All experimental procedures were performed in accordance with the guidelines established by the Animal Studies Committee at Iwate Medical University (Iwate, Japan). A total of four GFP-transgenic mice (24) were obtained from the Center for in vivo Science, Iwate Medical University (Iwate, Japan). The mice were sacrificed by excessive inhalation of CO₂. Cells were flushed from the tibia of three-week-old GFP-transgenic mice with phosphate-buffered saline (PBS) containing 0.5% fetal bovine serum (FBS; PAA Laboratories, GE Healthcare, Piscataway, NJ, USA) and 2 mM EDTA, and then seeded into plastic cell culture dishes (Nunc; Thermo Fisher Scientific, Waltham, MA, USA) with Dulbecco's modified Eagle's medium (DMEM; Sigma-Aldrich, St. Louis, MO, USA) containing 10% FBS. The cells were cultured for 1 week under hypoxic conditions (5% O₂, 5% CO₂ and 90% N₂). Cells were re-plated upon reaching 80% confluence.

The expanded cells were transfected with pBABE-neo-hTERT and pBABE-pur-SV40LT plasmids encoding neomycin and puromycin resistance (provided by Addgene, Cambridge, MA, USA) with Lipofectamine LTX (Invitrogen Life Technologies, Carlsbad, CA, USA) according to manufacturer instructions. Cells were then incubated in DMEM containing 10% FBS, 150 µg/ml G418 (Gibco-BRL, Thermo Fisher Scientific) or 1 µg/ml puromycin (Gibco-BRL) under hypoxic conditions for 12–15 days. The surviving cells were trypsinized and allowed to grow in 90-mm culture dishes.

Single-cell clones were obtained using the limited dilution method. After hTERT and SV40LT transfection and selection with G418 and puromycin, the surviving cells were seeded on a 96-well plate (Nunc) at 0.5% cells per well, and then cultured under hypoxic conditions. After 10 days, the cells were sub-cultured in 24-well plates (Nunc). This was repeated until confluence was reached at 20 days after single cell cloning. Population doubling (PD) was defined as the number of doublings required for a single cell to reach confluence in a 60-mm culture dish (Nunc) under hypoxic conditions. PD was estimated for clones SG-2, -3, -5 and -6.

Telomerase activity in bone marrow-derived cell lines was assayed by the stretch PCR method using the Quantitative Telomerase Detection kit (Allied Biotech, Vallejo, CA, USA) according to manufacturer's instructions. The PCR mixture contained QTD premixed buffer and SYBR green one dye (cat. no. MT3010; Allied Biotech, Inc., St. Benicia, CA, USA). Quantification of telomerase activity was performed under the following amplification conditions using a Thermal Cycler Dice Real Time system (Takara Bio, Otsu, Japan) according to manufacturer's instructions: 25°C for 20 min, initial activation at 95°C for 10 min, denaturation at 90°C for 30 s, annealing at 60°C for 30 s, and a final extension of 40 cycles at 72°C for 30 s. The PCR products were separated by 10–20% polyacrylamide gel electrophoresis and stained with ethidium bromide.

Bone marrow-derived cell lines were seeded onto eight-well culture slides (BD Biosciences, Franklin Lakes, NJ, USA). After 24 h, the cells were fixed with 4% paraformaldehyde and washed five times with PBS. For detection of SV40LT, cells were incubated with mouse monoclonal anti-SV40LT (1:100; cat. no. ab16879; Abcam, Cambridge, UK) antibody for 1 h at room temperature. The cells were then incubated with Alexa Fluor 594 goat polyclonal anti-mouse secondary antibodies (1:500; cat. no. A11005; Thermo Fisher Scientific) and DAPI (1:500; cat. no. D9542; Sigma-Aldrich) for 30 min at room temperature. Fluorescence was examined by using a fluorescence microscope (Olympus ix70; Olympus Corporation, Tokyo, Japan).

A total of 1.0×10⁵ bone marrow-derived cell lines (SG-2, -3, -5 and -6) were suspended in PBS containing 0.5 FBS and 2 mM EDTA and incubated with phycoerythrin-conjugated monoclonal anti-mouse Sca-1 (1:10; cat. no. 130-093-224), monoclonal anti-mouse CD44 (1:10; cat. no. 130-096-838), monoclonal anti-mouse CD11b (1:10; cat. no. 130-091-240) or monoclonal anti-mouse CD45 (1:10; cat. no. 130-091-610) antibody for 1 h at 4°C in the dark. All antibodies were purchased from Miltenyi Biotec (Bergisch Gladbach, Germany). Acquisition was performed using an EPICS XL ADC system (Beckman Coulter, Brea, CA, USA).

In vitro differentiation was performed according to a previous study by our group (25). To induce osteogenic differentiation, confluent cells were incubated in osteogenic differentiation medium (ODM) under hypoxic conditions for two weeks. Bone matrix mineralization was evaluated by Alizarin red S (Sigma-Aldrich) staining. To induce adipogenic differentiation, cells were cultured to near confluence and cultured in adipogenic differentiation medium (ADM) under hypoxic conditions for two weeks. At the end of the differentiation period, lipid droplets were stained with Oil Red O (Sigma-Aldrich).

Gene expression profiling was performed using a PrimerArray of mouse cytokine-cytokine receptor interaction (Takara Bio) in combination with a Thermal Cycler Dice Real Time System (Takara Bio) according to manufacturer's instructions. This PrimerArray is a set of real-time reverse transcription-polymerase chain reaction (RT-PCR) primers used for the analysis of RNA expression. The array contains a mixture of 96 primer pairs for 88 target genes and eight housekeeping genes. Quantification of gene expression was performed using a PrimerArray Analysis Tool version 2.0 (Takara Bio).

SG-2, -3 and -5 cells were serum-starved overnight and stimulated with 50 ng/ml BMP-2 (R&D Systems, Minneapolis, MN, USA) or 50 ng/ml TGF-β1 (R&D Systems). The cells were washed twice with ice-cold PBS and then lysed in radioimmunoprecipitation assay buffer (50 mM Tris-HCl, pH 7.2, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, and 0.1% SDS) containing protease and phosphatase inhibitor cocktails (Sigma-Aldrich). The protein content was quantified using the bicinchoninic acid method (Pierce; Thermo Fisher Scientific). Samples containing equal amounts of protein were separated by 12.5% SDS-PAGE and transferred onto a polyvinylidene difluoride membrane (Merck Millipore, Darmstadt, Germany). After blocking with 5% nonfat dry milk in 50 mM Tris-HCl, pH 7.2, containing 150 mM NaCl and 0.1% Tween-20 for 2 h at room temperature, the membrane was incubated with primary rabbit monoclonal anti-phospho-Smad 5 (1:1,000; cat. no. ab92698; Abcam), anti-Smad 1/5/8 (1:1,000, cat. no. 12656; Cell Signaling Technology, Danvers, MA, USA), rabbit monoclonal anti-phospho-Smad 2 (1:1,000; cat. no. 04-953; Merck Millipore), or anti-Smad 2/3 (1:1,000; cat. no. 610843; BD Biosciences) antibody overnight at 4°C, with mouse monoclonal anti-β-actin (1:1,000; cat. no. sc-47778; Santa Cruz Biotechnology, Dallas, TX, USA) antibody as a loading control. The blots were incubated with alkaline phosphatase-conjugated secondary antibody and developed using the 5-bromo-4-chloro-3′-indolyphosphate/nitro-blue tetrazolium membrane phosphatase substrate system (cat. no. 50-81-00; Kirkegaard & Perry Laboratories, Gaithersburg, MD, USA).

SG-2, -3 and -5 cells were stimulated with TGF-β1 (1–50 ng/ml; R&D Systems) for 24 h. Total RNA was isolated using ISOGEN reagent (Nippongene, Tokyo, Japan) according to the manufacturer's instructions. First-strand cDNA was synthesized from total RNA with the PrimeScript RT Reagent kit (Takara Bio). RT-qPCR was performed on a Thermal Cycler Dice Real Time System (Takara Bio) with SYBR Premix Ex Taq II (Takara Bio). Expression of CCAAT/enhancer binding protein-β (C/EBPβ) was normalized to β-actin and relative expression levels were calculated as a fold-increase or -decrease relative to the control. Transcripts were detected with primers (designed using a Perfect Real Time Support system; Takara Bio) for C/EBPβ (sense, 5′-GACAAGCTGAGCGACGAGTA-3′; and anti-sense, 5′-AGCTGCTCCACCTTCTTCTG-3′) and β-actin (sense, 5′-CATCCGTAAAGACCTCTATGCCAAC-3′ and anti-sense, 5′-ATGGAGCCACCGATCCACA-3′). For each PCR run, cDNA derived from 50 ng total RNA was used. Following initial denaturation at 95°C for 30 s, a two-step cycle procedure was used (denaturation at 95°C for 5 s and annealing and extension at 60°C for 30 s) for 40 cycles. The relative mRNA expression levels in each sample were calculated using the 2^−ΔΔCT method (26).

All experiments were repeated at least three times. Representative images or data are shown. Values are expressed as the mean ± standard deviation. Differences between control and test samples were analyzed using paired two-tailed Student's t-tests. P<0.05 was considered to indicate a statistically significant difference between values.

Bone marrow cells were flushed from the tibia of GFP mice and cultured under hypoxic conditions. Adherent cells were transformed with hTERT and SV40LT vectors, yielding four single cell-derived cell lines SG-2, -3, -5 and -6. As shown in Fig. 1, the cell lines exhibited fibroblastic morphology (Fig. 1, upper panel) and GFP fluorescence (Fig. 1, lower panel). All cell lines exhibited nuclear SV40LT expression (Fig. 2A). At PD 20, a stretch PCR assay indicated telomerase activity in all cell lines, but not in the negative control (Fig. 2B). Thus, the bone marrow-derived cell lines exhibited telomerase activity and SV40LT expression. All cell lines grew at a similar rate of ~1 PD every two days (Fig. 2C). The cells divided at least 60 times and were passaged >30 times, thus demonstrating successful immortalization.

To determine their MSC character, the SG-2, -3, 5, and -6 cell lines were analyzed for the expression of mouse MSC markers and differentiation potential. Sca-1 is the most reliable MSC marker in mice and was strongly expressed in SG-2, -3 and -5 cells, but only weakly expressed in SG-6 cells (Fig. 3). CD44 was detected at similar levels in all cell lines, although the hematopoietic stem cell markers CD11b and CD45 were not detected. The expression patterns of MSC markers suggested that the SG-2, -3 and -5 cell lines were MSCs. Next, the present study evaluated the osteogenic and adipogenic differentiation potentials of these cell lines. Bone matrix mineralization indicated by Alizarin red staining was highest in SG-5 cells (Fig. 4A). Although miner-alization was observed in SG-2 cells, it was markedly lower than that in SG-5 and was not detected in SG-3 and SG-6 cells. Thus, SG-2 and -5 cells retained their osteogenic differentiation potential, although at different levels of efficiency. Lipid droplet formation indicated by Oil Red O staining was more intense in SG-2 cells than in SG-3 and -5 cells (Fig. 4B). Thus, SG-2, -3, and -5 retained their adipogenic differentiation potential to various degrees.

In order to identify the expression profiles of cytokines and cytokine receptors in SG cells, the present study performed PrimerArray analyses and compared the results derived from SG-2, -3 and -5 cells. Differentially expressed genes are shown in Table I. BMP receptor 1B (Bmpr1b) was most highly expressed in SG-3 cells, while TGF-β receptor II (Tgfbr2) was most highly expressed in SG-3 and -5 cells, indicating differential sensitivities to BMP-2 and TGF-β. Smad 5, which is a major signaling factor activated by BMP, was most significantly phosphorylated in BMP 2-stimulated SG-3 cells (Fig. 5A). In addition, osteogenic differentiation in SG-3 cells was induced by stimulation with BMP-2 in a dose-dependent manner (Fig. 5B), suggesting these cells retained the capacity to differentiate into adipogenic (Fig. 4B) and osteogenic lineages. However, phosphorylation of Smad 2, which is activated by TGF-β, was highest in the MSC-like SG-2 cells in response to TGF-β1 (Fig. 6A). Of note, mRNA expression of C/EBPβ, which is an immune- and inflammatory response-associated as well as Smad-interacting transcription factor, was induced in SG-2 cells by TGF-β1 in a dose-dependent manner (Fig. 6B). Thus, the SG-3 cells were BMP-responsive and the SG-2 cells were TGF-β-responsive, while SG-5 cells were BMP/TGF-β-unresponsive MSCs.