In order to assess the

In order to assess the cellular proliferation potential colony forming efficiency was determined. As we decided to apply a low suction vacuum pressure in order to minimize potential discomfort during bone marrow harvest under local anesthesia in the control individuals the achieved mean cell yields ranging from 26 to 36CFU per 106 plated pak 4 into 25cm2 flasks (Fig. 2B) were in the lower range compared with other studies (Castro-Malaspina et al., 1980; Kasten et al., 2008; Mansilla et al., 2006). However, as the harvest protocol was standardized and applied to all probands of both groups, cell numbers of BMSCs from healthy and OA donors after 14days of cultivation were comparable (Fig. 2C). We noticed a trend towards higher proliferation rate in the OA study group. Singh et al. demonstrated increased proliferative capacity of osteophyte-derived BMSCs compared to normal BMSCs (Singh et al., 2008). Jones et al. examined CD271+ multipotential stromal cells from healthy and osteoarthritic trabecular bone and associated, contrary to this study, decreased cell proliferation but unchanged osteogenic potential with disease (Jones et al., 2010). No significant differences between BMSCs from OA versus healthy donors were observed concerning gene expression of ALP, BSP, and RUNX2 (Fig. 3A). ALP activity was not significantly stimulated by donor-specific OA comorbidity (Fig. 3B). These findings are in line with a study by Oreffo et al. who reported no significant differences in ALP activity in OA-BMSCs (Oreffo et al., 1998). In contrast to Morimoto et al. who demonstrated increased ALP activity levels in BMSCs derived from donors with rheumatoid arthritis upon osteogenic cultivation we did not observe disease-specific changes in osteogenic differentiation markers after cultivation with and without osteogenic supplements (Morimoto et al., 2009). We found no significant differences in the expression of COL2A1 and COL10A1 genes in OA- as opposed to control BMSCs. In contrast, OA-BMSCs showed significantly enhanced SOX9 gene expression levels (p<0.005) as well as a trend towards increased COL10A1 and COL2A1 expression after 14days of cultivation (Fig. 4A). Pathological changes in cartilage tissue hemostasis denote the first stage of a complex pathomechanism resulting in OA. In the context of clinical cell therapeutic strategies exaggerated chondrogenic differentiation susceptibility as observed in the present study may hamper adequate bone regeneration. Furthermore, an increased expression level of chondrogenic markers bears the potential of cellular hypertrophy thereby excluding the use of BMSCs for chondrogenic tissue regeneration applications (Mueller and Tuan, 2008; Pelttari et al., 2008; Yoneno et al., 2005). Coleman et al. described a concurrent increase in ALP, collagen I, collagen X, and osteopontin after application of growth differentiation factor-5 indicating a transition of BMSC pellet cultures to hypertrophy as a precursor to endochondral ossification (Coleman et al., 2013). Murphy et al. detected a reduced chondrogenic and adipogenic activity of mesenchymal stem cells from patients with advanced OA (Murphy et al., 2002). As opposed to our findings, Scharstuhl et al. found an adequate chondrogenic differentiation response in BMSCs irrespective of donor age and OA comorbidity (Scharstuhl et al., 2007). In contrast to the present study these authors harvested BMSCs from the femoral BM compartment of 98 patients with different OA etiologies, e.g. primary age-related OA, post-traumatic, joint dysplasia, rheumatoid arthritis, and aseptic osteonecrosis. However, a healthy positive control group, as is provided in the present study, is missing in these studies. Furthermore, the adipogenic and osteogenic differentiation capacity was not investigated by the authors. Instead, BMSC phenotype was only assessed in 25 donors by using flow cytometry and RT-PCR was only partially implemented. Dudics et al. reported of BMSCs from patients with rheumatoid arthritis or OA possessing similar chondrogenic potential as BMSCs isolated from healthy donors (Dudics et al., 2009). Compared to the present study the donor number was lower and only COL2A1 was analyzed by RT-PCR as a chondrogenic marker gene, whereas the present study compares proliferation, differentiation, and additionally surface marker expression in OA and age-matched healthy donors in a larger study population by use of several marker genes. Hence, contrary to Dudics and co-workers, our data suggest an impact of OA on the chondrogenic differentiation potential of BMSCs. The increase of SOX9 gene expression in BMSCs of OA donors indicates activation of the chondrogenic differentiation direction being considered favorable with regard to OA-BMSC application in cartilage regeneration therapy (Bernstein et al., 2010). To further enlighten the role of OA-BMSC in cartilage regeneration the evaluation of relevant signal transduction pathways is necessary. PPARγ and FABP4 were chosen as early and late adipogenic marker genes, respectively (Chamberlain et al., 2007; Dani, 1999; Gregoire et al., 1998). Justesen et al. found a non-age related adipogenic differentiation capacity and PPARγ gene expression of BMSCs (Justesen et al., 2001). Our findings are in line with this indicating an OA-independent PPARγ and FABP4 gene expression profile (Fig. 5A). In addition, lipid-rich vacuoles stained with Oil Red O were detectable in both study groups in similar intensity (Fig. 5B, C).