Human Bone Marrow Stromal Cells: A Reliable, Challenging Tool for In Vitro Osteogenesis and Bone Tissue Engineering Approaches

Adult human bone marrow stromal cells (hBMSC) are important for many scientific purposes because of their multipotency, availability, and relatively easy handling. They are frequently used to study osteogenesis in vitro. Most commonly, hBMSC are isolated from bone marrow aspirates collected in clinical routine and cultured under the “aspect plastic adherence” without any further selection. Owing to the random donor population, they show a broad heterogeneity. Here, the osteogenic differentiation potential of 531 hBMSC was analyzed. The data were supplied to correlation analysis involving donor age, gender, and body mass index. hBMSC preparations were characterized as follows: (a) how many passages the osteogenic characteristics are stable in and (b) the influence of supplements and culture duration on osteogenic parameters (tissue nonspecific alkaline phosphatase (TNAP), octamer binding transcription factor 4, core-binding factor alpha-1, parathyroid hormone receptor, bone gla protein, and peroxisome proliferator-activated protein γ). The results show that no strong prediction could be made from donor data to the osteogenic differentiation potential; only the ratio of induced TNAP to endogenous TNAP could be a reliable criterion. The results give evidence that hBMSC cultures are stable until passage 7 without substantial loss of differentiation potential and that established differentiation protocols lead to osteoblast-like cells but not to fully authentic osteoblasts.


Isolation and cultivation of human bone marrow stromal cells (hBMSC)
hBMSC were isolated from bone marrow aspirates. Briefly, about 10 ml of one marrow aspirate were diluted 1:5 in phosphate-buffered saline (PBS) (Gibco via ThemoFisher Scientific, Germany). About 20 ml of the diluted bone marrow were applied to a Percoll (Biochrom, Berlin, Germany) density gradient (d = 1.073 g/ml) and centrifuged at 900xg for 30 min at 25 °C. Mononuclear cells in the interface were harvested, filtered through a nylon cell strainer (100 µm, Becton Dickinson, Heidelberg, Germany), and washed twice with 0.5% human serum albumin in PBS. The cells were re-suspended in Dulbecco's modified essential medium (DMEM) (Gibco) containing 10% heat-inactivated fetal calf serum (HI-FCS; Biochrom, Berlin, Germany) and seeded into 75 cm² cell culture flasks (T75, Greiner bio-one, Frickenhausen, Germany) and incubated in 5% CO 2 -atmosphere at 37°C. After 24 h, nonadherent cells were removed by washing with PBS. The medium was changed twice per week. These passage 0 (P0; after isolation) cells need about 3-4 weeks to become confluent.

Gene expression analysis
For analysis of gene expression real time PCR was performed according to following protocol. After an initial activation step for 5 min at 95°C, 50 PCR cycles were run (denaturation at 95°C for 5 sec; annealing and synthesis at 60°C for 10 sec). Primers were constructed by use of Universal Probe Library (Roche, Mannheim Germany) (see Table 1 for detailed information). Identity of the PCR products was verified by sequence analysis. The  From day 4 after plating, the cells were cultured either in BM or in OM/D (osteogenic differentiation medium) and analyzed at day 15 after plating for TNAP activity as described.
The donors were grouped according their age into three subgroups (18-25yrs., 26-25 yrs., and > 35 yrs.) and analysed for statistical differences of endogenous TNAP activity, TNAP activity in OM/D, and inducibility. Significant differences between females and males were analysed by unpaired t-test and indicated with b (p<0.01) and c (p<0.001); the number of analysed donors is given in table 2.

Figure 1S3 Osteogenic differentiation potential of hBMSC (comparison of age groups).
hBMSC in passage 1 were plated in BM (basic medium) onto TCPS (=P2). From day 4 after plating, the cells were cultured either in BM or in OM/D (osteogenic differentiation medium) and analyzed at day 15 after plating for TNAP activity as described. The donors were grouped -37 -according their age into three subgroups (18-25yrs., 26-25 yrs., and > 35 yrs.) and analysed for statistical differences of endogenous TNAP activity, TNAP activity in OM/D, and inducibility. Significant differences between age groups were analysed by one-way ANAOVA with Bonferroni's post-test and indicated with b (p<0.01); the number of analysed donors is given in table 2.

Figure 2S
Correlation of TNAP inducibility to frequency of certain inducibility values.
hBMSC in passage 1 were plated in BM (basic medium) onto TCPS (=P2). From day 4 after plating, the cells were cultured either in BM or in OM/D (osteogenic differentiation medium) and analyzed at day 15 after plating for TNAP activity as described. For the age groups the inducibility of TNAP activity (TNAP activity in OM/D:TNAP activity in BM) was plotted against the frequency of certain inducibility value The data points were analysed by nonlinear curve fit and resulted in a Gaussian distribution. The peak characteristics (x max , y max ) are indicated in the graphs; the number of donors is given in graphs.

Figure 3S
Correlation of donor age with TNAP activity and inducibility. hBMSC in passage 1 were plated in BM (basic medium) onto TCPS (=P2). From day 4 after plating, the cells were cultured either in BM or in OM/D (osteogenic differentiation medium). At day 15 after plating, the cells were analyzed for TNAP activity as described above. Data for endogenous TNAP activity, TNAP activity in OM/D, and TNAP inducibility separately for females and males donor age group were plotted each in one graph. The lines indicate linear curve fit. Regression characteristics (p-value, R-square) and the number of analyzed donors are given in table 2. Figure 4S