[14] TISSUE ENGINEERING WITH HUMAN ES cells 303

[14] Tissue Engineering Using Human EmbryonicStem Cells

By SHAHAR COHEN, LUCY LESHANSKI, and JOSEPH ITSKOVITZ‐ELDOR

Abstract

The possibility of using stem cells for tissue engineering has greatlyencouraged scientists to design new platforms in the field of regenerativeand reconstructive medicine. Stem cells have the ability to rejuvenate andrepair damaged tissues and can be derived from both embryonic and adultsources. Among cell types suggested as a cell source for tissue engineering(TE), human embryonic stem cells (hESCs) are one of the most promisingcandidates. Isolated from the inner cell mass of preimplantation stage blas-tocysts, they possess the ability to differentiate into practically all adult celltypes. In addition, their unlimited self‐renewal capacity enables the genera-tion of sufficient amount of cells for cell‐based TE applications. Yet, severalimportant challenges are to be addressed, such as the isolation of the desiredcell type and gaining control over its differentiation and proliferation.Ultimately, combing scaffolding and bioactive stimuli, newly designedbioengineered constructs, could be assembled and applied to various clinicalapplications. Here we define the culture conditions for the derivation ofconnective tissue lineage progenitors, design strategies, and highlight thespecial considerations when using hESCs for TE applications.

Introduction

Tissue engineering (TE) is an evolving interdisciplinary area that com-bines biological and engineering principles aimed at mimicking the naturalprocesses of tissue formation and providing transplantable substitutes for thefield of reconstructive and regenerativemedicine (Vacanti andLanger, 1999).

The essentials of TE involve cells, scaffolds, and bioactive factors, suchas chemical substances and mechanical stimuli. For cells to be viable for TE,they must be easily isolated, sufficient in number, and have an appropriatedefined and controlled phenotype. A number of cell types have already beenused for TE applications, including fully matured cells derived from adulttissues and stem cells derived from either embryonic, fetal, or adult tissues(Sharma and Elisseeff, 2004).

Designing the right strategy for engineering tissues using stem cells isespecially challenging, because it requires plastic cells to form tissues while

METHODS IN ENZYMOLOGY, VOL. 420 0076-6879/06 $35.00Copyright 2006, Elsevier Inc. All rights reserved. DOI: 10.1016/S0076-6879(06)20014-4

304 tissue engineering and regenerative medicine [14]

differentiating. The formation of stem cell–derived neo‐tissues integratesseveral dynamic processes occurring simultaneously. Primarily, the basicbuilding block, a stem cell, is a very powerful and flexible unit. Its differen-tiation and proliferation characteristics have an inverse relationship; themore it differentiates, the less it proliferates.

In addition, the surrounding matrix builds and degrades at the same time,enabling cell migration and spatial organization. Thus, themicroenvironmentis continuously remodeling and, in turn, affects basic cellular processes.

Ultimately, in order for higher‐order engineered tissues to be suitable forgraftment into patients, it is essential to have a vascular network assimilated,enabling proper nutritional supply and waste disposal. This can be achievedthrough either stem cell–derived vascular progenitors setting up the infra-structure for functional blood vessels or by promoting host‐derived vascula-ture invading and integrating into the graft. Possibly, stem cell–based tissueengineered grafts, being growing and differentiating biological elements,will continue remodeling once transplanted in vivo, adjusting to the specifichost requirements.

As for having the ultimate cell source to accomplish these goals, humanembryonic stem cells (hESCs) hold great promise to be just that. Ever sincethey were first isolated (Thomson et al., 1998), they have been shown topossess a developmental potential to differentiate into cells representing allthree embryonic germ layers (Itskovitz‐Eldor et al., 2000), including neurons(Reubinoff et al., 2001), cardiomyocytes (Kehat et al., 2001), hematopoieticcells (Kaufman et al., 2001), endothelial cells (Levenberg et al., 2002), andmore. In addition, their practically unlimited self‐renewal capacity can pro-vide a sufficient amount of cells needed for TE applications. Nevertheless,these unique properties raise several critical issues. Because undifferentiatedhESCs have the potential to form tumors in vitro, it is crucial to study how todirect and control their differentiation and proliferation.

Special Considerations When Using hESCs as the Cell Source for TE

Previous studies have provided protocols for differentiating hESCsalong various lineages. Cell phenotype and functionality are usually testedto ensure that genuine differentiation has occurred. The following is anoutline of the general principles and special considerations when usinghESCs for TE applications.

Growing hESCs in Defined Animal‐Free Conditions

Traditionally, hESCs are cocultured with feeder layers made ofmouse embryonic fibroblasts (MEFs) and fed with serum‐containing media

[14] TISSUE ENGINEERING WITH HUMAN ES cells 305

(Pera et al., 2003); they can, therefore, be considered as xenografts.Although MEF feeders are essential to support the growth of undifferen-tiated hESCs, the use of animal products is associated with the risk ofpathogen transmission, thus major advantages were made in this field.Alternatives include several human‐derived feeder layers, feeder‐freeconditioned medium, and defined medium supplemented with growthfactors and matrix proteins (Amit et al., 2003, 2004; Mallon et al., 2006).

Ultimately, new hESC lines should be derived in defined, animal‐freeculture conditions to be suitable for clinical applications. Guidelines forsetting defined, animal‐free culture systems are beyond the scope of thischapter and may be found elsewhere (Ludwig et al., 2006).

Obtaining the Desired Cell Population

hESCs are the most potent stem cells available for TE; therefore,controlling their differentiation into the desired cell type is of great chal-lenge. In general, hESCs can be induced to differentiate once removedfrom the MEF feeders and introduced with bioactive signals. This is doneeither directly or through the formation of embryoid bodies (EBs). EBs aresmall clumps of hESC colonies grown in suspension, which form three‐dimensional (3‐D) spheroid bodies representing a differentiation modelwith the widest spectrum of cell types that can be achieved in vitro.Differentiating cells within the EBs enjoy cell–cell interaction and para-crine effects of the three embryonic germ layers, in a developing 3‐Dmicroenvironment that mimics to the closest extent the temporal and spatialprocesses taking place in the developing embryo (Dvash and Benvenisty,2004). Although direct differentiation is possible, most differentiationsystems rely on EB formation.

Once cells start to differentiate, a variety of bioactive manipulationscan be applied to control and direct the differentiation journey. In general,these include:

� Soluble signals, such as growth factors, hormones, and cell‐conditioned media

� Genetic modifications, such as overexpression of transcription factorsknown to derive stem cells into the desired cell type

� Direct or indirect coculturing with other developing or maturesomatic cell populations

� Physical stimuli, such as mechanical forces, temperature, and oxygena-tion changes.Whetherdifferentiation is allowed tooccur spontaneouslyor inadirected manner, the resultant cell population in most known protocols is stillheterogeneous, which limits its potential to be used as a cell source for TE.

306 tissue engineering and regenerative medicine [14]

Thus, the next challenge is the isolation of the desired cell type. Methods forisolating a specific cell type from heterogeneous populations include eitherpositively selecting the desired cells or negatively removing the undesiredcells. Both can be done using several approaches, including fluorescence‐activated and magnetic‐activated cell sorting or genetic modification andselection using antibiotics. Defining the target cell population and itsphenotype is crucial when using these strategies and planning the next stagesand appropriate time points for seeding cells onto scaffolds and transplantingthem into animal models.

Choosing the Right Scaffold

Scaffolds used in TE are designed to provide cells with a solid 3‐Dframework, allowing cells to attach, migrate, grow, and differentiate, whilemeeting their nutritional and biological needs (Lavik and Langer, 2004).They could be used as a means of delivering cells to the patient or supportcell growth and ex vivo tissue formation before transplantation.

Ideally, scaffolds should imitate the chemical and physical properties ofthe native extracellular matrix and provide the cells with the most ‘‘homey’’environment. They can be made of either natural materials, such as collagen,hydroxyapatite, alginate, and silica; synthetic materials, such as polyesters; orboth. Although natural materials are more biocompatible and recognizableby cells, synthetic materials offer more control over properties such as degra-dation rate, permeability, specific architecture, and mechanical properties.Their architecture can be processed in various techniques, attaining con-trolled porous structures, different‐scale fibrous matrices, and hydrogels,which are either chemically or physically cross‐linked water‐soluble polymersthat can be mixed with cells and potentially injected in a variety of clinicalsituations. In addition, the surface properties of a scaffold have a crucial effecton its biocompatibility. A surface can be modified in a range of physical andchemical ways, including applying biological molecules and binding peptides,such as RGD—an ubiquitous peptide found in many extracellular matrix(ECM) proteins (such as fibronectin and laminin), which binds to cellsthrough integrins. In summary, the properties of a scaffold can be specificallytailored and tightly controlled to meet the many biological and physicalrequirements.

Scaling Up a Regulatable Bioprocess

Large‐scale production of functional tissues to be suitable for biomedicalapplications requires that bioprocesses are scalable, tightly controlled, and

[14] TISSUE ENGINEERING WITH HUMAN ES cells 307

easily regulated. Standard, ‘‘investigative science’’—scale, static culturesystems for growing hESCs—offer limited control over the culture conditions.

Although different hESC lines show diversity in growth kinetics, phe-notype, and differentiation potential, the variety of culture protocols andlaboratory skills result in lack of consistency and different desirable yields.Each step of hESC‐based systems should be ultimately scaled up in aregulated and controlled manner. Perhaps the most challenging processto control is the formation and cultivation of EBs. EB remains the prefera-ble approach in many differentiation systems. Made of a small aggregate ofhESCs grown in suspension, EBs are independently growing and differen-tiating units, heterogeneous in size and in spatial arrangement, and mayaggregate between themselves to form agglomerates. In addition, cellswithin the growing EB respond differently and sometimes unpredictablyto bioactive stimuli such as growth factors and physical cues. Methods toobtain some control over these processes include encapsulation of EBs inbeads made of specific material and in specific size. Ultimately, tissueculture bioreactors, such as spinner flasks and rotating vessels, can bescaled in size to meet specific production needs and offer control overculture conditions such as nutrients and growth factors concentrations,pH, oxygen, and carbon dioxide levels.

hESC‐Derived Connective Tissue Progenitors for TE

Defining the desired target cell population derived from hESCs, onecould aim to either somatic cell type or earlier committed progenitor cell.The objective of the following protocols developed in our laboratory was todirect hESC progeny along the mesenchymal lineage and to achieve aprogenitor cell population that is committed to connective tissue deriva-tives and meets the basic requirements of being viable for cell‐based TEapplications. The general principles and special considerations can beapplied to other cell types and differentiation assays, with appropriatemodifications.

Culture and Maintenance of hESC on MEF Feeders

Preparation of Growth Media

MEF GROWTH MEDIUM

� High‐glucose Dulbecco’s modified eagle’s medium (DMEM),supplemented with:� 10% Fetal bovine serum (FBS)

308 tissue engineering and regenerative medicine [14]

HESC GROWTH MEDIUM

� Knockout DMEM, supplemented with:� 20% Knockout serum replacement� 1 mM Glutamine� 1% Nonessential amino acids� 0.1 mM 2‐mercaptoethanol� 4 ng/ml basic fibroblast growth factor (bFGF)

EB GROWTH MEDIUM

� Knockout DMEM, supplemented with:� 20% FBS� 1 mM glutamine� 1% nonessential amino acids

CTP GROWTH MEDIUM

� Minimum essential medium‐alpha (MEM‐�), supplemented with:� 15% FBS (selected lots)� 50 �g/ml ascorbic acid� 10 mM beta‐glycerophosphate� 10–7 M dexamethasone

Use low protein binding, 22‐�m pore size filters for sterilizing mediacomponents.

Preparation of MEF Feeder Layers

The procedure of deriving MEFs of 13‐day‐old ICR mouse embryosis described elsewhere (Pera et al., 2003). Once derived, MEFs shouldbe subcultured and used for feeder preparation at passages 3–4. Lowerpassage use is possible but wasteful.

1. Inactivate MEFs by incubating with 8 �g/ml mitomycin C for 2 h.2. Wash with Dulbecco’s phosphate‐buffered saline (PBS).3. Harvest cells by trypsinization.4. Plate 40,000 cells/cm2 on gelatin‐pretreated 6‐well plates, in MEF

growth medium.� Overnight incubation is recommended before plating hESCs.� The fresher the better; use plateswithin aweek, keep for 2weeks only.

Starting hESC Culture

1. Make sure MEF feeders are intact and healthy.2. Change feeders’ medium to hESC medium 1 h before plating.

[14] TISSUE ENGINEERING WITH HUMAN ES cells 309

3. Thaw out a frozen hESC vial and resuspend in fresh hESC medium.4. Gently centrifuge and resuspend pellet in final volume not exceeding

1 ml per feeder well.5. Plate on feeders, place inside the incubator, and shake plates for

evenly distributing hESCs on feeders.6. Feed cells with fresh medium on a daily basis.

Passaging hESCs

Frequency of Splitting hESC Cultures Changes BetweenDifferent Cell Lines

Generally, timing and splitting ratio should be determined according tothe following two principles.

Morphology of the Colonies

A high‐quality colony is round, has well‐defined edges, and shows nosigns of differentiation. Cells within the colony are small, with a highnucleus/cytoplasm ratio (Fig. 1).

Poor colonies could be selectively taken out, mechanically, using asterile needle or pipette tip. Alternately, good colonies could be saved inthe same manner.

Density of the Colonies

hESC colonies favor a crowded environment, where they support eachother. At the same time, avoid overly crowded cultures. Grow approximately40–60 medium‐size colonies per well (Fig. 2).

FIG. 1. The appearance of a high‐quality human embryonic stem cell (hESC) colony. Note

round‐shaped, well‐defined edges, and high nucleus/cytoplasm ratio of cells within the colony.

FIG. 2. The optimal density of human embryonic stem cell (hESC) colonies.

310 tissue engineering and regenerative medicine [14]

Passaging Protocol

1. Prepare feeders by removing MEF medium and incubating withhESC medium 1 hbefore hESC seeding.

2. Incubate hESCs with type 0.1% type IV collagenase for 20–40 min.3. Wait for the colonies’ edges to lift off the feeders before continuing.4. Add fresh medium.5. Use a pipette tip and thoroughly scratch out the colonies.6. Collect the scratched material into a conical tube.7. Gently centrifuge and resuspend pellet in final volume not

exceeding 1 ml per feeder well.8. Plate on feeders, place inside the incubator, and shake plates to

evenly distribute the hESCs on feeders.9. Feed cells with fresh hESC medium on a daily basis.

hEB Formation

1. Prepare 60‐mm petri dishes (bacterial grade, non‐tissue culturetreated).

2. Repeat steps 2–5 of passaging protocol.3. Gently centrifuge and resuspend pellet in fresh human EB medium.4. Plate into petri dishes.

hEBs are considered to be independently growing units, thus density ofculture is of lesser importance. In general, plate one well of a six‐well platecontent into one 60‐mm petri dish.

[14] TISSUE ENGINEERING WITH HUMAN ES cells 311

5. Place inside the incubator.6. Feed cells with fresh hEB medium every 3–4 days.

Changing hEB Medium

1. Collect the content of the petri dish into a conical tube.7. Gently centrifuge and resuspend pellet in fresh hEB medium.2. Plate into new petri dishes

Alternately, the following method can be used:

1. Place the petri dish inside the working hood, topless.2. Gently swirl the dish until all hEBs are centered.3. Aspirate off the medium from the edges of the dish.4. Add fresh hEB medium and place inside the incubator.

Derivation and Propagation of Connective Tissue Progenitors

1. Collect 10‐day‐old hEBs growing in suspension into a conical tube.2. Wash with PBS.3. Trypsinize.4. Thoroughly pipette up and down, and pass through a 40‐�m mesh

cell strainer.5. Resuspend in CTP medium, centrifuge, and resuspend again.6. Plate 5 � 104 cells per cm2 on tissue culture‐treated flask.7. On reaching subconfluence, incubate with 0.1% type IV collagenase

for 40–60 min.8. Wash with PBS.9. Harvest cells by trypsinization.

10. Resuspend in fresh CTP medium, centrifuge and resuspend again.11. Split 1:3 onto new flasks.

Choosing the Right Scaffold for Connective Tissue Engineering

In contrast to parenchymal organs, which are mainly cellular and func-tion bymeans of their cells, most of the volume of connective tissues consistsof their functional element—the ECM.

Connective tissue ECMs cope with tensile and compressive mechanicalstresses. Tension is transmitted and resisted by nanoscaled fibrous proteinssuch as collagen and elastin, whereas compression is opposed by water‐soluble proteoglycans, such as chondroitin sulphate (Scott, 2003). The pro-teoglycan part forms a highly hydrated, gel‐like ‘‘ground substance’’ in whichthe fibrous proteins are embedded.

312 tissue engineering and regenerative medicine [14]

So that cells would enjoy the most suitable 3‐D surrounding environ-ment resembling the native ECM, we have postulated that nanoscaledfabricated surface topography of a synthetic scaffold would be best oneto use. Electrospinning is the most common and practical way to fabricatepolymeric nanofiber matrix (reviewed by Ma et al. [2005]). We hypothe-sized that electrospun nanofiber biodegradable polymer scaffolds wouldsupport hESC‐derived CTPs’ organization into complex 3‐D tissues, as weshow in the following.

Scaffold Fabrication and Cell Seeding

Electrospun nanofiber mash scaffolds were made of a 1:1 blend of poly-caprolactone (PCL) and poly(lactic acids) (PLA) by a process describedelsewhere (Ma et al., 2005). The average thickness of the prepared scaffoldwas 500 �m; the fiber diameter ranged between 200–450 nm, with porosity of85%.

For preparation for cell seeding we recommend the following procedure:

1. Cutting scaffold mat into 0.5 � 0.5 cm2 squares, making them fit into24‐well plates.

2. Gas‐sterilizing with ethylene oxide.3. Immersing in 5 M sodium hydroxide and washing in PBS to increase

surface hydrophilicity.

CTP Seeding Protocol

1. Incubate subconfluent CTP cultures with 0.1% type IV collagenasefor 40–60 min.

2. Rinse with PBS.3. Harvest cells by trypsinization.4. Resuspend in fresh CTP medium, centrifuge, and resuspend again.5. Seeding volume should be minimal: count cells and resuspend to

obtain 5 � 105 cells per 10 �l.6. Seed 10 �l on each scaffold and allow cells to attach for 30 min

inside the incubator.7. Gently add fresh medium and change medium every 3–4 days.

Harvesting Samples for Analyses

Immunofluorescent Staining

To avoid misinterpretation of the staining results, assessing autofluorescenceprior to staining is highly recommended.

A B

FIG. 3. CTPs stained with antibody against type I collagen (A) and type II collagen (B).

DAPI was used for nuclear visualization. Note cells embedded in self‐produced extracellular

matrix.

[14] TISSUE ENGINEERING WITH HUMAN ES cells 313

1. Prewash samples with PBS.2. Soak in 4% paraformaldehyde fixative.3. Rinse with PBS.4. Apply primary antibody. Optimal dilution should be calibrated

individually.5. Rinse with PBS.6. Apply appropriate secondary antibody.7. Counterstain nuclei with appropriate dye, such as DAPI.8. Mount cells and view under fluorescent light microscope (Fig. 3A, B).

Electron Microscopy

1. Prewash samples with PBS.2. Soak in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer.3. Gradually dehydrate in ethanol followed by soaking in hexamethyl-

disilazane (HMDS).4. Coat with carbon and view under scanning electron microscope

(Fig. 4A–D).

Histological Analysis

1. Prerinse samples with PBS.2. Soak in 10% natural buffered formalin fixative.3. Gradually dehydrate in ethanol and embed in paraffin.4. Cut sections and stain with hematoxylin‐eosin (H&E).5. View under light microscope (Fig. 5A, B).

BA

FIG. 5. Hematoxylin‐eosin–stained histological images of cross‐sectioned CTPs seeded on

nanofiber scaffolds at low (A) and high (B) power magnifications.

C

D

�1000#0512 � 512

2 mm 4.00 kV 10 mm

NANO5L.TIF

50 mmAcc.V Spot Magn20.0 kV 3.6 1000x

Det WD ExpSE 10.0 7

A B

�3000#0512 � 512

10 mm 5 kV 5 mmF4.TIF

�1000#0512 � 512

20 mm 5 kV 5 mm

F7.TIF

FIG. 4. Scanning electron micrograph of electrospun PCL/PLA nanofiber mash scaffold

alone (A), and of seeded CTPs (B), producing extracellular matrix (C) and eventually forming

3‐D sheetlike tissue completely covering the scaffold (D).

314 tissue engineering and regenerative medicine [14]

[14] TISSUE ENGINEERING WITH HUMAN ES cells 315

References

Amit,M.,Margulets, V., Segev,H., Shariki, K., Laevsky, I., Coleman,R., and Itskovitz‐Eldor, J.(2003). Human feeder layers for human embryonic stem cells. Biol. Reprod. 68,

2150–2156.

Amit, M., Shariki, C., Margulets, V., and Itskovitz‐Eldor, J. (2004). Feeder layer‐ and serum‐free culture of human embryonic stem cells. Biol. Reprod. 70, 837–845.

Dvash, T., and Benvenisty, N. (2004). Human embryonic stem cells as a model for early

human development. Best Pract. Res. Clin. Obstet. Gynaecol. 18, 929–490.Itskovitz‐Eldor, J., Schuldiner, M., Karsenti, D., Eden, A., Yanuka, O., Amit, M., Soreq, H.,

and Benvenisty, N. (2000). Differentiation of human embryonic stem cells into embryoid

bodies compromising the three embryonic germ layers. Mol. Med. 6, 88–95.

Kaufman, D. S., Hanson, E. T., Lewis, R. L., Auerbach, R., and Thomson, J. A. (2001).

Hematopoietic colony‐forming cells derived from human embryonic stem cells. Proc. Natl.

Acad. Sci. USA 98, 10716–10721.

Kehat, I., Kenyagin‐Karsenti, D., Snir, M., Segev, H., Amit, M., Gepstein, A., Livne, E.,

Binah, O., Itskovitz‐Eldor, J., and Gepstein, L. (2001). Human embryonic stem cells can

differentiate into myocytes with structural and functional properties of cardiomyocytes.

J. Clin. Invest. 108, 407–414.

Lavik, E., and Langer, R. (2004). Tissue engineering: Current state and perspectives.

Appl. Microbiol. Biotechnol. 65, 1–8.

Levenberg, S., Golub, J. S., Amit, M., Itskovitz‐Eldor, J., and Langer, R. (2002).

Endothelial cells derived from human embryonic stem cells. Proc. Natl. Acad. Sci. USA

99, 4391–4396.Ludwig, T. E., Levenstein, M. E., Jones, J. M., Berggren, W. T., Mitchen, E. R., Frane, J. L.,

Crandall, L. J., Daigh, C. A., Conard, K. R., Piekarczyk, M. S., Llanas, R. A., and

Thomson, J. A. (2006). Derivation of human embryonic stem cells in defined conditions.

Nat. Biotechnol. 24, 185–187.Ma, Z., Kotaki, M., Inai, R., and Ramakrishna, S. (2005). Potential of nanofiber matrix as

tissue‐engineering scaffolds. Tissue Eng. 11, 101–109.

Mallon, B. S., Park, K. Y., Chen, K. G., Hamilton, R. S., and McKay, R. D. (2006).

Toward xeno‐free culture of human embryonic stem cells. Int. J. Biochem. Cell Biol.

38, 1063–1075.

Pera, M. F., Filipczyk, A. A., Hawes, S. M., and Laslett, A. L. (2003). Isolation,

characterization, and differentiation of human embryonic stem cells. Methods Enzymol.

365, 429–446.

Reubinoff, B. E., Itsykson, P., Turetsky, T., Pera,M. F., Reinhartz, E., Itzik,A., andBen‐Hur, T.

(2001). Neural progenitors from human embryonic stem cells. Nat. Biotechnol.

19, 1134–1140.Scott, J. E. (2003). Elasticity in extracellular matrix ‘shape modules’ of tendon, cartilage, etc.

A sliding proteoglycan‐filament model. J. Physiol. 553, 335–343.

Sharma, B., and Elisseeff, J. H. (2004). Engineering structurally organized cartilage and bone

tissues. Ann. Biomed. Eng. 32, 148–159.

Thomson, J. A., Itskovitz‐Eldor, J., Shapiro, S. S., Waknitz, M. A., Swiergiel, J. J., Marshall,

V. S., and Jones, J. M. (1998). Embryonic stem cell lines derived from human blastocysts.

Science 282, 1145–1147.Vacanti, J. P., and Langer, R. (1999). Tissue engineering: The design and fabrication of living

replacement devices for surgical reconstruction and transplantation. Lancet 354(Suppl. 1),

SI32–SI34.