current opportunities and challenges in skeletal muscle tissue engineering

JOURNAL OF TISSUE ENGINEERING AND REGENERATIVE MEDICINE R E V I E W A R T I C L EJ Tissue Eng Regen Med 2009; 3: 407–415.Published online 2 July 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/term.190

Current opportunities and challenges in skeletalmuscle tissue engineering

Merel Koning1,2, Martin C. Harmsen2, Marja J. A. van Luyn2 and Paul M. N. Werker1*1Department of Plastic Surgery, University Medical Centre Groningen, University of Groningen, The Netherlands2Department of Pathology and Medical Biology, University Medical Centre Groningen, University of Groningen, The Netherlands

Abstract

The purpose of this article is to give a concise review of the current state of the art in tissueengineering (TE) of skeletal muscle and the opportunities and challenges for future clinicalapplicability. The endogenous progenitor cells of skeletal muscle, i.e. satellite cells, show a highproneness to muscular differentiation, in particular exhibiting the same characteristics and functionas its donor muscle. This suggests that it is important to use an appropriate progenitor cell, especiallyin TE facial muscles, which have a exceptional anatomical and fibre composition compared to otherskeletal muscle. Muscle TE requires an instructive scaffold for structural support and to regulatethe proliferation and differentiation of muscle progenitor cells. Current literature suggests thatoptimal scaffolding could comprise of a fibrin gel and cultured monolayers of muscle satellite cellsobtained through the cell sheet technique. Tissue-engineered muscle constructs require an adequateconnection to the vascular system for efficient transport of oxygen, carbon dioxide, nutrients andwaste products. Finally, functional and clinically applicable muscle constructs depend on adequateneuromuscular junctions with neural cells. To reach this, it seems important to apply optimalelectrical, chemotropic and mechanical stimulation during engineering and discover other factorsthat influence its formation. Thus, in addition to approaches for myogenesis, we discuss the currentstatus of strategies for angiogenesis and neurogenesis of TE muscle constructs and the significancefor future clinical use. Copyright 2009 John Wiley & Sons, Ltd.

Received 26 November 2008; Revised 20 May 2009; Accepted 27 May 2009

Keywords tissue engineering; regenerative medicine; stem cells; satellite cells; myoblasts; skeletalmuscle; facial muscle

Introduction

Paralysis of the facial muscles, which provide facialexpression, may lead to great physical and socialdistress for the patient. Despite advancing knowledgeand improved surgical techniques, the outcome of currenttreatment remains at best suboptimal (Terzis and Noah,1997; Kumar and Hassan, 2002; Werker, 2007).

Tissue engineering (TE) may offer an alternative futuresolution. In general, TE employs progenitor cells and scaf-folds that together generate the appropriate environmentto functionally repair, replace or regenerate lost or dam-aged tissue. Engineered muscle tissue may be customized

*Correspondence to: Paul M. N. Werker, University MedicalCentre Groningen, Hanzeplein 1, PO Box 30001, 9700 RBGroningen, The Netherlands.E-mail: [email protected]

to the individual patient and offers an opportunity forpatients to improve physical and psychological symptomsand reach near to normal mimical function.

Nevertheless, muscle TE has limitations and chal-lenges. Specific myogenic progenitor cells, scaffolds tosupport growth and differentiation of progenitor cells,and bioactive factors are required. Functional skeletalmuscle requires the parallel alignment of muscle fibreswith myosin/actin filaments and acetylcholine receptorsto create direct forces. Besides, the tissue engineeredmuscle must be vascularized and innervated for futureclinical application. Furthermore, the scaffold must bebiocompatible (Vandenburgh, 2002; Bach et al., 2003).

The purpose of this article is to give a concise reviewof the current state of art in TE of skeletal muscle and itssignificance for future clinical use.

Copyright 2009 John Wiley & Sons, Ltd.

408 M. Koning et al.

Facial muscle characteristics

The human face holds 23 paired facial muscles and oneunpaired, the orbicularis oris muscle (May and Klein,1991). Unlike other skeletal muscles, which are spannedout between bones, facial muscles attach at least on oneside to skin. Contraction of the facial muscles gives facialexpression by moving the skin. Each facial muscle is madeup out of 75–150 muscle fibres (Rubin, 1999). They arearranged into parallel bundles, which run from origin toinsertion (Happak et al., 1997).

In general, two types of fibre can be distinguished: typeI, or slow-twitch muscle fibres, that produce large amountsof energy at a slow pace, which allows them to work fora long time without getting exhausted; and type II, orfast-twitch muscle fibres, which produce small amountsof energy very quickly. This enables rapid movements,but causes great fatiguability. Type II fibres are furthersubdivided in types IIA and IIB (Table 1).

Muscles are made up of a mixture of types I andII muscle fibres. Muscles consisting mainly of type Ifibres are red and well perfused because they containmany capillaries; they depend on a rich supply ofoxygenated blood. Type II fibres are primarily dependenton anaerobic metabolism; no rich blood supply is neededand consequently muscles that consist mainly of type IIfibres are more white-coloured (Guyton and Hall, 2000).

The zygomaticus minor has the largest proportion oftype II fibres (89.1%) reported in human skeletal muscle.The zygomaticus major and minor muscles also have amarked predominance of subtype II fibres intermediatebetween types IIA and IIB. The presence of these oxidativetype IIAB fibres indicates a high resistance to fatigue.Besides, facial muscles lack a firm insertion, so staticcontraction and subsequent development of high tensionis prevented, therefore blood circulation is not obstructedand fatigability is further postponed (Stal, 1994).

Progenitor cells

An important prerequisite for engineering skeletal muscleis using suitable, autologous progenitor cells. These

Table 1. Characteristics of skeletal muscle fibre type

Type I Type IIA Type IIX Type IIB

Contractiontime

Slow Moderately fast Fast Very fast

Size of motorneuron

Small Medium Large Very large

Resistance tofatigue

High Fairly high Intermediate Low

Maximumduration ofuse

Hours <30 min <5 min <1 min

Forceproduction

Low Medium High Very high

Capillarydensity

High Intermediate Low Low

Oxidativecapacity

High High Intermediate Low

cells should be cultured, expanded and capable ofdifferentiating into the cell types of muscle tissue.

Mesenchymal stem cells from several different sourcesharbour myogenic potency. Bone marrow-derived mes-enchymal stem cells (MSCs) are able to differentiate intomyoblasts with high efficiency (Dezawa et al., 2005).Myoblasts fuse with each other to form myotubes, whichdifferentiate to muscle fibres. MSCs have a high prolifera-tion capacity and are capable of self-renewal. This rendersthem suitable as progenitor cells for skeletal muscle TE.Furthermore, they can be transplanted autologously andexpanded efficiently in vitro to reach a therapeutic scale(Caplan, 2005; Dezawa et al., 2005). Expansion of har-vested cells is necessary in TE to create a sufficientvolume of skeletal muscle in order to deliver enoughforce to be functional in future clinical application. How-ever, myogenic lineage induction in MSCs is an additionaldifferentiation step. Furthermore, harvesting MSCs frombone marrow is a relatively invasive procedure for thepatient.

Skeletal muscles, on the other hand, harbour their ownendogenous organ-specific mesenchymal stem cells, i.e.satellite cells. These, unlike MSCs, are already proneto myogenic differentiation and therefore seem moresuitable for skeletal muscle TE.

Satellite cells are sequestered between the sarcolemmaand mature muscle fibres and normally do not proliferate.In response to specific local challenges, such as muscledamage, myogenic satellite cells start to proliferate. Theymigrate through the basal lamina sheets to areas ofinjury, differentiate into myoblasts and fuse with pre-existing, damaged fibres or with each other and ultimatelydifferentiate to muscle fibres. Remarkably, if satellite cellsfail to fuse they dedifferentiate back to quiescent satellitecells (DiEdwardo et al., 1999). Myogenic satellite cells canbe harvested from muscle tissue, which is easy accessible,and propagated in vitro.

Unfortunately, the differentiation processes towardsmyofibres still appear difficult to induce and controlin vitro. Therefore, frequently cell lines are used suchas C2C12, which are satellite cells of C3H mice (Denniset al., 2001; Borschel et al., 2004; Levenberg et al., 2005;Riboldi et al., 2005; Huang et al., 2006a; Boontheekulet al., 2007; Matsumoto et al., 2007). Also, satellite cellsharvested from the soleus muscle of rats are used forTE of muscle tissue (Dennis and Kosnik, 2000; Beieret al., 2004; Stern-Straeter et al., 2005; Das et al., 2006;Borschel et al., 2006; Larkin et al., 2006; Bach et al.,2006; Huang et al., 2006b; Boontheekul et al., 2007).Other muscles harvested for TE research include thelatissimus dorsi and rectus femoris of rats (Kamelgeret al., 2004), the flexor digitorum brevis of rats (De Coppiet al., 2005), the tibialis anterior of rats (Dennis et al.,2001; Huang et al., 2005; Boontheekul et al., 2007), theextensor digitorum longus of mice (Dennis et al., 2001),the human masseter (Lewis et al., 2000; Sinanan et al.,2004; Shah et al., 2005; Stern-Straeter et al., 2008; Bradyet al., 2008a) and brachioradialis (Alessandri et al., 2004)

Copyright 2009 John Wiley & Sons, Ltd. J Tissue Eng Regen Med 2009; 3: 407–415.DOI: 10.1002/term

Current opportunities and challenges in skeletal muscle tissue engineering 409

and the iliofibularis from female Xenopus laevis frogs(Jaspers et al., 2006) (Table 2).

Even in vitro, the predetermination of satellite cellsis preserved. In other words, the source of muscle fibretype predisposes satellite cells differentiation: Huang et al.(2006b) showed that characteristics of muscle fibres tissueengineered from soleus muscle satellite cells are differentfrom those obtained from tibialis anterior muscle satellitecells. For example, the contraction/relaxation time isslower in muscle constructs derived of satellite cells ofsoleus muscle compared to muscle constructs derived ofsatellite cells of tibialis anterior muscle. This suggests astrong imprinting of myogenic satellite cells.

McLoon et al. (2007) suggested that facial musclescontain satellite cells with unique abilities, amongst whicha higher quantity and resistance to apoptosis, whichrenders them more favourable for TE than other skeletalmuscle-derived satellite cells.

A difficulty lies in culturing differentiated skeletalmuscle. Sinanan et al. (2004) investigated an approach topurify the population of myogenic satellite cells throughCD56+, which is a cell adhesion molecule (NCAM)expressed in proliferating myoblasts (Illa et al., 1992).Although the resulting cell population has a higherpercentage of desmin (Sinanan et al., 2004), which isan intermediate filament protein, force production of thispopulation was significantly lower than in the mixed andCD56− cell populations (Brady et al., 2008b).

The previously mentioned MSCs harbour a sidepopulation of CD31− CD45− mesenchymal stem cellsthat reside in bone marrow (Gussoni et al., 1999;

Luth et al., 2008), but also in skeletal muscle itself(Asakura et al., 2002; Montarras et al., 2005; Uezumiet al., 2006; Motohashi et al., 2008). A co-culture ofmyogenic satellite cells and this side population of CD31−CD45− mesenchymal stem cells is another approachin stimulating the proliferation and differentiation ofmyogenic satellite cells.

However, hurdles still remain concerning differenti-ation towards functional skeletal muscle before TE isclinically applicable in patients with facial paralysis. Theproneness of satellite cells for myogenic differentiation,the strong imprinting and the unique characteristics offacial muscles suggest that it is important for the func-tional outcome in TE of facial muscles to use appropriateprogenitor cells, such as satellite cells derived of facialmuscle.

Scaffolds

In vivo, the extracellular matrix (ECM) of musclesprovides muscle fibres with the architecture to supportdevelopment and function. Thus, TE of functional skeletalmuscle requires a scaffold to mimic the ECM, tosupport proliferation and differentiation of progenitorcells (Kamelger et al., 2004).

Different types of scaffolds have been developed withregard to their physiochemical features and compositions,as well as their biological characteristics (Figure 1).Although several studies have addressed the developmentof an optimal non-biodegradable scaffold, such as

Table 2. Source of progenitor cells

Reference Species Type Progenitor cells

Caplan, 2005 Human Bone marrow Mesenchymal stem cellDezawa et al., 2005 RatLewis et al., 2000 Human Masseter muscle Satellite cellShah et al., 2005Sinanan et al., 2004Stern-Straeter et al., 2008Alessandri et al., 2004 Human Brachioradialis muscle Satellite cellLewis et al., 2000 C3H mice C2C12 Cell lineBorschel et al., 2004Dennis et al., 2000Huang et al., 2005Levenberg et al., 2005Riboldi et al., 2005Boontheekul et al., 2007Borschel et al., 2004 Rat Soleus muscle Satellite cellLarkin et al., 2006Dennis et al., 2000Huang et al., 2005Das et al., 2006Beier et al., 2004Bach et al., 2003Stern-Staeter et al., 2008Boontheekul et al., 2007Huang et al., 2005 Rat Tibialis anterior muscle Satellite cellDennis et al., 2000Boontheekul et al., 2007Kamelger et al., 2004 Rat Latissimus dorsi muscle Satellite cellKamelger et al., 2004 Rat Rectus femoralis muscle Satellite cellDe Coppi et al., 2005 Rat Flexor digitorum brevis muscle Satellite cellDennis et al., 2000 Mice Extensor digitorum longus muscle Satellite cellJaspers et al., 2006 Frog Iliofibularis muscle Satellite cell



Figure 1. Different designs for scaffolds currently used in muscle TE

phosphate-based glass (Shah et al., 2005), biodegradablescaffolds are preferred because, upon degradation,remodelling to the natural muscular ECM can occur.

Both synthetic and natural scaffolds have beendeveloped. Synthetic biodegradable three-dimensional(3D) scaffolds that hold promise for muscle TE arepolymers made of polylactic-co-glycolic acid (PGA) fibremesh sheets (Saxena et al., 1999). These have been foundto provide both appropriate rigidity and connection.Aligned nanoscale and microscale topographic featuresof a polymer scaffold both cause alignment of myoblastsand cytoskeletal proteins and promote myotube assemblyalong the nanofibres and microgrooves to mimic themyotube organization in muscle fibres. This alignment isan important requirement of functional skeletal muscle.Furthermore, they enhance myotube striation, restrictcell spreading and finally suppress myoblast proliferationduring differentiation and cell fusion. Nanoscale featuresare more efficient in promoting the assembly of longermyotubes than microscale features (Huang et al., 2006a).Parallel alignment can also be induced by applyingmechanical or magnetic strain: muscle fibres reorganizealong the line of force (Sekine et al., 2006; Matsumotoet al., 2007) and through static magnetic fields (Colettiet al., 2007).

Also, in a natural biodegradable 3D scaffold such ascollagen, aligned topographic features cause alignmentof myoblasts and cytoskeletal proteins (Yan et al., 2007;Kroehne et al., 2008). A natural biodegradable 3D scaffoldcan also be made of acellular muscle ECM (Borschelet al., 2004). A drawback of this type of scaffold isits extreme fragility and the associated difficulties ofhandling. Another natural biodegradable 3D scaffoldcan be created by using fibrin. Satellite cells are mixedwith a growth medium, fibrin and thrombin to create afibrin gel (Beier et al., 2004, 2006; Stern-Straeter et al.,2005; Bach et al., 2006; Matsumoto et al., 2007). Muscleprogenitor cells produce their own ECM proteins andover 3–4 weeks degrade and replace the original fibrinmatrix (Huang et al., 2005), which is important because

myoblasts proliferate more extensively in fast-degradinggels (Boontheekul et al., 2007). This process is muchsimilar to wound healing, i.e. repair, in which fibrin formsa temporary scaffold to serve tissue regeneration. Uponcompletion of wound healing, i.e. repair, the fibrin hasbeen replaced by the physiological ECM. Fibrin has theadditional advantage that it binds growth factors, suchas fibroblast growth factor (FGF-2), vascular endothelialgrowth factor (VEGF) and, indirectly, insulin growthfactor (IGF-1), which all augment myogenesis. Myoblastsseeded on fibrin gels have been shown to differentiate intocontracting muscle fibres and to demonstrate a normallength–tension and force–frequency relationship (Huanget al., 2005). In vitro seeding of myogenic progenitor cellson a scaffold has been found to enhance differentiationinto muscle fibres after implantation in vivo and resultsin ingrowth of a nourishing capillary network. This mightbe part of the foreign body reaction that occurs afterimplantation of a skeletal muscle construct in vivo (Saxenaet al., 2001; Kamelger et al., 2004; Luttikhuizen et al.,2006).

An alternative technique which does not require a3D scaffold is cell sheet technology (Tsuda et al., 2004;Nishida et al., 2004; Yang et al., 2005, 2006; Hatakeyamaet al., 2007; Ohashi et al., 2007; Masuda et al., 2008).Although originally developed for myocardial TE purposes(Masuda et al., 2008), the technique has found its way toother fields of TE, such as liver (Ohashi et al., 2007)and cornea (Nishida et al., 2004), and is promisingfor engineering facial muscles. It exercises the useof temperature-responsive 2D scaffolds made out of apolymer, poly(N-isoproplyacrylamide). By increasing thetemperature, the cell layer becomes afloat, which allowsnon-invasive harvesting, i.e. without the use of enzymessuch as trypsin. By means of a stamp, the cultured cells canbe processed as intact sheets, along with their depositedECM. Out of these cell sheets, 3D constructs can be createdby layering these cell sheets (Figure 2). However, thenumber of cell sheet layers is limited because developingmyoblasts are unable to proliferate and differentiate



Figure 2. Concept of in vitro TE of skeletal muscle with three different types of scaffolds

Figure 3. In culturing progenitor cells for skeletal muscle TE, an ideal scaffold can be developed by combining the advantagesof aligned topographic features, fibrin and the cell sheet technique. Vascularized tissue may develop by layering cell sheets on avascular bed and functional muscle tissue might be constructed with appropriate stimuli in vitro, ready for implantation in patients

further than ∼150 µm from a nutrient source (Dennisand Kosnik, 2000).

The mentioned scaffolds all have their own advantages.Parallel alignment can be reached through alignedtopographic features, fibrin promotes cell proliferation,the cell sheet technique offers an excellent architectureof monolayers. Combined, they would seem to form theideal scaffold for skeletal muscle TE (Figure 3).

Innervation

One of the key features for innervation of a muscleconstruct is formation of a motor end plate, the

neuromuscular junction between the motorneuron andevery individual muscle fibre of the construct. It is impor-tant to understand the influence on, and the effect ofchemotropic, electrical and mechanical stimulation oncultured myoblasts with respect to their proliferation anddifferentiation. During in vivo myogenesis acetylcholinereceptors (AChRs) are expressed randomly along thesurface of myoblasts. Stimulation induces the accumu-lation of AChR in the plasma membrane of developingmuscle fibres (Larkin et al., 2006). Whenever a motornerve comes into the vicinity of such an AChR spot, amotor end plate may develop. Continuous communica-tion between the muscle fibres and the nerve is necessaryin maintaining the neuromuscular junction, because these



junctions disappear upon loss of communication (Nevilleet al., 1992).

In co-cultures of myotubes and neural cells,neuromuscular-like junctions appear to form sponta-neously. On muscle cells AChRs accumulate and aresurrounded by neural extensions that harbour neuro-filaments. These nerve–muscle constructs show sponta-neous contractions and can also be stimulated (Wagneret al., 2003). Larkin et al. showed in vitro that stimu-lated nerve–muscle constructs had better contractilitycharacteristics than muscle-only constructs: muscle-onlyconstructs produced a twitch force of 40 µN and a tetanicforce of 95 µN, while stimulated nerve–muscle constructsproduced a twitch force of 100 µN and a tetanic forceof 200 µN (Larkin et al., 2006). Dhawan et al. (2007)showed that in vivo neurotization of skeletal muscleconstructs led to better contractibility ex vivo than non-neurotizated skeletal muscle constructs. This indicatedthat skeletal muscle constructs display a greater maturity,and thereby a higher force production in the presence ofneural cells and neuromuscular junctions.

The most important contraction proteins of musclefibres are desmin and myosin heavy chain (MHC). Besides,myogenic regulator factors (MRFs), such as MyoD andmyogenin, regulate the expression of AChR as part ofmotor endplate development. It has been found that theexpression of desmin, MHC and MRFs can be stronglyinfluenced by chronic electrical stimulation, whichimitates in vivo neuronal activity during myogenesisin vitro (Powell et al., 2002; Pedrotty et al., 2005). Huanget al. (2006b) used this concept on muscle fibre constructsfrom soleus and tibialis anterior muscle satellite cells andfound an increase in force production of 61–80% insoleus muscle constructs, without an increase in totalMHC. They hypothesized that the increase in forceproduction is caused by reorganization of sarcomereswithin muscle fibres, promoted by chronic low-frequencyelectrical stimulation and/or a change in matrix producedby muscle cells, which allowed for better transmission offorce to the anchor materials specifically designed to serveas artificial tendons. Unfortunately, this protocol was notfound to be universally effective, since it did not show thiseffect in the tibialis anterior muscle constructs. This onceagain emphasizes that the characteristics of satellite cellsused to generate muscle constructs, amongst other things,determine the contractile force of a muscle construct.

Although proof-of-principle experiments showed thatelectrical stimulation was essential for the formationand maintenance of functional signalling between motornerve and muscle fibre, other experiments showed that,depending on the nature of this electrical stimulation,the opposite effects can also be achieved (Neville et al.,1992; Pedrotty et al., 2005). When contractions exceeded800/day, loss of muscle mass and force production wasfound within 24 h. In vivo this is within the normal rangefor fast-twitch muscle, but apparently in this culture it isinappropriate (Huang et al., 2006b).

It is a fundamental feature of skeletal muscle, in vivoas well as in vitro, that optimal force production is a

function of stimulation voltage and frequency. Therefore,electrical stimulation needs to be further optimized interms of frequency, pulse width and train durationto minimize tissue damage and maximize beneficialphenotypic alterations.

As said before, the formation of the neuromuscularjunction is also influenced by chemotropic stimulation.Myoblasts appear to undergo dramatic phenotypicchanges upon prolonged culturing and passaging. Thiswas found to be most prominent for the expression ofMRFs such as MyoD and myogenin, and thus for AChRexpression. Myoblasts that underwent low passage hada higher expression level of MyoD, myogenin and AChRcompared to myoblasts that underwent high passage.It also had a diminishing effect on force production(Stern-Straeter et al., 2005). This indicates that frequentpassaging of cells delays synaptic signalling between themotor nerve ending and the muscle fibres.

In conclusion, for the development of a fully functionalneuromuscular junction with neural cells, it seemsimportant to apply the optimal electrical, chemotropicand mechanical stimulation and to discover other factorsthat influence its formation.

Vascularization

In contrast to the limited research into innervatingtissue-engineered skeletal muscle constructs, there hasbeen extensive research into the vascularization of theseconstructs. This is very important, since a major limitingfactor in creating skeletal muscle via TE is that myoblastsare unable to proliferate and differentiate further than150 µm from a nutrient source and oxygen supply (Dennisand Kosnik, 2000). Obviously, most muscles are thickerthan 300 µm; thus, every muscle construct must beconnected to a vascular system for efficient transport ofoxygen, carbon dioxide and nutrients and waste products.

Different options have been studied. First, implantationin vivo of muscle constructs gives the foreign bodyreaction that results in ingrowth of a supportive capillarynetwork that subsequently promotes and maintainsviability (Tanaka et al., 2000; Mian et al., 2000; Saxenaet al., 2001; Beier et al., 2006). Nevertheless, to engineerthicker, functional tissue, an integrated vascular system isrequired.

The second option includes fibrin gel fixed arounda preformed ectopic arteriovenous loop in rats (Bachet al., 2006). If primary expanded myoblasts are injectedinto this prevascularized fibrin gel, vascularized muscletissue is expected to develop. However, Bach et al. showedthat although the cells survived and kept their myogenicphenotype, muscle fibre formation was not yet reached.Differentiation of myoblasts failed in these experiments(Bach et al., 2006), opposite to the experiments of Huanget al., which showed that fibrin gel as a scaffold promotesthe proliferation and differentiation of myoblasts (Huanget al., 2005). It might be that potent stimuli, suchas chemotropic, electrical and mechanical stimulation,



which are necessary for differentiation of myoblasts intomuscle fibres, were lacking in the experiments of Bachet al. (2006).

Third, Borschel et al. (2006) implanted engineeredskeletal muscle constructs around femoral vessels in rats.This led to the formation of vascularized skeletal muscletissue which, when electrically stimulated, produced alongitudinal specific force of 35.4 ± 62.2 N/m2.

Fourth, Levenberg et al. (2005) developed so-calledprevascularized skeletal muscle constructs from culturesof myoblasts, endothelial cells and embryonic fibroblasts,in which enhancement of the vascular network wasperceived. This appears of major interest because thesein vitro prevascularized muscle constructs showed lessapoptosis of cells after implantation in vivo than muscleconstructs that were not prevascularized. Reduction ofapoptosis is correlated to vessels in muscle constructs thatform anastomoses with the host vasculature (Levenberget al., 2005). Further research is needed to establishwhether this model is applicable in humans. The useof a xenogenic culture system (human and animal)in combination with embryonic cells is more thanquestionable, since embryonic tissue might give rise tomalignant proliferation and implantation of animal tissuegives rise to an immunological response.

The cell sheet method mentioned earlier also holdspromise for the concerted vascularization of muscleconstructs. By implanting the maximum of cell sheetsover a vascular system, neovascularization occurs in vivo(Shimizu et al., 2006b). If sufficient neovascularizationoccurs, a new layer of cell sheets can be applied ontop of the previous one, after which neovascularizationof the new layer of cell sheets takes place. Throughthis kind of polysurgery, which applies the use ofrepeated transplantations after neovascularization, thickmuscle tissue with connectable vessels may be produced(Shimizu et al., 2006a, 2006b; Yang et al., 2007; Kuboet al., 2007). This technique has not yet been used forengineering skeletal muscle, and the current method is notpractical in a human situation. Combining the cell sheettechnique with a fibrin coating and aligned topographicalfeatures with a co-culture of myogenic satellite cells andendothelial progenitor cells, this technique can be madeapplicable in a human situation.

Concluding remarks and perspective

Human myogenic satellite cells are at present the firstchoice for skeletal muscle TE purposes. The characteristicsof satellite cells used for TE have a direct effect onthe characteristics and contractility of resulting muscletissue. Further research is required to determine whethersatellite cells from facial muscles are applicable to tissueengineering new facial tissue and whether a combinationwith other mesenchymal stem cells is preferred indeveloping the specific characteristics necessary forclinical application in patients with facial paralysis.

A scaffold should promote the proliferation of satellitecells, differentiation and parallel alignment. Alignedtopographic features, fibrin and the cell sheet techniqueall have their own advantages which, combined, wouldseem to form an ideal scaffold for engineering functionalskeletal muscle.

It seems possible to create a functional neuromuscularjunction through co-culture of myotubes and neuraltissue or by means of electrical stimulation. However,innervating the muscle construct remains a major hurdlefor the clinical applicability of skeletal muscle created viaTE. Vascularization of the muscle construct is essentialto accomplish the volume necessary for the constructto be functional in clinical application. This could bereached through in vivo implantation of a preformedmuscle construct around a vascular pedicle, throughimplantation of myoblasts in a prevascularized fibringel, or through the cell sheet method. These optionsare currently not suitable for human application and anin vitro model seems to have the most potential. Weanticipate that by combining different scaffold techniquesand co-culturing different cell types, further progresscan be made in engineering functional skeletal muscle.Together, these new technologies provide a novel insightin an alternative and improved method for reconstructivesurgery in patients with facial paralysis.

References

Alessandri G, Pagano S, Bez A, et al. 2004; Isolation and cultureof human muscle-derived stem cells able to differentiate intomyogenic and neurogenic cell lineages. Lancet 364: 1872–1883.

Asakura A, Seale P, Girgis-Gabardo A, et al. 2002; Myogenicspecification of side population cells in skeletal muscle. J CellBiol 159: 123–134.

Bach AD, Arkudas A, Tjiawi J, et al. 2006; A new approach to tissueengineering of vascularized skeletal muscle. J Cell Mol Med 10:716–726.

Bach AD, Stem-Straeter J, Beier JP, et al. 2003; Engineering ofmuscle tissue. Clin Plast Surg 30: 589–599.

Beier JP, Kneser U, Stern-Strater J, et al. 2004; Y chromosomedetection of three-dimensional tissue-engineered skeletal muscleconstructs in a syngeneic rat animal model. Cell Transpl 13: 45–53.

Beier JP, Stern-Straeter J, Foerster VT, et al. 2006; Tissueengineering of injectable muscle: three-dimensional myoblast-fibrin injection in the syngeneic rat animal model. Plast ReconstrSurg 118: 1113–1121.

Boontheekul T, Hill EE, Kong HJ, et al. 2007; Regulating myoblastphenotype through controlled gel stiffness and degradation. TissueEng 13: 1431–1442.

Borschel GH, Dennis RG, Kuzon WM Jr. 2004; Contractile skeletalmuscle tissue-engineered on an acellular scaffold. Plast ReconstrSurg 113: 595–602.

Borschel GH, Dow DE, Dennis RG, et al. 2006; Tissue-engineeredaxially vascularized contractile skeletal muscle. Plast Reconstr Surg117: 2235–2242.

Brady MA, Lewis MP, Mudera V. 2008b; Synergy between myogenicand non-myogenic cells in a 3D tissue-engineered craniofacialskeletal muscle construct. J Tissue Eng Regen Med 2: 408–417.

Brady MA, Lewis MP, Mudera V. 2008a; Synergy between myogenicand non-myogenic cells in a 3D tissue-engineered craniofacialskeletal muscle construct. J Tissue Eng Regen Med 2: 408–417.

Caplan AI. 2005; Review: mesenchymal stem cells: cell-basedreconstructive therapy in orthopedics. Tissue Eng 11: 1198–1211.

Coletti D, Teodori L, Albertini MC, et al. 2007; Static magnetic fieldsenhance skeletal muscle differentiation in vitro by improvingmyoblast alignment. Cytometry A 71A: 846–856.



Das M, Gregory CA, Molnar P, et al. 2006; A defined system to allowskeletal muscle differentiation and subsequent integration withsilicon microstructures. Biomaterials 27: 4374–4380.

De Coppi P, Delo D, Farrugia L, et al. 2005; Angiogenic gene-modified muscle cells for enhancement of tissue formation. TissueEng 11: 1034–1044.

Dennis RG, Kosnik PE. 2000; Excitability and isometric contractileproperties of mammalian skeletal muscle constructs engineeredin vitro. In Vitro Cell Dev Biol Anim 36: 327–335.

Dennis RG, Kosnik PE, Gilbert ME, et al. 2001; Excitability andcontractility of skeletal muscle engineered from primary culturesand cell lines. Am J Physiol Cell Physiol 280: C288–295.

Dezawa M, Ishikawa H, Itokazu Y, et al. 2005; Bone marrow stromalcells generate muscle cells and repair muscle degeneration. Science309: 314–317.

Dhawan V, Lytle IF, Dow DE, et al. 2007; Neurotization improvescontractile forces of tissue-engineered skeletal muscle. Tissue Eng13: 2813–2821.

DiEdwardo CA, Petrosko P, Acarturk TO, et al. 1999; Muscle tissueengineering. Clin Plast Surg 26: 647–656.

Gussoni E, Soneoka Y, Strickland CD, et al. 1999; Dystrophinexpression in the mdx mouse restored by stem cell transplantation.Nature 401: 390–394.

Guyton AC, Hall JE. 2000; Textbook of Medical Physiology. Saunders:Philadelphia.

Happak W, Liu J, Burggasser G, et al. 1997; Human facial muscles:dimensions, motor endplate distribution, and presence of musclefibers with multiple motor endplates. Anat Rec 249: 276–284.

Hatakeyama H, Kikuchi A, Yamato M, et al. 2007; Patternedbiofunctional designs of thermoresponsive surfaces forspatiotemporally controlled cell adhesion, growth, and thermallyinduced detachment. Biomaterials 28: 3632–3643.

Huang NF, Patel S, Thakar RG, et al. 2006a; Myotube assembly onnanofibrous and micropatterned polymers. Nano Lett 6: 537–542.

Huang YC, Dennis RG, Baar K. 2006b; Cultured slow vs. fast skeletalmuscle cells differ in physiology and responsiveness to stimulation.Am J Physiol Cell Physiol 291: C11–17.

Huang YC, Dennis RG, Larkin L, et al. 2005; Rapid formation offunctional muscle in vitro using fibrin gels. J Appl Physiol 98:706–713.

Illa I, Leon-Monzon M, Dalakas MC. 1992; Regenerating anddenervated human muscle fibers and satellite cells express neuralcell adhesion molecule recognized by monoclonal antibodies tonatural killer cells. Ann Neurol 31: 46–52.

Jaspers RT, Feenstra HM, van Beek-Harmsen BJ, et al. 2006;Differential effects of muscle fibre length and insulin on muscle-specific mRNA content in isolated mature muscle fibres duringlong-term culture. Cell Tissue Res 326: 795–808.

Kamelger FS, Marksteiner R, Margreiter E, et al. 2004; Acomparative study of three different biomaterials in theengineering of skeletal muscle using a rat animal model.Biomaterials 25: 1649–1655.

Kroehne V, Heschel I, Schugner F, et al. 2008; Use of a novel collagenmatrix with oriented pore structure for muscle cell differentiationin cell culture and in grafts. J Cell Mol Med 12: 1640–1648.

Kubo H, Shimizu T, Yamato M, et al. 2007; Creation of myocardialtubes using cardiomyocyte sheets and an in vitro cell sheet-wrapping device. Biomaterials 28: 3508–3516.

Kumar PA, Hassan KM. 2002; Cross-face nerve graft with free-muscletransfer for reanimation of the paralyzed face: a comparative studyof the single-stage and two-stage procedures. Plast Reconstr Surg109: 451–462.

Larkin LM, Van der Meulen JH, Dennis RG, et al. 2006; Functionalevaluation of nerve–skeletal muscle constructs engineered in vitro.In Vitro Cell Dev Biol Anim 42: 75–82.

Levenberg S, Rouwkema J, Macdonald M, et al. 2005; Engineeringvascularized skeletal muscle tissue. Nat Biotechnol 23: 879–884.

Lewis MP, Tippett HL, Sinanan AC, et al. 2000; Gelatinase-B matrixmetalloproteinase-9; MMP-9 secretion is involved in the migratoryphase of human and murine muscle cell cultures. J Muscle Res CellMotil 21: 223–233.

Luth ES, Jun SJ, Wessen MK, et al. 2008; Bone marrow sidepopulation cells are enriched for progenitors capable of myogenicdifferentiation. J Cell Sci 121: 1426–1434.

Luttikhuizen DT, Harmsen MC, van Luyn MJ. 2006; Cellular andmolecular dynamics in the foreign body reaction. Tissue Eng 12:1955–1970.

Masuda S, Shimizu T, Yamato M, et al. 2008; Cell sheet engineeringfor heart tissue repair. Adv Drug Deliv Rev 60: 277–285.

Matsumoto T, Sasaki J, Alsberg E, et al. 2007; Three-dimensionalcell and tissue patterning in a strained fibrin gel system. PLoS ONE2: e1211.

May M, Klein SR. 1991; Differential diagnosis of facial nerve palsy.Otolaryngol Clin North Am 24: 613–645.

McLoon LK, Thorstenson KM, Solomon A, et al. 2007; Myogenicprecursor cells in craniofacial muscles. Oral Dis 13: 134–140.

Mian R, Morrison WA, Hurley JV, et al. 2000; Formation of newtissue from an arteriovenous loop in the absence of addedextracellular matrix. Tissue Eng 6: 595–603.

Montarras D, Morgan J, Collins C, et al. 2005; Direct isolation ofsatellite cells for skeletal muscle regeneration. Science 309:2064–2067.

Motohashi N, Uezumi A, Yada E, et al. 2008; Muscle CD31− CD45−side population cells promote muscle regeneration by stimulatingproliferation and migration of myoblasts. Am J Pathol 173:781–791.

Neville CM, Schmidt M, Schmidt J. 1992; Response of myogenicdetermination factors to cessation and resumption of electricalactivity in skeletal muscle: a possible role for myogenin indenervation supersensitivity. Cell Mol Neurobiol 12: 511–527.

Nishida K, Yamato M, Hayashida Y, et al. 2004; Cornealreconstruction with tissue-engineered cell sheets composed ofautologous oral mucosal epithelium. N Engl J Med 351:1187–1196.

Ohashi K, Yokoyama T, Yamato M, et al. 2007; Engineeringfunctional two- and three-dimensional liver systems in vivo usinghepatic tissue sheets. Nat Med 13: 880–885.

Pedrotty DM, Koh J, Davis BH, et al. 2005; Engineering skeletalmyoblasts: roles of three-dimensional culture and electricalstimulation. Am J Physiol Heart Circ Physiol 288: H1620–1626.

Powell CA, Smiley BL, Mills J, et al. 2002; Mechanical stimulationimproves tissue-engineered human skeletal muscle. Am J PhysiolCell Physiol 283: C1557–1565.

Riboldi SA, Sampaolesi M, Neuenschwander P, et al. 2005;Electrospun degradable polyesterurethane membranes: potentialscaffolds for skeletal muscle tissue engineering. Biomaterials 26:4606–4615.

Rubin LR. 1999; The anatomy of the nasolabial fold: the keystone ofthe smiling mechanism. Plast Reconstr Surg 103: 687–691.

Saxena AK, Marler J, Benvenuto M, et al. 1999; Skeletal muscletissue engineering using isolated myoblasts on syntheticbiodegradable polymers: preliminary studies. Tissue Eng 5:525–532.

Saxena AK, Willital GH, Vacanti JP. 2001; Vascularized three-dimensional skeletal muscle tissue-engineering. Biomed Mater Eng11: 275–281.

Sekine H, Shimizu T, Yang J, et al. 2006; Pulsatile myocardial tubesfabricated with cell sheet engineering. Circulation 114: I87–193.

Shah R, Sinanan AC, Knowles JC, et al. 2005; Craniofacial muscleengineering using a three-dimensional phosphate glass fibreconstruct. Biomaterials 26: 1497–1505.

Shimizu T, Sekine H, Isoi Y, et al. 2006a; Long-term survival andgrowth of pulsatile myocardial tissue grafts engineered by thelayering of cardiomyocyte sheets. Tissue Eng 12: 499–507.

Shimizu T, Sekine H, Yang J, et al. 2006b; Polysurgery of cell sheetgrafts overcomes diffusion limits to produce thick, vascularizedmyocardial tissues. FASEB J 20: 708–710.

Sinanan AC, Hunt NP, Lewis MP. 2004; Human adult craniofacialmuscle-derived cells: neural-cell adhesion molecule NCAM; CD56-expressing cells appear to contain multipotential stem cells.Biotechnol Appl Biochem 40: 25–34.

Stal P. 1994; Characterization of human oro-facial and masticatorymuscles with respect to fibre types, myosins and capillaries.Morphological, enzyme-histochemical, immuno-histochemicaland biochemical investigations. Swed Dent J Suppl 98: 1–55.

Stern-Straeter J, Bach AD, Stangenberg L, et al. 2005; Impact ofelectrical stimulation on three-dimensional myoblast cultures – areal-time RT–PCR study. J Cell Mol Med 9: 883–892.

Stern-Straeter J, Bran G, Riedel F, et al. 2008; Characterization ofhuman myoblast cultures for tissue engineering. Int J Mol Med 21:49–56.

Tanaka Y, Tsutsumi A, Crowe DM, et al. 2000; Generation of anautologous tissue matrix flap by combining an arteriovenous shuntloop with artificial skin in rats: preliminary report. Br J Plast Surg53: 51–57.

Terzis JK, Noah ME. 1997; Analysis of 100 cases of free-muscletransplantation for facial paralysis. Plast Reconstr Surg 99:1905–1921.



Tsuda Y, Kikuchi A, Yamato M, et al. 2004; Control of cell adhesionand detachment using temperature and thermoresponsivecopolymer grafted culture surfaces. J Biomed Mater Res A 69:70–78.

Uezumi A, Ojima K, Fukada S, et al. 2006; Functional heterogeneityof side population cells in skeletal muscle. Biochem Biophys ResCommun 341: 864–873.

Vandenburgh HH. 2002; Functional assessment and tissue design ofskeletal muscle. Ann N Y Acad Sci 961: 201–202.

Wagner S, Dorchies OM, Stoeckel H, et al. 2003; Functionalmaturation of nicotinic acetylcholine receptors as an indicator ofmurine muscular differentiation in a new nerve–muscle co-culturesystem. Pflugers Arch 447: 14–22.

Werker PM. 2007; [Plastic surgery in patients with facial palsy]. NedTijdschr Geneeskd 151: 287–294.

Yan W, George S, Fotadar U, et al. 2007; Tissue engineering ofskeletal muscle. Tissue Eng 13: 2781–2790.

Yang J, Yamato M, Kohno C, et al. 2005; Cell sheet engineering:recreating tissues without biodegradable scaffolds. Biomaterials26: 6415–6422.

Yang J, Yamato M, Nishida K, et al. 2006; Cell delivery inregenerative medicine: the cell sheet engineering approach.J Control Release 116: 193–203.

Yang J, Yamato M, Shimizu T, et al. 2007; Reconstruction offunctional tissues with cell sheet engineering. Biomaterials 28:5033–5043.


current opportunities and challenges in skeletal muscle tissue engineering

Documents