Repair of Osteochondral Lesions - Engineered Bone: Tissue Engineering Intelligence Unit - page 175

by dynamic compression, 109,110 and (iii) cell-polymer constructs by intermittent hydrostatic
pressure, 111 hydrodynamic shear, 107 and cyclic loading. 112
Despite the great efforts currently dedicated to the development and use of bioreactors for
the engineering of functional cartilage tissue, it remains unclear which specific physical stimu-
lation regimen is required to induce a specific effect on cultured chondrocytes. The scenario is
further complicated by the fact that different scaffolds may transduce similar physical signals in
different ways to cells, making it difficult to compare results obtained by different groups.
Engineering Osteochondral Composite Tissues
As previously described in the section “Autografts and Allografts”, a promising approach
for the correction of large defects involves transplantation of osteochondral units consisting of
an articular cartilage layer and the underlying subchondral bone. In vitro engineering of an
osteochondral composite of predefined size and shape starting from autologous cells would
eliminate the main limitations of this technique, namely (i) the amount of material available,
(ii) the donor site morbidity, and (iii) the difficulty to match the topology of the grafts with the
injured site. In addition, since ingrowth of trabecular bone into an osteoconductive material or
a preformed bone-like tissue is generally highly efficient, the approach would allow easier fixa-
tion and anchorage of the engineered graft into the joint lesion, in a way bypassing the
well-known problem of cartilage-cartilage integration. 113
Osteochondral composite tissues were first generated in vitro by culturing differentiated
chondrocytes onto PGA meshes and periosteal-derived cells onto foams made of a blend of
poly-lactic-coglycolic acid and polyethylene glycol. 113 The two tissue parts were cultured first
separately and then sutured together for different periods of time. The study demonstrated the
feasibility of generating composites of cartilaginous and bone-like tissues in vitro, and pointed
out that the maturation and integration of the two parts can be modulated by the cultivation
time.
Osteochondral composites were also generated using mesenchymal progenitor cells, ex-
ploiting their capacity to differentiate along different mesenchymal lineages. 114 Rat
marrow-derived cells were first “committed” in monolayers with chondrogenic or osteogenic
culture supplements and then seeded respectively on hyaluronic acid-based sponges or porous
ceramic. The two parts were sealed together using fibrin glue and implanted subcutaneously
into nude mice. The two compartments supported a selective differentiation of the mesenchy-
mal progenitor cells and did not separate even when the fibrin sealant was totally degraded.
Lamellar bone was abundant in the ceramic part, but only fibrocartilage and not hyaline carti-
lage filled the voids of the hyaluronic acid sponge. Implantation in a site better resembling the
mechanical and biological environment of a joint was therefore concluded to be essential.
A recent study reported the implantation of engineered osteochondral composites in rabbit
joints. 116 A layer of functional cartilage, generated by bioreactor culture of chondrocytes seeded
onto PGA meshes, was combined with an osteoconductive support made of a ceramic/collagen
sponge (Collagraft). The composites were press-fit in the femoropatellar groove of adult rab-
bits, in the largest ever created defects (7 x 5 x 5 mm) (Fig. 4). The engineered cartilage, which
was 5 fold thicker than the native rabbit cartilage, withstood physiological loading immedi-
ately after implantation and remodelled within 6 months into osteochondral tissue consisting
of a cartilaginous surface and new trabecular bone that was well anchored to the host bone. The
cartilage surface contained evenly distributed GAG and type II collagen and displayed charac-
teristic architectural features, uniform thickness as well as Young's moduli almost normal. Con-
trol defects left empty or treated with cell-free scaffolds healed with irregularly shaped fibrocar-
tilaginous tissue. The study left many open questions, including the mechanism by which the
cartilage was remodelled, the origin of the cells forming the repaired cartilage surface (from the
implant or from the host), and the long-term stability of the engineered cartilage. However, it
demonstrated for the first time that functional engineered cartilage grafts, combined with an
osteoconductive support, might provide a template permitting the orderly repair of very large