15502-m-pleumeekers

of alginate itself. Surprisingly and in contrast to our previous work with MSCs in collagen scaffolds [71] or in pellets without scaffold [218], constructs containing MSCs (both BMSCs and AMSCs) never mineralized in vivo , although signs of endochondral differentiation were observed prior to implantation. The absence of endochondral ossification during in vivo implantation was accompanied by lack of neovascularization or vessel ingrowth within the matrix, which is known to be extremely important for endochondral ossification [219]. We believe that alginate prevented this process, by the fact that endothelial cells lack receptors to bind to alginate polymers, prohibiting neovascularization or vessel ingrowth. [220] Therefore, it seemed that alginate could be an excellent cell-carrying gel for cartilage regeneration although future work needs to clarify which approach is required to exclude alginate calcification after in vivo implantation. In addition to a 3D-culture environment, specific growth and differentiation factors will help to regain and induce a chondrocyte-like phenotype. In vitro , culture-expanded cells of all sources studied failed to differentiate towards the chondrogenic lineage in the absence of TGFβ1, as assessed by an almost negligible deposition of sGAGs and the inferior expression of both ACAN and COL2A1 in alginate constructs. The presence of TGFβ1 induced chondrogenic differentiation in vitro , where ACs exhibit a superior chondrogenic capacity in vitro , compared to the other cell sources. The beneficial effect of culturing with TGFβ1 during in-vivo chondrogenesis was present, although less obvious. Constructs cultured without TGFβ1 increased their production of matrix components after in-vivo implantation, but were not able to reach levels found in constructs cultured with TGFβ1. Even after four passages of culture-expansion, chondrocytes demonstrated some clear subtype specific differences. Firstly, ACs possessed the highest chondrogenic capacity in vitro , but were not able to further increase their cartilage matrix in vivo . The inability of ACs to promote cartilage formation in vivo may be due to the lack of mechanical loading or growth factor stimulation after subcutaneous implantation which may have led to a loss of chondrogenic capacity. ACs, different from the other cell sources, are exposed to mechanical loading within native articular cartilage and unloading is known to induce sGAG-release from the cartilage matrix and to reduce cell proliferation and sGAG-synthesis within the matrix. [221] Secondly, chondrocytes from ear cartilage were able to form an elastin network after subcutaneous implantation in vivo . Elastin was predominantly found in constructs which were cultured with TGFβ1. In-vitro culture did not demonstrate elastin deposition at all, which was in accordance with our previous work. [56] The capability of culture-expanded ECs to produce elastin in vivo suggests that these cells retain their capability to form an elastic cartilage matrix. Both findings - the inability of ACs to promote cartilage formation in vivo without mechanical loading and the ability of cultured expanded ECs to produce elastin - indicate that both cell types preserved their subtype specific phenotype after culture expansion, confirming our previous study where gene expression profiles of culture expanded NCs and ECs displayed clear differences that were related to their developmental origin. [56] Besides chondrocytes, MSCs have been demonstrated to be an attractive cell source for cartilage TE. [62, 64, 66, 222] Although bone marrow offers the most common source of MSCs, adipose tissue has been proven to be an attractive alternative in respect to the abundant and easily accessible pool of MSCs. [67, 68] We have demonstrated that both BMSCs 79 CARTILAGE-FORMING CAPACITY OF SEVERAL CELL SOURCES 4