Tissue engineering - Biomedical Engineering Desk Reference - page 631

myocardial infarction, intracoronary infusion of autolo-
gous progenitor cells beneficially affected postinfarction
left-ventricle remodeling processes ( Assmus et al. , 2002 ).
An alternative approach to cell grafting techniques is
the generation of cardiac tissue grafts in vitro and
implanting them as spontaneously and coherently
contracting tissues. As a model system, rat neonatal or
embryonic chicken cardiomyocytes may be seeded on
three-dimensional polymeric scaffolds ( Carrier et al. ,
1999 ) or collagen disks formed as a sponge ( Radisic et al. ,
2003 ) or by layering cell sheets three-dimensionally
( Shimizu et al. , 2002 ). The latter two approaches are
suitable for producing thicker cardiac tissue with more
evenly distributed cells at a higher density. A principally
different approach to generate engineered heart tissue
was developed by Eschenhagen and colleagues. Neonatal
or embryonic cardiomyocytes were mixed with freshly
neutralized collagen I and cast into a cylindrical mold.
After a few days the tissue patches were transferred to
a stretching device, which induced hypertrophic growth
and increased cell differentiation ( Eschenhagen et al. ,
2002b , 1997; Zimmermann et al. , 2000 ). Interestingly,
the response to isoprenalin of stretched tissue was much
more pronounced than in unstretched tissue ( Eschen-
hagen et al. , 2002b ; Zimmermann et al. , 2002 ).

To overcome the limitations just mentioned, tissue
engineering procedures could lead to completely bi-
ological vascular grafts. In fact, there have already been
case reports regarding first human pediatric applications
of tissue-engineered large-diameter vascular grafts ( Naito
et al. , 2003 ; Shin'oka et al. , 2001). As to small-caliber
grafts, there are three principal approaches involving
(1) synthetic biodegradable scaffolds, (2) biological scaf-
folds, and (3) completely autologous methods.
1. Niklason et al. have shown in animal models that by
utilizing flow bioreactors to condition biodegradable
polymers loaded with vascular cells, it is possible to
generate arbitrary lengths of functional vascular
grafts with significant extracellular matrix
production, contractile responses to pharmacological
agents, and tolerance of supraphysiologic burst
pressures (Mitchell and Niklason, 2003; Niklason
et al. , 2001 , 1999). Similar in vitro experiments
based on human vascular-derived cells seeded on
PGA/PHA copolymers demonstrated the feasibility
of viable, surgically implantable human small-caliber
vascular grafts and the important effect of a
''biomimetic'' in vitro environment on tissue matura
tion ( Hoerstrup et al. , 2001 ).
2. A different approach to tissue engineering of vascular
grafts comprises the use of decellularized natural ma-
trices as initially introduced by Rosenberg et al. (1996).
Histological examination of chemically decellularized
carotid arteries revealed well-preserved structural
matrix proteins. This provides an acellular scaffold that
can be successfully repopulated in vitro prior to
implantation ( Teebken et al. ,2000 ). Such scaffolds
have also been shown to be repopulated in vivo ( Bader
et al. , 2000 ). Recently, successful utilization of
endothelial precursor cells for tissue engineering of
vascular grafts based on decellularized matrices has
been demonstrated ( Kaushal et al. , 2001 ).
3. L'Heureux et al. cultured and conditioned sheets
of vascular smooth muscle cells and their native
extracellular matrix without any scaffold material in
a flow system. Subsequently these sheets were placed
around a tubular support device and after maturation
the tubular support was removed and endothelial cells
were seeded into the lumen. Thereby a complete
scaffold-free vessel was created with a functional
endothelial layer and a burst strength of more than
2000 mm Hg ( L'Heureux et al. , 1998 ).
Angiogenesis (the formation of new blood vessels) is
essential for growth, tissue repair, and wound healing.
Therefore, many tissue-engineering concepts involve an-
giogenesis for the vascularization of the newly generated
tissues. Unfortunately, so far advances have been com-
promised by the inability to vascularize thick, complex

Blood vessels
Peripheral vascular disease represents a growing health
and socioeconomic burden in most developed countries
( Ounpuu et al. , 2000 ). Today, artificial prostheses made
of expanded polytetrafluoroethylene (ePTFE) and
poly(ethylene terephthalate) (PET, Dacron) are the most
widely used synthetic materials. Although successful
in large diameter ( > 5-mm) high-flow vessels, in low-flow
or smaller diameter sites they are compromised by
thrombogenicity and compliance mismatch ( Edelman,
1999 ). To circumvent these problems numerous modi-
fications and techniques to enhance hematocompatibility
and graft patency have been evaluated both in vitro and
in vivo. These include chemical modifications, coatings
( Gosselin et al. , 1996 ; Ye et al. , 2000 ), and surface
seeding with endothelial cells ( Pasic et al. , 1996 ; Zilla
et al. , 1999 ). In vitro endothelialization of ePTFE grafts
may result in patency rates comparable to state-of-the-
art veinous autografts (Meinhart et al. , 1997). Polymer
surface modifications involving protein adsorption may
also be desirable. Unfortunately, materials that promote
endothelial cell attachment often simultaneously pro-
mote attachment of platelets and smooth muscle cells
associated with the adverse side effects of clotting and
pseudointimal thickening. A possible solution has been
demonstrated with polymers containing adhesion mole-
cules (ligands) specific for endothelial cells ( Hubbell
et al. , 1991 ).