Male Sprague-Dawley rats aged 6 weeks were obtained from the Department of Experimental Animals of the Fudan University (Shanghai, China). The rats were housed at 20–26°C with 16-h light and 8-h dark cycle, and provided ad libitum food and water. All the animal handling and experimental procedures were approved by the Committee for Ethical Use of Experimental Animals at Fudan University (Shanghai, China), and the ethical approval registration number for animal study was 20150559A183.

The rats were locally irradiated with 2 Gy by an Animal X-ray Irradiator (X-RAD320, PXi, USA) at a rate of 185.5 cGy/min after being anesthetized. The irradiation site was restricted in the region between distal femur and proximal tibia, which was sensitive to irradiation (20). In addition, the irradiation area is limited with the collimators, only the left hind limbs of rats were exposed to irradiation directly, and the remaining parts of the rats (including opposite right hind limbs) were covered by a shielding device. The Source-to-Surface-Distance of irradiation is 60 cm, and the irradiation field is 2.54 cm². Control groups of rats were anesthetized and placed in the irradiator but no exposure to irradiation.

The BMMSCs were obtained from three male Sprague-Dawley rats in each group at 3, 7 and 14 days after irradiation. The rats were sacrificed by cervical dislocation and the left and right femur and tibia of the local irradiated rats and control rats were dissected, and then bone marrow mesenchymal stem cells (BMMSCs) were obtained by density gradient centrifugation. The BMMSCs isolated at 3, 7 and 14 days post-irradiation were named D3, D7 and D14, respectively. BMMSCs isolated from direct irradiated bones (left hind limbs of local irradiated rats) and indirect irradiated bones (right hind limbs of local irradiated rats) were named direct irradiated group and indirect irradiated group. BMMSCs obtained from the non-irradiated rats were named control group. The cells were cultured in culture media (DMEM; Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS; Gibco/Invitrogen, Grand Island, NY, USA), 100 U/ml penicillin and streptomycin (North China Pharmaceutical Co., Ltd., Shijiazhuang, China) at 37°C in 5% humidified CO₂. As soon as the cells reached 80% confluence, they were detached using 0.25% trypsin-EDTA and replated into cell culture flasks.

The proliferation ability of BMMSCs was evaluated by 3-(4,5-dimethylithiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. The cells isolated from direct, indirect irradiated bones as well as the control group at 3, 7 and 14 days post-irradiation were seeded in 96-well plates with 3,000 cells per well. When the cells were cultured for 2, 3, 5, 7, 9 and 11 days, the MTT assay were performed as described below. The culture medium was removed and 100 µl of fresh culture medium containing 10 µl of MTT (5 mg/ml; Sigma-Aldrich, St. Louis, MO, USA) was added to each well (21,22). The cells were then incubated at 37°C for 4 h. The color was extracted with 100 µl 10% sodium dodecyl sulfate (China Pharmaceutical Shanghai Chemical Reagent Co., Ltd., Shanghai, China) at 37°C for 2 h. OD values, which were directly related to the viable cell numbers, were determined at room temperature using a microplate reader (Epoch 2 Microplate Spectrophotometer; BioTek, Milan, Italy) and cell growth curves were plotted.

When the third passage of BMMSCs isolated from direct, indirect irradiated bones at 3, 7 and 14 days post-irradiation were cultured for 48 h, we changed the medium into osteogenic inductive medium. Specifically, the medium was prepared by supplementing DMEM with 15% FBS, 50 mg/l ascorbic acid and 10⁻⁸ M dexamethasone (both from Sigma-Aldrich), 10 mM β-glycerol phosphate (China Pharmaceutical Shanghai Chemical Reagent Co., Ltd.) (17,23) and 100 U/ml penicillin and streptomycin (North China Pharmaceutical Co., Ltd.). The medium was changed every 3 days.

The BMMSCs isolated from direct, indirect irradiated bones as well as the control group at 3, 7 and 14 days post-irradiation were seeded in 96-well plates with 3,000 cells per well for ALP activity assay. On the 7th day after osteogenic induction, ALP activity assay was performed as described below. After being rinsed twice with phosphate-buffered saline (PBS), the BMMSCs were lysed with 0.1% Triton X-100 (Sigma-Aldrich) at 4°C for 2 h. Then the cells were lysed three times by ultrasonication (VCX130PB Serial; Sonics & Materials, Inc., Newtown, CT, USA) for 10 sec at 20 kHz on ice. The cell supernatant was collected. Next, 2-amino-2-methyl-1-propanol buffer containing p-nitrophenyl phosphate (Fluka, Co., Milwaukee, WI, USA) were subsequently added to the supernatant and the mixture was incubated for 30 min at 37°C. The reaction was stopped by adding 0.1 M NaOH. Finally, the ALP activity was determined at the wavelength of 405 nm. Protein content was measured using a BCA assay kit (Beyotime Institute of Biotechnology, Shanghai, China) according to the manufacturer's instructions. All results were normalized in relation to protein content (24). The ALP staining assay was also performed on the 7th day after osteogenic induction, and the BMMSCs isolated from direct, indirect irradiated bones and unirradiated bones at 3, 7 and 14 days post-irradiation were seeded in 48-well plate with 6,000 cells per well. The staining procedure was washed the cells twice with PBS and fixed them in 2.5% glutaraldehyde solution for 5 min, then, a BCIP/NBT Alkaline Phosphatase Color Development kit (Beyotime Institute of Biotechnology) was performed according to the manufacturer's instructions.

The BMMSCs from direct and indirect irradiated groups as well as the control group in D3, D7 and D14 were cultured in 48-well plates (5×10⁴ cells/well) with osteogenic inductive medium for 3 weeks. Subsequently, the differentiated cells were stained with Alizarin Red for osteogenic mineralized nodules on the 21th day after osteogenic induction. Cells were first washed with PBS and fixed in cold 95% (v/v) ethanol for an hour, then the fixed BMMSCs were incubated with staining solution (Alizarin Red-Tris-HCl, pH 8.3) at room temperature for 10 min. The areas of positively stained mineral nodules were measured and analyzed (25) with an optical microscope and Simple PCI 5.2.1. imaging software.

The BMMSCs isolated from direct, indirect irradiated bones as well as the control group at 3, 7 and 14 days post-irradiation were cultured in 6-well plates (5×10⁵ cells/well) with osteogenic inductive medium for a week. Then the total RNA was harvested from cells using RNAprep Pure Cell/Bacteria kit (Tiangen Biotech Co., Ltd., Beijing, China) according to the manufacturer's protocols. The RNA was then used to synthesize complementary DNA (cDNA) by using a Quantscript RT kit (Tiangen Biotech Co., Ltd.). According to the manufacturer's instructions, RT was performed in a 20-µl reaction mixture. Then cDNA was analyzed on Real Time PCR Amplifier (Light Cycler 2.0; Roche Diagnostics GmbH, Mannheim, Germany) with SYBR Premix Ex Taq Mix (Takara Bio Inc., Otsu, Japan) for the expression of RUNX2 (also called Cbfa1, initially expressed in osteoprogenitor cells) and osteocalcin (OCN, also called Bglap, a late period marker in matrix maturation). The PCR primers were designed by Primer 5. The gene sequences were searched in MEDLINE. Rat RUNX2 forward, 5′-TGCCACCTCTGACTTCTGC-3′ and reverse, 3′-GATGAAATGCCTGGGAACTG-5′; Rat OCN forward, 5′-GAACAGACAAGTCCCACAC-3′ and reverse, 3′-GAGCTCACACACCTCCCTG-5′; Rat β-actin forward, 5′-CACCCGCGAGTACAACCTTC-3′ and reverse, 3′-CCCATACCCACCATCACACC-5′. All detections were in triplicate for each sample and data were normalized to β-actin levels (ΔΔCq). The qPCR performed 40 cycles at 95°C for 5 sec, and then at 60°C for 20 sec.

Western blot analysis was used to detect protein expression levels of RUNX2 and ALP. The BMMSCs isolated from direct, indirect irradiated bones as well as the control group at 3, 7 and 14 days post-irradiation were cultured in osteogenic inductive medium for a week. Then total cytoplasmic protein was isolated with RIPA lysis buffer and supplemented with phenylmethylsulfonylfluoride (PMSF) (both from Beyotime Institute of Biotechnology). Next, the samples were centrifuged at 20,000 × g for 10 min at 4°C. The protein contents were determined using a BCA protein assay kit (Beyotime) according to manufacturer's instructions. Equal quantities of total proteins were separated by 10% (w/v) Tris glycine SDS/PAGE (Beyotime) and transferred to PVDF membranes (Millipore). The membranes were then blocked with 5% (w/v) non-fat milk in Tris-buffered saline with Tween-20 for an hour at room temperature. We then incubated the membranes with an optimal concentration of the primary antibodies: anti-RUNX2 (1:1000, cat. no. 8486S; Cell Signaling Technology, Inc., Danvers, MA, USA), anti-ALP (1:1000, cat. no. sc-98652; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and anti-Tubulin (Beyotime) overnight at 4°C. Immunoreactive bands were detected with anti-rabbit (Cell Signaling Technology, Inc.) or goat anti-mouse (Beyotime) fluorescein-conjugated secondary antibody and visualized with chemiluminescent substrate (BeyoECL Plus; Beyotime) by Gel Imaging system (Omega Lum C; Aplegen, San Francisco, CA, USA). The protein expression levels were quantified by the optical density ratio of the target and loading control proteins using Image-Pro Plus 6.0 software.

Data are expressed as the mean ± standard deviation (SD) for separate experiments. Statistical analysis was done using a one-way ANOVA by SPSS 20.0 software (SPSS, Inc., Chicago, IL, USA). Values of P<0.05 were considered to be statistically significant.

The BMMSCs isolated from the rats 3, 7 and 14 days after local irradiation were named D3, D7 and D14, respectively. Under a light microscope, no obvious morphological change was observed 3 days post-irradiation in the direct and indirect irradiated groups. However, there were significant morphological changes in the cell shape in the direct and indirect irradiated groups at 14 days post-irradiation compared to the control group. In detail, the control group was presented with long spindle-shaped adherent growth, with equal cytoplasm and oval nuclei (Fig. 1A), while the BMMSCs of direct and indirect irradiated group were irregularly shaped, such as short fusiform or polygons, with bigger cell bodies and increased granules at 14 days post-irradiation (Fig. 1B and C).

The proliferation ability of BMMSCs showed different trends at 3, 7 and 14 days post-irradiation. At 3 days post-irradiation, the proliferation ability of direct and indirect irradiated group decreased slightly, with the decrease rate reaching 20.9 and 32.0%, respectively (after 7 days culture). However, at 7 days post-irradiation, the proliferation of BMMSCs in the direct irradiated group was damaged significantly by 35.9%, and the viability of BMMSCs in the indirect irradiated group was still impaired (Fig. 2B). At 14 days post-irradiation, the proliferation of BMMSCs obtained from direct and indirect irradiated bones were both markedly inhibited by 40.4 and 39.9%, respectively (Fig. 2C).

ALP is a marker for early osteoblast differentiation, and its activity was analyzed to reflect the early osteogenic potential of BMMSCs. The results of ALP activity normalized by protein content (IU/mg protein) were shown as Fig. 3A, and the ALP staining was shown as Fig. 3C. In direct irradiated groups, a transient increased ALP activity was observed at 3 days post-irradiation, then a sharp drop of ALP activity was observed at 7 days post-irradiation, the decrease rate reaching 67.2% compared with the control group, and only a little recovery was observed at 14 days post-irradiation compared to 7 days post-irradiation. Interestingly, in indirect irradiated groups, ALP activity was found to increase both at 3 and 7 days post-irradiation, with the increase rate reaching 29.9 and 31.3%, respectively (P<0.05), and a decline of ALP activity was observed only at 14 days post-irradiation, (decrease rate reaching 16.0%, P<0.05). The results of western blot analysis were shown as Fig. 3B, and the results were well consistent with ALP activity and staining.

Alizarin Red staining was applied to reflect the capacity of mineralization of osteoblast in vitro (Fig. 4A), and the area of bone mineralization nodules quantified by Simple PCI 5.2.1. imaging software were showed as Fig. 4B. Typical nodular structure of mineralized tissue with clear cell accumulation around the nodules can be observed in the control group. In the direct irradiated group, a slight increase of staining for mineralized nodules were observed 3 days post-irradiation (the increase rate reaching 19.6%), while almost no mineralization structure was observed 7 days post-irradiation (the area of bone mineralization nodules decreased by 91.9%), and only a little recovery was observed at 14 days post-irradiation (the decrease rate reaching 57.9%). Additionally, in the indirect irradiated group, there has been a modest decrease of the calcium-richer mineralized matrix at 3, 7 and 14 days post-irradiation, with the decrease rate reaching 19.7, 27.9 and 36.1%, respectively.

Differential expression of lineage-specific genes and proteins. By using the real-time PCR, we found the expressions of osteogenesis-related genes RUNX2 (Runt-related transcription factor 2) and OCN (osteocalcin) altered in BMMSCs after irradiation (Fig. 5A). In the direct irradiated group, the mRNA expression levels of RUNX2 and OCN were clearly higher than that of the control group at 3 days post-irradiation, while a sharp decrease was observed at 7 days post-irradiation, and no obvious recovery was found at 14 days post-irradiation. In other words, in the direct irradiated group there has been a more powerful osteogenic potential at 3 days post-irradiation, but the enhanced potential decreased at 7 and 14 days post-irradiation. In the indirect irradiated group, the expressions of RUNX2 and OCN dropped in varying degrees at 3, 7 and 14 days post-irradiation. Thus, the osteogenic potential of BMMSCs obtained from indirect irradiated bones was impaired at early stages of irradiation. The results of western blot analysis were well consistent with the gene expression levels (Fig. 5B). The protein expression of RUNX2 at D3, D7, D14 groups were shown as Fig. 5B with the control groups showed the basic level. The optical density of the protein bands of RUNX2 (related to tubulin) was quantified, and the results were shown as Fig. 5C.