Characteristic change and loss of in vivo osteogenic abilities of human bone marrow stromal cells during passage.

Although human bone marrow stromal cells (BMSCs) have the ability to form bone when transplanted, the responsible factors for in vivo osteogenic abilities are poorly understood. Here we report conditions that are required for human BMSCs to demonstrate their in vivo osteogenic abilities. BMSCs were obtained from healthy donors and their in vivo osteogenic abilities were analyzed. Transplantation analyses revealed that the passage number and length of osteogenic induction significantly affected ectopic bone formation. Although 2-week induction increased the percentage of success in bone formation compared with the 1-week induction, BMSCs completely lost their in vivo osteogenic ability after passage 4 regardless of the length of osteogenic induction. Despite their in vivo osteogenic ability, no significant difference was observed in alkaline phosphatase activity or gene expression of osteogenic markers between BMSCs at passages 1 and 3. Differences were only observed in in vitro mineralizing abilities. Application of basic fibroblast growth factor helped to maintain the BMSCs in vivo osteogenic ability; basic fibroblast growth factor altered cell growth and expression of HLA-DR. The results strongly suggest that there are several required conditions for human BMSCs to demonstrate their bone-forming capabilities, which should be further investigated and considered when designing a protocol for clinical bone tissue engineering.

Feasibility and efficacy of bone tissue engineering using human bone marrow stromal cells cultivated in serum-free conditions.

Current standard techniques for bone tissue engineering utilize ex vivo expanded osteogenic cells. However, ex vivo expansion requires serum, which may hinder clinical applications. Here, we report the feasibility and efficacy of bone tissue engineering with human bone marrow stromal cells (BMSCs) expanded in serum-free conditions. Bone marrow was aspirated from 4 healthy donors and adherent cells were cultured in either serum-free medium (STEMPRO((R)) MSC SFM) or conventional serum-containing medium (alpha-MEM supplemented with 10% serum). Efficacy of expansion was greater in serum-free medium. Phenotypically, serum-free expanded BMSCs were smaller in cell-size and showed expression of CD105(++) and CD146(dim). After osteogenic induction, serum-free expanded BMSCs showed lower alkaline phosphatase activity. However, they showed higher responsiveness to induction. In vivo bone-forming ability was also confirmed. In conclusion, bone tissue engineering with serum-free expanded BMSCs is feasible and as efficient as that obtained with BMSCs expanded in conventional serum-containing medium.