Healthy female SD rats (4–6-week-old) were used for the isolation and culture of BMSCs. The rats were obtained from the Laboratory Animal Center of China Medical University (Liaoning, China). The animal procedures were approved by the Animal Care and Use Committee of China Medical University and conducted in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals.

The primary and secondary antibodies used in this study were: monoclonal anti-integrin α4 for blocking (Abcam, Hong Kong, China); anti-integrin α4 for immunofluorescence and western blot analysis (Cell Signaling Technology, Danvers, MA, USA); anti-GAPDH (Santa Cruz Biotechnology, Dallas, TX, USA); FITC-labeled anti-rat CD34, CD45, CD73, CD90 and CD105 (Bioss, Beijing, China); and the secondary antibodies for western blot analysis, HRP-conjugated anti-rabbit IgG and HRP-conjugated anti-mouse IgG (both from ZsBio, Beijing, China). An enhanced chemiluminescence kit (ECL) was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Fetal bovine serum (FBS) and Dulbecco’s modified Eagle’s medium of low glucose (L-DMEM) were both obtained from Gibco (Waltham, MA, USA).

The BMSCs were isolated and expanded by their characteristics of adherence to plastic as previously described (18,19). Briefly, the rats were euthanized by CO₂ inhalation following cervical dislocation after intraperitoneal injection of 10% chloral hydrate (3.5 ml/kg). Bilateral tibias and femurs were excised by aseptic procedures. Bone marrow cells were flushed out from the bones using L-DMEM containing 10% FBS. Then the cells were transferred into a T75 cm² flask and incubated at 37°C with 5% CO₂. The non-adherent cells were discarded by changing the media after 48 h. When the adherent cells grew to 90% confluence, the cells were detached by 0.25% trypsin and passaged at a ratio of 1:2. The cells at passage 3 were applied to the subsequent experiments.

Osteogenic, adipogenic and chondrogenic induction were performed as previously described (20,21). To demonstrate the multilineage differentiation capacity of the cells used in this study, P3 BMSCs (1×10⁵ cells in 1 ml medium) were seeded in 6-well plates. After the cell grew to 80% confluence, the culture media were changed to osteogenic (dexamethasone 0.1 µM, β-glycerol phosphatase 10 mM and ascorbate 50 µM, 10% FBS) (19), adipogenic (IBMX 0.5 mM, dexamethasone 1 µM, insulin 10 µg/ml and indomethacin 200 µM, 10% FBS) (19) and chondrogenic differentiation media (insulin 6.25 µg/ml, TGF-β2 10 ng/ml and 2-ascorbic acid 50 nM) (20), separately. The media were refreshed every 3 days. After two weeks, the cells were fixed with 4% paraformaldehyde (PFA) and then stained with 0.1% alizarin red S, 3% oil red O or 1% alcian blue, respectively, to show the formation of mineral, lipid deposits or glycosaminoglycans. The images were captured under an upright microscope (IX71; Olympus, Tokyo, Japan).

The rat C6 glioma cell line and human U87 and U251 glioblastoma cell lines, were purchased from the Chinese Academy of Medical Sciences. The cells were cultured and maintained in L-DMEM with 10% FBS under the same conditions of temperature and humidity as BMSCs. The CM of C6, U251 and U87 glioma cells was prepared as previously described (22). Briefly, when the cells grew to 80% confluence in the T75 flask, the complete media were removed and the cells were washed twice with PBS and followed by incubation with 10 ml serum-free L-DMEM for 24 h. The cells were collected and centrifuged at 1,000 × g for 10 min to remove cell debris. The aliquots of the supernatant were stored at −80°C until required.

Flow cytometry was carried out as previously described (18), to determine the expression of CD markers in the isolated cells. Third-passage BMSCs were collected and fixed in 70% ethanol at 4°C for 30 min. The cells were incubated with FITC-labeled anti-CD34, CD45, CD73, CD90 and CD105 antibodies for 2 h, washed and resuspended with PBS and then analyzed by FACScan flow cytometer (Becton-Dickinson, San Jose, CA, USA).

Transwell cell culture chambers of 8-µm pore size (Corning Costar, MA, USA) were applied to assess the role of integrin α4 in the glioma-induced motility of BMSCs. After BMSCs were trypsinized, the cells were collected and resuspended in serum-free L-DMEM at a density of 5×10⁵/ml. Then, 200 µl of BMSCs suspension was added into the upper chambers. Glioma CM was placed into the lower chambers. The blocking antibody of integrin α4 was added into the upper chamber to a final concentration of 20 µg/ml to neutralized integrin α4 bioactivity. After the cells were co-cultured for 24 h, the cells migrating to the underside of the membrane were stained with Giemsa and the number of cells was evaluated as previously described (16). To assess the role of NF-κB in the migration of BMSCs towards glioma, BAY11-7082 (NF-κB inhibitor; Enzo, New York, NY, USA) was applied to incubate BMSCs at a final concentration of 5 µM during the stimulation of glioma CM.

A cell proliferation assay was used to investigate the effect of glioma CM incubation on the growth of BMSCs. An in vitro growth assay was used to prevent the proliferation of BMSCs affecting the evaluation. BMSCs were seeded in 96-well plates at a density of 5×10³ cells/well in 100 µl L-DMEM containing 10% FBS with five replicate wells for each group. The cells were incubated overnight, and when the cells adhered to the plastic, the culture media were changed to C6 CM, U251 CM and U87 CM, respectively. After incubation of 0 and 24 h, 20 µl MTS solution (Promega, Beijing, China) was added into the wells and the cells were then incubated for 1.5 h in the dark. The absorbance at 490 nm was measured by a multi-channel microplate reader (SpectraMax M5).

BMSCs were treated with C6, U87 and U251 CM for 24 h, respectively. Total RNA was isolated by TRIzol reagent (Life Technologies, Danvers, MA, USA), following the manufacturer’s instructions. The concentration of RNA was determined by a NanoDrop 2000 (Thermo Fisher Scientific). The Takara RNA PCR kit (AMV) ver. 3.0 was used for reverse transcription (Takara, Dalian, China). For routine PCR, the primer sequences of were: integrin α4, 5′-ATCTAGTTTTTACACACAGGATTT-3′ (forward) and, 5′-TGTCAATGTCGCCAAGATT-3′ (reverse) (23); GAPDH, 5′-CCGTATCGGACGCCTGGTTA-3′ (forward) and, 5′-TCTCGCTCCTGGAAGATGGTG-3′ (reverse). The length of the amplified fragment was 533 bp for integrin α4 and 207 bp for GAPDH. The PCR reaction was carried out under the following conditions: denaturation at 94°C for 30 sec, annealing at 52°C for 30 sec, and extension at 72°C for 1 min, and 35 cycles. GAPDH was applied as an internal control. The PCR product was identified using 1% agarose gel electrophoresis containing Genefinder (Zeesan, Xiamen, China). The ratio of the integrated density values (IDV) of the integrin α4 and GAPDH gene was then calculated. qPCR was performed on a TP800 Real-Time PCR system using the SYBR Premix Ex TaqTM II kit (both from Takara) and specific integrin α4 primers as follows: 5′-CGGCACGCTGTTTGGCTACT-3′ (forward) and, 5′-ATCGCCCCAGGATTGACCAC-3′ (reverse). GAPDH served as an internal control. The primers for GAPDH were the same as those for routine PCR. Amplification was performed for 40 cycles as follows: 30 sec at 95°C, and for each cycle 5 sec at 95°C for denaturation and 30 sec at 60°C for annealing. Relative mRNA quantification of integrin α4 to GAPDH was determined by the 2^−ΔΔCt method. In addition, to investigate the roles of PI3K, NF-κB, MEK, p38 MAPK and JNK in the changes of the integrin α4 expression of BMSCs induced by glioma CM, LY294002 (PI3K inhibitor, 30 µM; Cell Signaling Technology), BAY11-7082 (5 µM), PD98059 (MEK inhibitor, 10 µM), SB203580 (p38MAPK inhibitor, 10 µM) and SP600125 (JNK inhibitor, 10 µM; all from Enzo) were applied to incubate BMSCs 30 min before, and for the duration of glioma CM stimulation, separately. Serum-free L-DMEM was used as a negative control.

For immunofluorescence, BMSCs at 70% confluence were trypsinized and seeded onto cover slips coated with 1.5% gelatin. After 48 h, the cells were treated with glioma CM for 24 h, followed by fixation with 4% PFA. The cells were blocked with 5% BSA for 2 h and incubated with monoclonal rabbit anti-rat integrin α4 antibody at 4°C overnight. TRITC conjugated anti-rabbit secondary antibody was then used for visualization. Nuclei were counterstained with DAPI. Fluorescent photomicrographs were captured using an upright microscope (DP71; Olympus). Identical filters, objectives and acquisition parameters were used for each experiment.

BMSCs were incubated with C6, U251 and U87 CM as described above. The cell lysate of each group was prepared in lysis buffer containing protease and phosphatase inhibitor cocktail on ice (Sigma Aldrich). The sample protein concentrations were determined using a Bicinchoninic Acid Protein Assay kit (Beyotime Biotechnology, Shanghai, China). An equal amount of protein lysates (25 µg/lane) were fractionated on 8% SDS-polyacrylamide gel and transferred to PVDF membrane. Subsequently, the membranes were blocked with 5% skimmed milk for 2 h and then treated with anti-integrin α4 and anti-GAPDH antibodies at 4°C overnight, respectively. The protein bands were visualized with an ECL Detection system. The IDVs were measured by the Fluor Chen 2.0 software and GAPDH protein served as an internal control.

Triplicate replications were performed for each experiment. Data were analyzed by SPSS 19.0 and presented as mean ± SD. Statistical differences between two groups were assessed using the Student’s t-test. One-way analysis of variance test (ANOVA) followed by Dunnett’s post test were used to compare differences among multiple groups. P<0.05 was considered statistically significant.

The isolated cells demonstrated the ability of adherence to plastic and exhibited spindle-like shape (Fig. 1A, upper panel). After being passaged, the cells grew to typical fibroblast colonies (Fig. 1A, lower panel). The results obtained by the flow cytometry showed that the isolated cells of passage 3 were positive for CD73 (98.7%), CD90 (99.1%) and CD105 (98.3%), and negative for CD34 (1%) and CD45 (0.966%) (Fig. 1B). Furthermore, following incubation with differentiation medium for 2 weeks, the cells showed positive staining for mineral nodules formation, lipid deposits and glycosaminoglycans. These results indicated that the cells had the capacity of differentiation into osteoblasts, adipocytes and chondroblasts in vitro (Fig. 1C). Based on these results, the cells used in this study were consistent with the definition of BMSCs as previously described (19–21).

Our previous findings showed that the stimulation of glioma CM induced the directional migration of BMSCs and VCAM-1 functioned as an important adhesion molecule in the glioma-induced migration of BMSCs (16). Since VCAM-1 is the counter ligand of VLA-4 and VLA-4 is an integrin dimer composed of integrin α4 and β1, we investigated the relationship between integrin α4 and the glioma-stimulated migration of BMSCs. The results of the in vitro migration assay showed that the number of migrating BMSCs towards glioma CM was significantly decreased with the addition of an integrin α4-blocking antibody, when compared with the control group (Fig. 2A and B). These results suggested that integrin α4 contributed to the glioma-induced motility of BMSCs. We also examined the growth-stimulating activity of glioma CM on BMSCs by MTS assay to prevent the influence of glioma CM on BMSC proliferation using an in vitro migration assay. The results showed that the incubation of glioma CM for 24 h did not significantly increase the number of BMSCs in comparison with the control group (L-DMEM) (Fig. 2C).

We studied the impact of glioma CM on the expression of integrin α4 of BMSCs. In order to assess the effect of glioma CM on the integrin α4 expression of BMSCs, BMSCs were treated with C6, U87 and U251 CM for 24 h, and RT-PCR and immunofluorescence were performed to determine the mRNA and protein expression of integrin α4. As shown in Fig. 3, incubation with glioma CM for 24 h significantly increased the expression of integrin α4 at the mRNA (Fig. 3A and B) and protein levels (Fig. 3C).

To clarify the regulatory mechanisms associated with the glioma-induced integrin α4 upregulation of BMSCs, NF-κB inhibitor BAY11-7082, PI3K inhibitor LY294002, p38 MAPK inhibitor SB203580, JNK inhibitor SP600125 and MEK inhibitor PD98059 were applied in our study. As shown in Fig. 4, the data demonstrated that the glioma CM-induced upregulation of integrin α4 was significantly reduced with the application of BAY11-7082 and LY294002. By contrast, there was no significant inhibition obtained by the addition of SB203580, SP600125 and PD98059 (Fig. 4). These results suggested that NF-κB and PI3K participated in the intracellular signal transduction of glioma CM-induced upregulation of integrin α4.

We previously demonstrated that PI3K played a crucial role in the glioma-induced migration of BMSCs (16). As demonstrated above, integrin α4 contriuted to mediating the glioma-induced migration of BMSCs, and NF-κB and PI3K were key signaling molecules in glioma-induced upregulation of integrin α4. Thus, we hypothesized that NF-κB is also associated with the regulation of glioma-induced migration of BMSCs. To assess the role of NF-κB in the glioma-induced migration of BMSCs, an in vitro migration assay was performed. As shown in Fig. 5, the results of the in vitro migration showed that the treatment of 5 µM BAY11-7082 as mentioned above significantly decreased the motility of BMSCs under the stimulation of C6, U87 and U251 glioma CM (P<0.05 for C6 CM, U251 CM, and U87 CM group), which indicated that NF-κB played an important role in the glioma-induced migration of BMSCs.