenedo, M. D. Ungrin, P. W.Zandstra, and T. C. Mcdevitt. Incorporation of bioma-terials in multicellular aggregates modulates pluripotentstem cell differentiation. Biomaterials 32:48–56, 2011.

16Bratt-Leal, A. M., K. L. Kepple, R. L. Carpenedo, M. T.Cooke, and T. C. Mcdevitt. Magnetic manipulation andspatial patterning of multi-cellular stem cell aggregates.Integr. Biol. 3:1224–1232, 2011.

17Bratt-Leal, A. M., A. H. Nguyen, K. A. Hammersmith, A.Singh, and T. C. Mcdevitt. A microparticle approach tomorphogen delivery within pluripotent stem cell aggre-gates. Biomaterials 34:7227–7235, 2013.

18Burdsal, C. A., C. H. Damsky, and R. A. Pedersen. Therole of E-cadherin and integrins in mesoderm differenti-ation and migration at the mammalian primitive streak.Development 118:829–844, 1993.

19Burridge, P., D. Anderson, H. Priddle, and M. Barbadillo.Improved human embryonic stem cell embryoid bodyhomogeneity and cardiomyocyte differentiation from anovel V-96 plate aggregation system highlights interlinevariability. Stem Cells 25(4):929–938, 2007.

20Cameron, C. M., W.-S. Hu, and D. S. Kaufman. Im-proved development of human embryonic stem cell-der-ived embryoid bodies by stirred vessel cultivation.Biotechnol. Bioeng. 94:938–948, 2006.

21Carpenedo, R. L., S. A. Seaman, and T. C. Mcdevitt.Microsphere size effects on embryoid body incorporationand embryonic stem cell differentiation. J. Biomed. Mater.Res., Part A 94:466–475, 2010.

22Carpenedo, R. L., et al. Homogeneous and organizeddifferentiation within embryoid bodies induced bymicrosphere-mediated delivery of small molecules. Bio-materials 30:2507–2515, 2009.

23Carrier, R. L., et al. Perfusion improves tissue architectureof engineered cardiac muscle. Tissue Eng. 8:175–188, 2002.

24Chambers, S. M., et al. Highly efficient neural conversionof human ES and ips cells by dual inhibition of smadsignaling. Nat. Biotechnol. 27:275–280, 2009.

25Chan, D. N., S. F. Azghadi, J. Feng, and W. E. Lowry.Ptk7 marks the first human developmental emt in vitro.PLoS ONE 7:e50432, 2012.


26Chen, X., and S. L. Thibeault. Cell–cell interactionbetween vocal fold fibroblasts and bone marrow mesen-chymal stromal cells in three-dimensional hyaluronanhydrogel. J. Tissue Eng. Regen. Med., 2013. DOI:10.1002/term.1757.

27Chen, T., et al. E-cadherin-mediated cell–cell contact iscritical for induced pluripotent stem cell generation. StemCells 28:1315–1325, 2010.

28Choi, J., et al. Wnt5a-mediating neurogenesis of humanadipose tissue-derived stem cells in a 3D microfluidic cellculture system. Biomaterials 32:7013–7022, 2011.

29Chung, H. J., and T. G. Park. Injectable cellular aggre-gates prepared from biodegradable porous microspheresfor adipose tissue engineering. Tissue Eng. Part A15:1391–1400, 2009.

30Cimetta, E., et al. Microfluidic device generating stableconcentration gradients for long term cell culture: appli-cation to wnt3a regulation of b-catenin signaling. LabChip 10:3277–3283, 2010.

31Ciruna, B., and J. Rossant. FGF signaling regulatesmesoderm cell fate specification and morphogeneticmovement at the primitive streak. Dev. Cell 1:37–49, 2001.

32Constantinescu, D., H. L. Gray, P. J. Sammak, G. P.Schatten, and A. B. Csoka. Lamin a/c expression is amarker of mouse and human embryonic stem cell differ-entiation. Stem Cells 24:177–185, 2006.

33Cormier, J. T., N. I. Zur Nieden, D. E. Rancourt, and M.S. Kallos. Expansion of undifferentiated murine embry-onic stem cells as aggregates in suspension culture biore-actors. Tissue Eng. 12:3233–3245, 2006.

34Cox, T. R., and J. T. Erler. Remodeling and homeostasisof the extracellular matrix: implications for fibrotic dis-eases and cancer. Dis. Models Mech. 4:165–178, 2011.

35Csaszar, E., et al. Rapid expansion of human hemato-poietic stem cells by automated control of inhibitoryfeedback signaling. Cell Stem Cell 10:218–229, 2012.

36Cukierman, E., R. Pankov, D. R. Stevens, and K. M.Yamada. Taking cell–matrix adhesions to the thirddimension. Science 294:1708–1712, 2001.

37Dahl, K. N., A. J. S. Ribeiro, and J. Lammerding.Nuclear shape, mechanics, and mechanotransduction.Circ. Res. 102:1307–1318, 2008.

38Daley, W. P., J. M. Kohn, and M. Larsen. A focaladhesion protein-based mechanochemical checkpointregulates cleft progression during branching morphogen-esis. Dev. Dyn. 240:2069–2083, 2011.

39Daley, W. P., S. B. Peters, and M. Larsen. Extracellularmatrix dynamics in development and regenerative medi-cine. J. Cell Sci. 121:255–264, 2008.

40Dang, S. M., S. Gerecht-Nir, J. Chen, J. Itskovitz-Eldor,and P. W. Zandstra. Controlled, scalable embryonic stemcell differentiation culture. Stem Cells 22:275–282, 2004.

41Davey, R. E., and P. W. Zandstra. Spatial organization ofembryonic stem cell responsiveness to autocrine gp130ligands reveals an autoregulatory stem cell niche. StemCells 24:2538–2548, 2006.

42Desroches, B. R., et al. Functional scaffold-free 3-D car-diac microtissues: a novel model for the investigation ofheart cells. Am. J. Physiol. Heart Circ. Physiol.302:H2031–H2042, 2012.

43Doetschman, T. C., H. Eistetter, M. Katz, W. Schmidt,and R. Kemler. The in vitro development of blasto-cyst-derived embryonic stem cell lines: formation of vis-ceral yolk sac, blood islands and myocardium. J. Embryol.Exp. Morphol. 87:27–45, 1985.

44Dufort, C. C., M. J. Paszek, and V. M. Weaver. Balancingforces: architectural control of mechanotransduction. Nat.Rev. Mol. Cell Biol. 12:308–319, 2011.

45Eckermann, C. W., K. Lehle, S. A. Schmid, D. N.Wheatley, and L. A. Kunz-Schughart. Characterizationand modulation of fibroblast/endothelial cell co-culturesfor the in vitro preformation of three-dimensional tubularnetworks. Cell Biol. Int. 35:1097–1110, 2011.

46Eiraku, M., et al. Self-organizing optic-cup morphogene-sis in three-dimensional culture. Nature 472:51–56, 2011.

47Eldar, A., et al. Robustness of the BMP morphogengradient in drosophila embryonic patterning. Nature419:304–308, 2002.

48Engler, A. J., S. Sen, H. L. Sweeney, and D. E. Discher.Matrix elasticity directs stem cell lineage specification.Cell 126:677–689, 2006.

49Engler, A., et al. Substrate compliance versus ligand densityin cell on gel responses. Biophys. J . 86:617–628, 2004.

50Ewald, A. J., A. Brenot, M. Duong, B. S. Chan, and Z.Werb. Collective epithelial migration and cell rearrange-ments drive mammary branching morphogenesis. Dev.Cell 14:570–581, 2008.

51Farhadifar, R., J.-C. Roper, B. Aigouy, S. Eaton, and F.Julicher. The influence of cell mechanics, cell–cell inter-actions, and proliferation on epithelial packing. Curr.Biol. 17:2095–2104, 2007.

52Fletcher, D. A., and R. D. Mullins. Cell mechanics andthe cytoskeleton. Nature 463:485–492, 2010.

53Fok, E. Y. L., and P. W. Zandstra. Shear-controlledsingle-step mouse embryonic stem cell expansion andembryoid body-based differentiation. Stem Cells 23:1333–1342, 2005.

54Franke, J. D., R. A. Montague, and D. P. Kiehart.Nonmuscle myosin II generates forces that transmit ten-sion and drive contraction in multiple tissues duringdorsal closure. Curr. Biol. 15:2208–2221, 2005.

55Fridley, K. M., M. A. Kinney, and T. C. Mcdevitt.Hydrodynamic modulation of pluripotent stem cells. StemCell Res. Ther. 3:45, 2012.

56Friedl, P., and D. Gilmour. Collective cell migration inmorphogenesis, regeneration and cancer. Nat. Rev. Mol.Cell Biol. 10:445–457, 2009.

57Frith, J. E., B. Thomson, and P. G. Genever. Dynamicthree-dimensional culture methods enhance mesenchymalstem cell properties and increase therapeutic potential.Tissue Eng. C Methods 16:735–749, 2010.

58Fu, J., et al. Mechanical regulation of cell function withgeometrically modulated elastomeric substrates. Nat.Methods 7:733–736, 2010.

59Fung, W.-T., A. Beyzavi, P. Abgrall, N.-T. Nguyen, andH.-Y. Li. Microfluidic platform for controlling the dif-ferentiation of embryoid bodies. Lab Chip 9:2591–2595,2009.

60Gallego-Perez, D., et al. High throughput assembly ofspatially controlled 3D cell clusters on a micro/nanoplat-form. Lab Chip 10:775–782, 2010.

61Gardner, R. L., and J. Rossant. Investigation of the fateof 4-5 day post-coitum mouse inner cell mass cells byblastocyst injection. J. Embryol. Exp. Morphol. 52:141–152, 1979.

62Go, M. J., D. S. Eastman, and S. Artavanis-Tsakonas.Cell proliferation control by notch signaling in drosophiladevelopment. Development 125:2031–2040, 1998.

63Goh, S.-K., P. Olsen, and I. Banerjee. Extracellular matrixaggregates from differentiating embryoid bodies as a

KINNEY et al.

http://dx.doi.org/10.1002/term.1757

scaffold to support ESC proliferation and differentiation.PLoS ONE 8:e61856, 2013.

64Gomez, E. W., Q. K. Chen, N. Gjorevski, and C. M.Nelson. Tissue geometry patterns epithelial-mesenchymaltransition via intercellular mechanotransduction. J. Cell.Biochem. 110:44–51, 2010.

65Grskovic, M., A. Javaherian, B. Strulovici, and G. Q.Daley. Induced pluripotent stem cells—opportunities fordisease modelling and drug discovery. Nat. Rev. DrugDiscov. 10:915–929, 2011.

66Gupta, P., et al. Structurally specific heparan sulfatessupport primitive human hematopoiesis by formation of amultimolecular stem cell niche. Blood 92:4641–4651, 1998.

67Haider, K. H., and M. Ashraf. Preconditioning approachin stem cell therapy for the treatment of infarcted heart.Prog. Mol. Biol. Transl. Sci. 111:323–356, 2012.

68Heng, B. C., C. M. Cowan, and S. Basu. Comparison ofenzymatic and non-enzymatic means of dissociatingadherent monolayers of mesenchymal stem cells. Biol.Proced. Online 11:161–169, 2009.

69Hong, S.-H., T. Werbowetski-Ogilvie, V. Ramos-Mejia, J.B. Lee, and M. Bhatia. Multiparameter comparisons ofembryoid body differentiation toward human stem cellapplications. Stem Cell Res 5:120–130, 2010.

70Hongisto, H., et al. Laminin-511 expression is associatedwith the functionality of feeder cells in human embryonicstem cell culture. Stem Cell Res 8:97–108, 2012.

71Hsiao, A. Y., et al. 384 hanging drop arrays give excellentz-factors and allow versatile formation of co-culturespheroids. Biotechnol. Bioeng. 109:1293–1304, 2012.

72Hu, X., et al. Transplantation of hypoxia-preconditionedmesenchymal stem cells improves infarcted heart functionvia enhanced survival of implanted cells and angiogenesis.J. Thorac. Cardiovasc. Surg. 135:799–808, 2008.

73Huang, R. Y.-J., P. Guilford, and J. P. Thiery. Earlyevents in cell adhesion and polarity during epithelial–mesenchymal transition. J. Cell Sci. 125:4417–4422, 2012.

74Huang, N. F., and S. Li. Regulation of the matrixmicroenvironment for stem cell engineering and regener-ative medicine. Ann. Biomed. Eng. 39:1201–1214, 2011.

75Hufnagel, L., A. A. Teleman, H. Rouault, S. M. Cohen,and B. I. Shraiman. On the mechanism of wing sizedetermination in fly development. Proc. Natl Acad. Sci.U. S. A. 104:3835–3840, 2007.

76Hwang, Y.-S., et al. Microwell-mediated control ofembryoid body size regulates embryonic stem cell fate viadifferential expression of WNT5a and WNT11. Proc. NatlAcad. Sci. U. S. A. 106:16978–16983, 2009.

77Hynes, R. O. The extracellular matrix: not just prettyfibrils. Science 326:1216–1219, 2009.

78Ingber, D. E. Tensegrity: the architectural basis of cellularmechanotransduction.Annu. Rev. Physiol. 59:575–599, 1997.

79Ingber, D. E. Cellular mechanotransduction: putting allthe pieces together again. FASEB J. 20:811–827, 2006.

80Jackson, S. A., J. Schiesser, E. G. Stanley, and A. G.Elefanty. Differentiating embryonic stem cells passthrough ‘temporal windows’ that mark responsiveness toexogenous and paracrine mesendoderm inducing signals.PLoS ONE 5:e10706, 2010.

81Kamiya, D., et al. Intrinsic transition of embryonic stem-cell differentiation into neural progenitors. Nature 470:503–509, 2011.

82Keller, R., L. A. Davidson, and D. R. Shook. How we areshaped: the biomechanics of gastrulation. Differentiation71:171–205, 2003.

83Kelm, J. M., et al. A novel concept for scaffold-free vesseltissue engineering: self-assembly of microtissue buildingblocks. J. Biotechnol. 148:46–55, 2010.

84Kim, H. Y., and L. A. Davidson. Punctuated actin con-tractions during convergent extension and their permissiveregulation by the non-canonical Wnt-signaling pathway.J. Cell Sci. 124:635–646, 2011.

85Kim, L., M. D. Vahey, H.-Y. Lee, and J. Voldman.Microfluidic arrays for logarithmically perfused embry-onic stem cell culture. Lab Chip 6:394–406, 2006.

86Kim, J. M., et al. Assessment of differentiation aspects bythe morphological classification of embryoid bodiesderived from human embryonic stem cells. Stem CellsDev. 20:1925–1935, 2011.

87Kingsmore, S. F. Multiplexed protein measurement:technologies and applications of protein and antibodyarrays. Nat. Rev. Drug Discov. 5:310–320, 2006.

88Kinney, M. A., R. Saeed, and T. C. Mcdevitt. Systematicanalysis of embryonic stem cell differentiation in hydro-dynamic environments with controlled embryoid bodysize. Integr. Biol. 4:641–650, 2012.

89Kinney, M. A., C. Y. Sargent, and T. C. Mcdevitt. Themultiparametric effects of hydrodynamic environments onstem cell culture. Tissue Eng. B Rev. 17:249–262, 2011.

90Kinney, M. A., C. Y. Sargent, and T. C. Mcdevitt.Temporal modulation of b-catenin signaling by multicel-lular aggregation kinetics impacts embryonic stem cellcardiomyogenesis. Stem Cells Dev. 22(19):2665–2677,2013.

91Kirouac, D. C., et al. Cell–cell interaction networks reg-ulate blood stem and progenitor cell fate. Mol. Syst. Biol.5:293, 2009.

92Knospe, W. H., S. A. Gregory, S. G. Husseini, W. Fried,and F. E. Trobaugh. Origin and recovery of colony-forming units in locally curetted bone marrow of mice.Blood 39:331–340, 1972.

93Kumachev, A., et al. High-throughput generation ofhydrogel microbeads with varying elasticity for cellencapsulation. Biomaterials 32:1477–1483, 2011.

94Kurosawa, H. Methods for inducing embryoid body for-mation: in vitro differentiation system of embryonic stemcells. J. Biosci. Bioeng. 103:389–398, 2007.

95Laflamme, M. A., et al. Cardiomyocytes derived fromhuman embryonic stem cells in pro-survival factors en-hance function of infarcted rat hearts. Nat. Biotechnol.25:1015–1024, 2007.

96Lammerding, J., et al. Lamin a/c deficiency causes defec-tive nuclear mechanics and mechanotransduction. J. Clin.Invest. 113:370–378, 2004.

97Lancaster, M. A., et al. Cerebral organoids model humanbrain development and microcephaly. Nature 501:373–379, 2013.

98Larue, L., et al. A role for cadherins in tissue formation.Development 122:3185–3194, 1996.

99Le Blanc, K., et al. Mesenchymal stem cells for treatmentof steroid-resistant, severe, acute graft-versus-host disease:a phase ii study. Lancet 371:1579–1586, 2008.

100Lee, W.-Y., et al. The use of injectable spherically sym-metric cell aggregates self-assembled in a thermo-respon-sive hydrogel for enhanced cell transplantation.Biomaterials 30:5505–5513, 2009.

101Lee, W.-Y., et al. Enhancement of cell retention andfunctional benefits in myocardial infarction using humanamniotic-fluid stem-cell bodies enriched with endogenousecm. Biomaterials 32:5558–5567, 2011.


102Li, L., and W. B. Neaves. Normal stem cells and cancerstem cells: the niche matters. Cancer Res. 66:4553–4557,2006.

103Li, T.-S., et al. Cardiospheres recapitulate a niche-likemicroenvironment rich in stemness and cell-matrix inter-actions, rationalizing their enhanced functional potencyfor myocardial repair. Stem Cells 28:2088–2098, 2010.

104Lian, X., et al. Directed cardiomyocyte differentiationfrom human pluripotent stem cells by modulating Wnt/b-catenin signaling under fully defined conditions. Nat.Protoc. 8:162–175, 2013.

105Liu, Z., et al. Mechanical tugging force regulates the sizeof cell–cell junctions. Proc. Natl Acad. Sci. U. S. A.107:9944–9949, 2010.

106Maniscalco, W. M., and M. H. Campbell. Transforminggrowth factor-beta induces a chondroitin sulfate/dermatansulfate proteoglycan in alveolar type ii cells. Am. J.Physiol. 266:L672–L680, 1994.

107Manning, M. L., R. A. Foty, M. S. Steinberg, and E.-M.Schoetz. Coaction of intercellular adhesion and corticaltension specifies tissue surface tension. Proc. Natl Acad.Sci. U. S. A. 107:12517–12522, 2010.

108Matthews, B. D., D. R. Overby, R. Mannix, and D. E.Ingber. Cellular adaptation to mechanical stress: role ofintegrins, rho, cytoskeletal tension and mechanosensitiveion channels. J. Cell Sci. 119:508–518, 2006.

109Mcbeath, R., D. M. Pirone, C. M. Nelson, K. Bhadriraju,and C. S. Chen. Cell shape, cytoskeletal tension, andRhoA regulate stem cell lineage commitment. Dev. Cell6:483–495, 2004.

110Mohr, J. C., et al. The microwell control of embryoid bodysize in order to regulate cardiac differentiation of humanembryonic stem cells. Biomaterials 31:1885–1893, 2010.

111Morrison, S. J., and A. C. Spradling. Stem cells and ni-ches: mechanisms that promote stem cell maintenancethroughout life. Cell 132:598–611, 2008.

112Mummery, C. L., et al. Visceral-endoderm-like cell linesinduce differentiation of murine p19 embryonal carcinomacells. Differentiation 46:51–60, 1991.

113Nair, R., A. V. Ngangan, M. L. Kemp, and T. C. Mcdevitt.Gene expression signatures of extracellular matrix andgrowth factors during embryonic stem cell differentiation.PLoS ONE 7:e42580, 2012.

114Nair, R., S. Shukla, and T. C. Mcdevitt. Acellular matricesderived from differentiating embryonic stem cells. J. Bio-med. Mater. Res A 87:1075–1085, 2008.

115Nakano, T., et al. Self-formation of optic cups and stor-able stratified neural retina from human ESCs. Cell StemCell 10:771–785, 2012.

116Nelson, C. M., et al. Emergent patterns of growth con-trolled by multicellular form and mechanics. Proc. NatlAcad. Sci. U. S. A. 102:11594–11599, 2005.

117Niebruegge, S., et al. Generation of human embryonicstem cell-derived mesoderm and cardiac cells using size-specified aggregates in an oxygen-controlled bioreactor.Biotechnol. Bioeng. 102:493–507, 2009.

118Ong, S.-M., et al. A gel-free 3D microfluidic cell culturesystem. Biomaterials 29:3237–3244, 2008.

119Orlichenko, L. S., and D. C. Radisky. Matrix metallo-proteinases stimulate epithelial-mesenchymal transitionduring tumor development. Clin. Exp. Metastasis 25:593–600, 2008.

120Page-Mccaw, A., A. J. Ewald, and Z. Werb. Matrixmetalloproteinases and the regulation of tissue remodel-ling. Nat. Rev. Mol. Cell Biol. 8:221–233, 2007.

121Parker, T., Z. Upton, and D. Leavesley. Vitronectinmodulates human mesenchymal stem cell response toinsulin-like growth factor-i and transforming growth fac-tor beta 1 in a serum-free environment. Tissue Eng. Part A15:1415–1426, 2009.

122Paszek, M. J., D. Boettiger, V. M. Weaver, and D. A.Hammer. Integrin clustering is driven by mechanicalresistance from the glycocalyx and the substrate. PLoSComput. Biol. 5:e1000604, 2009.

123Purpura, K. A., A. M. Bratt-Leal, K. A. Hammersmith, T.C. Mcdevitt, and P. W. Zandstra. Systematic engineeringof 3D pluripotent stem cell niches to guide blood devel-opment. Biomaterials 33:1271–1280, 2012.

124Radisic, M., et al. Oxygen gradients correlate with celldensity and cell viability in engineered cardiac tissue.Biotechnol. Bioeng. 93:332–343, 2006.

125Rago, A. P., D. M. Dean, and J. R. Morgan. Controllingcell position in complex heterotypic 3D microtissues bytissue fusion. Biotechnol. Bioeng. 102:1231–1241, 2009.

126Rebustini, I. T., et al. Laminin alpha5 is necessary forsubmandibular gland epithelial morphogenesis and influ-ences fgfr expression through beta1 integrin signaling.Dev. Biol. 308:15–29, 2007.

127Ruiz, S. A., and C. S. Chen. Emergence of patterned stemcell differentiation within multicellular structures. StemCells 26:2921–2927, 2008.

128Sachlos, E., and D. T. Auguste. Embryoid body mor-phology influences diffusive transport of inductive bio-chemicals: a strategy for stem cell differentiation.Biomaterials 29:4471–4480, 2008.

129Saha, K., and R. Jaenisch. Technical challenges in usinghuman induced pluripotent stem cells to model disease.Cell Stem Cell 5:584–595, 2009.

130Sahai, E., and C. J. Marshall. Rock and Dia haveopposing effects on adherens junctions downstream ofrho. Nat. Cell Biol. 4:408–415, 2002.

131Saleh, F. A., J. E. Frith, J. A. Lee, and P. G. Genever.Three-dimensional in vitro culture techniques for mesen-chymal stem cells. Methods Mol. Biol. 916:31–45, 2012.

132Sargent, C. Y., G. Y. Berguig, and T. C. Mcdevitt. Car-diomyogenic differentiation of embryoid bodies is pro-moted by rotary orbital suspension culture. Tissue Eng.Part A 15:331–342, 2009.

133Sargent, C. Y., et al. Hydrodynamic modulation ofembryonic stem cell differentiation by rotary orbital sus-pension culture. Biotechnol. Bioeng. 105:611–626, 2010.

134Sasai, Y. Next-generation regenerative medicine: organo-genesis from stem cells in 3D culture. Cell Stem Cell12:520–530, 2013.

135Sasaki, D., et al. Mass preparation of size-controlledmouse embryonic stem cell aggregates and induction ofcardiac differentiation by cell patterning method. Bioma-terials 30:4384–4389, 2009.

136Sauer, H., et al. Characteristics of calcium sparks incardiomyocytes derived from embryonic stem cells. Am. J.Physiol. Heart Circ. Physiol. 281:H411–H421, 2001.

137Scadden, D. T. The stem-cell niche as an entity of action.Nature 441:1075–1079, 2006.

138Schroeder, M., et al. Differentiation and lineage selectionof mouse embryonic stem cells in a stirred bench scalebioreactor with automated process control. Biotechnol.Bioeng. 92:920–933, 2005.

139Selimovic, S., et al. Microfabricated polyester conical mi-crowells for cell culture applications. Lab Chip 11:2325–2332, 2011.

KINNEY et al.

140Shen, F. H., et al. Implications of adipose-derived stromalcells in a 3D culture system for osteogenic differentiation:an in vitro and in vivo investigation. Spine J. 13:32–43,2013.

141Shimozono, S., T. Iimura, T. Kitaguchi, S.-I. Higashijima,and A. Miyawaki. Visualization of an endogenous retinoicacid gradient across embryonic development. Nature496:363–366, 2013.

142Shraiman, B. I. Mechanical feedback as a possible regu-lator of tissue growth. Proc. Natl Acad. Sci. U. S. A.102:3318–3323, 2005.

143Shukla, S., et al. Synthesis and organization of hyaluronanand versican by embryonic stem cells undergoing embry-oid body differentiation. J. Histochem. Cytochem. 58:345–358, 2010.

144Sims, J. R., S. Karp, and D. E. Ingber. Altering the cel-lular mechanical force balance results in integrated chan-ges in cell, cytoskeletal and nuclear shape. J. Cell Sci.103(Pt 4):1215–1222, 1992.

145Singh, A., et al. Adhesion strength-based, label-free iso-lation of human pluripotent stem cells. Nat. Methods10:438–444, 2013.

146Si-Tayeb, K., et al. Highly efficient generation of humanhepatocyte-like cells from induced pluripotent stem cells.Hepatology 51:297–305, 2010.

147Son, M.-Y., H.-J. Kim, M.-J. Kim, and Y. S. Cho.Physical passaging of embryoid bodies generated fromhuman pluripotent stem cells. PLoS ONE 6:e19134, 2011.

148Son, Y. S., et al. Brief report: L1 cell adhesion molecule, anovel surface molecule of human embryonic stem cells, isessential for self-renewal and pluripotency. Stem Cells29:2094–2099, 2011.

149Spence, J. R., et al. Directed differentiation of humanpluripotent stem cells into intestinal tissue in vitro. Nature470:105–109, 2011.

150Stankovich, B. L., E. Aguayo, F. Barragan, A. Sharma,and M. G. Pallavicini. Differential adhesion moleculeexpression during murine embryonic stem cell commit-ment to the hematopoietic and endothelial lineages. PLoSONE 6:e23810, 2011.

151Suga, H., et al. Self-formation of functional adenohy-pophysis in three-dimensional culture. Nature 480:57–62,2011.

152Suri, S., S. Ankur, A. H. Nguyen, A. M. Bratt-Leal, T. C.Mcdevitt, and H. Lu. Microfluidic patterning of embry-onic multicellular aggregates for in vitro developmentstudies. Lab Chip 13(23):4617–4624, 2013.

153Symes, K., E. M. Smith, M. Mitsi, and M. A. Nugent.Sweet cues: how heparan sulfate modification of fibro-nectin enables growth factor guided migration of embry-onic cells. Cell Adh Migr. 4:507–510, 2010.

154Takebe, T., et al. Vascularized and functional human liverfrom an ipsc-derived organ bud transplant. Nature499:481–484, 2013.

155Talukdar, S., Q. T. Nguyen, A. C. Chen, R. L. Sah, and S.C. Kundu. Effect of initial cell seeding density on 3D-engineered silk fibroin scaffolds for articular cartilage tis-sue engineering. Biomaterials 32:8927–8937, 2011.

156Taylor, G., M. S. Lehrer, P. J. Jensen, T. T. Sun, and R.M. Lavker. Involvement of follicular stem cells in formingnot only the follicle but also the epidermis. Cell 102:451–461, 2000.

157Ten Berge, D., et al. Wnt signaling mediates self-organi-zation and axis formation in embryoid bodies. Cell StemCell 3:508–518, 2008.

158Toole, B. P., A. Zoltan-Jones, S. Misra, and S. Ghatak.Hyaluronan: a critical component of epithelial-mesen-chymal and epithelial-carcinoma transitions. Cells TissuesOrgans 179:66–72, 2005.

159Trappmann, B., et al. Extracellular-matrix tethering reg-ulates stem-cell fate. Nat. Mater. 11:642–649, 2012.

160Tremoleda, J. L., et al. Assessment of a preclinical modelfor studying the survival and engraftment of human stemcell derived osteogenic cell populations following ortho-topic implantation. J. Musculoskelet. Neuronal Interact.12:241–253, 2012.

161Tse, H.-F., et al. Paracrine effects of direct intramyocar-dial implantation of bone marrow derived cells to enhanceneovascularization in chronic ischaemic myocardium. Eur.J. Heart Fail. 9:747–753, 2007.

162Ueno, S., et al. Biphasic role for Wnt/beta-catenin signalingin cardiac specification in zebrafish and embryonic stemcells. Proc. Natl Acad. Sci. U. S. A. 104:9685–9690, 2007.

163Ungrin, M. D., C. Joshi, A. Nica, C. Bauwens, and P. W.Zandstra. Reproducible, ultra high-throughput formationof multicellular organization from single cell suspen-sion-derived human embryonic stem cell aggregates. PLoSONE 3:e1565, 2008.

164Ungrin, M. D., et al. Rational bioprocess design forhuman pluripotent stem cell expansion and endodermdifferentiation based on cellular dynamics. Biotechnol.Bioeng. 109:853–866, 2012.

165Van Winkle, A. P., I. D. Gates, and M. S. Kallos. Masstransfer limitations in embryoid bodies during humanembryonic stem cell differentiation. Cells Tissues Organs196:34–47, 2012.

166Vinci, M., et al. Advances in establishment and analysis ofthree-dimensional tumor spheroid-based functional assaysfor target validation and drug evaluation. BMC Biol.10:29, 2012.

167Von Dassow, M., J. A. Strother, and L. A. Davidson.Surprisingly simple mechanical behavior of a complexembryonic tissue. PLoS ONE 5:e15359, 2010.

168Wan, C.-R., S. Chung, and R. D. Kamm. Differentiationof embryonic stem cells into cardiomyocytes in a compli-ant microfluidic system. Ann. Biomed. Eng. 39:1840–1847,2011.

169Wang, X., R. E. Harris, L. J. Bayston, and H. L. Ashe.Type IV collagens regulate BMP signalling in Drosophila.Nature 455:72–77, 2008.

170Wartenberg, M., et al. Tumor-induced angiogenesis stud-ied in confrontation cultures of multicellular tumorspheroids and embryoid bodies grown from pluripotentembryonic stem cells. FASEB J. 15:995–1005, 2001.

171White, D. E., M. A. Kinney, T. C. Mcdevitt, and M. L.Kemp. Spatial pattern dynamics of 3D stem cell loss ofpluripotency via rules-based computational modeling.PLoS Comput. Biol. 9:e1002952, 2013.

172Wilson, J. L., and T. C. Mcdevitt. Stem cell microencap-sulation for phenotypic control, bioprocessing, andtransplantation. Biotechnol. Bioeng. 110(3):667–682, 2012.

173Wolfenson, H., I. Lavelin, and B. Geiger. Dynamic regu-lation of the structure and functions of integrin adhesions.Dev. Cell 24:447–458, 2013.

174Wong, S. F., et al. Concave microwell based size-con-trollable hepatosphere as a three-dimensional liver tissuemodel. Biomaterials 32:8087–8096, 2011.

175Wozniak, M. A., and C. S. Chen. Mechanotransduction indevelopment: a growing role for contractility. Nat. Rev.Mol. Cell Biol. 10:34–43, 2009.


176Wu, K. H., X. M. Mo, Z. C. Han, and B. Zhou. Stem cellengraftment and survival in the ischemic heart. Ann.Thorac. Surg. 92:1917–1925, 2011.

177Yirme, G., M. Amit, I. Laevsky, S. Osenberg, and J.Itskovitz-Eldor. Establishing a dynamic process for theformation, propagation, and differentiation of humanembryoid bodies. Stem Cells Dev. 17:1227–1241, 2008.

178Yoon, B. S., et al. Enhanced differentiation of humanembryonic stem cells into cardiomyocytes by combininghanging drop culture and 5-azacytidine treatment. Dif-ferentiation 74:149–159, 2006.

179Zhang, L. Glycosaminoglycan (gag) biosynthesis and gag-binding proteins. Prog. Mol. Biol. Transl. Sci. 93:1–17,2010.

180Zhang, D., et al. Highly efficient differentiation of humanES cells and ips cells into mature pancreatic insulin-pro-ducing cells. Cell Res. 19:429–438, 2009.

181Zhao, F., and T. Ma. Perfusion bioreactor system forhuman mesenchymal stem cell tissue engineering: dynamiccell seeding and construct development. Biotechnol. Bio-eng. 91:482–493, 2005.

182Zhou, J., H. Y. Kim, and L. A. Davidson. Actomyosinstiffens the vertebrate embryo during crucial stages ofelongation and neural tube closure. Development 136:677–688, 2009.

183Zhu, Y., A. Oganesian, D. R. Keene, and L. J. Sandell.Type IIA procollagen containing the cysteine-rich aminopropeptide is deposited in the extracellular matrix of pre-chondrogenic tissue and binds to TGF-beta1 and BMP-2.J. Cell Biol. 144:1069–1080, 1999.

184Zweigerdt, R., R. Olmer, H. Singh, A. Haverich, andU. Martin. Scalable expansion of human pluripotentstem cells in suspension culture. Nat. Protoc. 6:689–700, 2011.

KINNEY et al.

engineering three-dimensional stem cell morphogenesis for the development of tissue models and...

Documents