BMSCs were isolated according to the method described by Panepucci et al (14). In brief, femurs and tibias from SD rats (male, weighing 150±5 g) were removed. Muscle and extraosteal tissue were trimmed under sterilized conditions. Bone marrow cells were flushed and were transferred into culture flasks in 5% CO₂ incubator at 37°C. The culture medium contained 10% fetal calf serum (FCS), (HyClone, Tauranga, New Zealand) and DMEM/F12 (Gibco, Grand Island, NY, USA) containing 100 U/ml penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine (Sigma-Aldrich, St. Louis, MO, USA). Three days later, BMSCs adhered to the bottom of culture plates, and the hematopoietic cells remained suspended in the medium. Fresh medium was changed every 3 days. The sub-confluent cells in the seed cultures were removed from the flasks by 0.25 trypsin (Sigma-Aldrich) treatment 7 days after the initial plating. They were labeled as P1 and continued to culture until P6.

I-OMe AG538 was purchased from Sigma-Aldrich. Stock solution of this drug was prepared in DMSO and stored at −20°C. Working dilutions of all drugs were prepared immediately before use.

To study the inhibition effects of I-OMe AG538 in standard or no-serum medium, 1,000–10,000 cells were plated into 96-well plates in DMEM/F12 plus 10% FCS. After 24 h, medium was replaced by DMEM/F12 plus 10% FCS or without (control) various concentrations of the compound (10 nmol/l-100 µmol/l) for ≤3 days. MTT solution was added to the plate (5 mg/ml) 20 µl/well, then incubated for 4 h and washed. In order to monitor at OD 490 nm, 150 µl DMSO was added to the plate for 10 min. IC₅₀ (drug concentration resulting in 50% inhibition of growth) values of inhibitor was determined using the GraphPad Prism 5 Demo program, GraphPad Software, Inc. (La Jolla, CA, USA).

When BMSCs were cultured at P6, they were already purified. To identify if these cells were BMSCs, cells cultured on 35 mm culture dish were fixed with 4% paraformaldehyde for 20 min. After being washed 3 times with PBS for 5 min, the culture dish was covered with 0.01% Triton X-100 (Gen-View Scientific, Inc., El Monte, CA, USA) for 10 min then were covered with 3% H₂O₂ for 10 min and blocked with normal goat serum for 20 min at room temperature. After removal of serum, rat monoclonal CD29 antibody (dilution, 1:200; cat. no. 121409), rat monoclonal CD44 antibody (dilution, 1:200; cat. no. 203901) and rat monoclonal CD45 antibody (dilution, 1:200; cat. no. 202211) were added followed by HRP goat anti-rat IgG secondary antibody (dilution, 1:500; cat. no. 405405) (all from BioLegend, Inc., San Diego, CA, USA) after washing with PBS. The cells were stained using AEC staining kit and then haematoxylin. PBS was added for the control group.

In the second experiment, to evaluate the ability of IGF-1 induced BMSCs to transdifferentiate into CLCs, 15 ng/ml IGF-1 group and control group containing 10% FCS were used in induction of BMSCs. These cells were observed for morphological changes under an inverted microscope (BX-42; Olympus, Tokyo, Japan). The expression of troponin-T and troponin-I was detected by immunocytochemistry.

In the third experiment, to evaluate the ability of IGF-1 to induce cell recovery from the cytotoxic effects of I-OMe AG538, 2×10⁴/ml cells were plated into 35 mm culture dish in DMEM/F12 plus 10% FCS. After 24 h, 15 ng/ml IGF-1 was added to the medium and 300 nmol/l, replaced by DMEM/F12 plus 10% FBS or the compound (300 nmol/l) every 6 h ≤3 days. After 72 h of treatment, the identification of BMSCs by immunocytochemical stain was performed as previously described. The primary antibody was rat monoclonal troponin-T antibody (dilution, 1:500; cat. no. ab50576) and rat polyclonal troponin-I antibody (dilution, 1:500; cat. no. ab47003) (both from Abcam, Cambridge, UK). PBS was used in the control group.

Constitutive activation of IGF-1R was evaluated on a panel of BMSCs. To evaluate the ability of IGF-1 to induce cell recovery from the cytotoxic effects of I-OMe AG538 toward IGF-1R kinase and signaling, starved BMSCs were treated with doses of 300 nmol/l (corresponding to IC₅₀ value) I-OMe AG538 for 3 h followed by stimulation with IGF-1 (Peprotech, Inc., Rocky Hill, NJ, USA) for 5 to 30 min. In a second experiment, I-OMe AG538 inhibitory effects were followed on IGF-1R-related signaling pathways by exposing BMSCs to 300 nmol/l compound for 1 to 48 h in standard medium. To determine phosphorylation status of Erk1/2 and Akt, two downstream mediators of mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) pathways, cell lysates were prepared with a buffer containing 50 mmol/l MOPS, 320 mmol/l Sucrose, 100 mmol/l Kcl, 0.5 mmol/l MgCl₂, 1 mmol/l DTT, 50 mmol/l NaF, 20 mmol/l NaPPi, 20 mmol/l β-glycerophosphate, 1 mmol/l Na₃VO₄, 1 mmol/l EDTA, and protease inhibitors (0.5 mmol/l phenylmethylsulfonyl fluoride, 0.5 mmol/l leupeptin).

In the third experiment, to evaluate the ability of IGF-1 to induce BMSCs to transdifferentiate into CLCs which had been added with 300 nmol/l I-OMe AG538, we collected BMSCs which were induced by IGF-1 for 3 days. The experiment followed as above. Protein concentration was determined by BCA protein assay (Thermo Labsystems, Santa Rosa, CA, USA) and equivalent amounts of total cell lysate (50 µg) were separated by 4 or 10% SDS-PAGE under denaturing conditions and transferred onto nitrocellulose membrane. Membranes were incubated with 5% bovine serum albumin confining liquid overnight, then with primary antibodies. Rabbit monoclonal phospho-IGF-1R antibody (dilution, 1:1,000; cat. no. 3918S); rabbit monoclonal IGF-1R antibody (dilution: 1:1,000; cat. no. 14534S); rabbit monoclonal phospho-Akt antibody (dilution, 1:1,000; cat. no. 12178S); rabbit monoclonal Akt antibody (dilution, 1:1,000; cat. no. 11848) and rabbit monoclonal Erk1/2 antibody (dilution, 1:1,000; cat. no. 4348S) were purchased from Cell Signaling Technology, Inc. (Beverly, MA, USA). Then the goat anti-rabbit secondary antibody (dilution, 1:2,000; cat. no. ab6721; Abcam, Cambridge, MA, USA) was added. In all experiments, final concentration of DMSO in the medium was <0.001%, and in the present study, it had no effect on cell growth inhibition. This method was described by Scotlandi et al (15).

BMSCs were successfully expanded and used for subsequent experiments. At first, single cells of spindle morphology were observed 3–4 days after the initial seeding. The morphologic appearance of the cell population was fairly homogeneous (Fig. 1A). About 1 week later, most cells appeared as fibroblasts with spindle or shuttle shape. The colonies continuously grew, reaching 80–90% confluence in 2 weeks. However, there were some blood cells in the view. When they were labeled at P6, cells were already purified and there were no blood cells in the view and the confluence was whirlpool-like (Fig. 1B).

All P6 BMSCs express CD29 (Fig. 2C) and CD44 (Fig. 2D), but there was no staining for CD45 (Fig. 2B). This demonstrated that these cells were BMSCs but haemopoietic stem cells. We confirmed that the method to isolate BMSCs was feasible and highly purified cells could be obtained with this method.

After BMSCs were exposed to 15 ng/ml of IGF-1 for 21 days, the morphology of BMSCs did not change. The expression of troponin-T and troponin-I were significantly higher in the 14-day group than control group (Fig. 3) and the levels of pIGF-1R were higher in the 14-day group than other groups (Fig. 4).

Fig. 5 demonstrates that similar inhibitory effects were obtained in BMSCs with various concentrations using I-OMe AG538. Growth inhibitory activity of the compound was maintained for ≥72 h after its removal (50% of growth inhibition with the dose of 331.7±2.3 nmol/l; P<0.05) (Fig. 5). To confirm the inhibitory activity of I-OMe AG538 toward IGF-1R kinase and signaling, starved BMSCs were treated with doses of 300 nmol/l for 3 h followed by stimulation with IGF-1 for 5 to 30 min. Fig. 6 shows that both IGF-1R autophosphorylation and the two major IGF-1R-related intracellular signaling pathways, MAPK and PI3K pathways, were inhibited by I-OMe AG538.

To confirm the inhibitory activity of I-OMe AG538 and whether it blocks the differentiation of BMSCs into CLCs induced by IGF-1, BMSCs were treated with 15 ng/ml IGF-1 and 300 nmol/l or 10 µmol/l I-OMe AG538. Fig. 7 shows that the expression of troponin-T and troponin-I were inhibited by 300 nmol/l I-OMe AG538.

A time course evaluation of inhibitory effects of 300 nmol/l I-OMe AG538 on MAPK and PI3K signaling pathways in standard medium, however, revealed transient effects on MAPK and PI3K pathway, particularly for the dose of 300 nmol/l (Fig. 8).