Towards osteochondral regeneration with human bone marrow derived mesenchymal stromal cells in a functionalized hydrogel system

There is the need of alternative treatment strategies for osteochondral injuries that include a defect of articular cartilage and the underlying bone. Human bone marrow derived mesenchymal stromal cells (BMSCs) due to their ease of isolation and multipotent differentiation capacity have been investigated for a long time as cell source candidate for osteochondral tissue engineering. However, their clinical application has been hampered by several limitations most importantly such as intrinsic tendency to acquire a (pre-) hypertrophic chondrogenic phenotype leading to endochondral ossification in vivo, lack of spatial control of the differentiated cell phenotypes and vast donor-to-donor variability, as well as unpredictability of differentiation outcome potentially due to the crude isolation procedure and lack of selective markers.
Part I of the thesis addressed the optimization of the protocol to generate endochondral bone by BMSCs and the assessment of the formation of bone-cartilage composites by combination of BMSCs with nasal chondrocytes (NCs). To this end, an enzymatically cross-linked and cell-degradable poly(ethylene glycol) (PEG) based hydrogel system served as a scaffolding material. By functionalization of the hydrogel with TGFß3 employing an affinity binding strategy, encapsulated BMSCs were induced to undergo endochondral ossification resulting in the efficient formation of ossicles including a cortical rim and bone marrow upon immediate subcutaneous implantation in immunocompromised mice. This demonstrated that the otherwise needed lengthy in vitro culture step can be circumvented. In bi-layered hydrogels endochondral ossification of BMSCs occurred similarly to the single-layered configuration, while NCs formed cartilaginous tissue, however, unexpectedly acquired hypertrophic features under the influence of the TGFß3 from the BMSC-layer. Replacing TGFß3 with BMP-2 allowed the formation of an osteochondral construct including hyaline cartilage corroborating the potential of our approach to generate cartilage-bone composites. In future, these bi-layered gels need to be tested in an orthotopic model with special focus on how an interface closely resembling the native one can be generated.
Part II of the thesis aimed at elucidating the existence of an expanded BMSC subpopulation with superior chondrogenic differentiation potential. It was hypothesized that retrospective analysis of single clones with high chondrogenic capacity have a different gene expression profile than clones with low capacity and that differential gene expression would guide to prospectively isolate superior chondrogenic potential clones from bulk BMSCs. For one of the tested donors a segregation of clones of high and low CC based on their transcriptomic profile could be observed. Comparison of sorted multiclonal BMSCs based on CD56/NCAM1 - the most promising surface marker identified by the transcriptomic analysis - in chondrogenic in vitro culture assays showed a trend of better chondrogenesis in the CD56+ cells, however, it necessitates confirmation with additional donors. In a further analysis of clones from other donors, intra-donor variability compromised the revelation of transcriptional signatures of clones with high versus low chondrogenic capacity. In future, RNA sequencing as well as cell sorting are required to be performed at earlier time points to exclude confounding effects from extensive cell expansion. Ultimately, identification of a cell subset with superior chondrogenic potential may aid to develop improved BMSC based osteochondral tissue engineering approaches.