The human lung adenocarcinoma A549 cells and bone marrow stromal HS-5 cells were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured in Dulbecco's modified Eagle's medium (DMEM; Hyclone, USA) containing 10% fetal bovine serum (FBS; Gibco, USA) and 1% penicillin/streptomycin at 37°C in a 5% CO₂ incubator under humidified conditions. Recombinant adenovirus AdBMP9 and negative control AdGFP were kindly provided by Dr Tongchuan He (University of Chicago Medical Center, Chicago, IL, USA), AdBMP9 or AdGFP was transfected into the A549 or HS-5 cells with Polybrene (Sigma, USA). The medium was replaced with fresh medium after 8 h of cultivation.

The co-culture system was set up using 6-well Transwell inserts (Corning, USA) with a 0.4-µm pore size. In the Transwell chamber, A549 and HS-5 cells were plated in the upper chambers at the density of 1×10⁵ cells/well in 1 ml, whereas HS-5 and A549 cells were plated in 6-well plates at the density of 2×10⁵ cells/well in 2 ml. Cells became adherent after 6 h, and then were placed together in DMEM supplemented with 2% FBS for 3 days. Cells alone were used as the control.

Cell viability was detected by MTT [3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay. A total of 4×10⁴ A549 cells/well were cultured in 6-well plates and 2×10⁴ HS-5 cells were cultured in a chamber. The A549 cells were treated with AdGFP or AdBMP9, and the cells were cultured for 24, 48 and 72 h. At the indicated time, MTT reagent (Sigma) was added (200 µl/well), and incubation was carried out at 37°C for 4 h. Next, 3 ml of dimethyl sulfoxide (DMSO) was added to the 6-well plates to dissolve the formazan product for 10 min at room temperature, and 150 µl lysate was added into the 96-well plate. Finally, the absorbance was measured at 492 nm using a microplate reader. Each group was conducted in quintuplicate, and the overall experiment was repeated 3 times.

Log-phase A549 cells treated with AdGFP or AdBMP9 were collected and seeded in 10% FBS medium at 200 cells/well in 6-well plates for 14 days. When clones were observed, the cells were washed twice with PBS and stained with 1% crystal violet. The colony formation rate was calculated as: (Colony number/seeded cell number) × 100%. Each experiment was repeated thrice.

Cell apoptosis analysis was assessed by flow cytometry. The A549 cells were seeded into 6-well plates at a density of 2×10⁵ cells/well and treated with AdGFP or AdBMP9 for 48 h. Then the cells were harvested and re-suspended in 1 ml cold PBS, and samples were added with propidium iodide (PI) and FITC-Annexin V for cell apoptosis analysis according to the manufacturer's protocol (Invitrogen, USA). Respectively, data were analyzed using FACS Sorter (Becton Dickinson, San Jose, CA, USA).

A549 cells were seeded into 6-well plates and HS-5 cells were seeded into chambers. A549 cells were treated with AdGFP or AdBMP9. When A549 cells had grown to 95% confluency, the monolayer was scratched with a sterile pipette tip, and then washed with PBS twice to remove cellular debris. Culture medium was replaced with fresh DMEM containing 1% FBS, and then co-cultured with the HS-5 cells. Cells that migrated into the scratched area were compared using bright field microscopy at 48 h. Experiments were performed in triplicate.

HS-5 cells are seeded into a 24-well plate containing 10% FBS. After the cells became adherent, the A549 cells (4×10⁴/400 µl) in FBS-free DMEM were seeded into the chambers (24-well Transwell chambers, 8-µm pore size; Corning). The Transwell membrane was coated with 1:3 diluted Matrigel (Sigma) beforehand. After 24 h, the cells that invaded to the underside of the filter were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet, and counted by bright field microscopy. All the experiments were repeated 3 times.

A549 cells were seeded into 6-well plates and incubated at 37°C in a CO₂ incubator until they were 60–80% confluent, then the cells were transfected with 50 nM IL-6 or IL-8 siRNA according to the manufacturer's instructions. Transfected cells were incubated for 3 days. The siRNA sequences which aimed to target IL-6 and IL-8 were designed by RiboBio and are shown in Table I.

A549 cells were co-cultured with HS-5 cells or alone or were treated with AdBMP9 in FBS-free DMEM for 3 days. Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. Total RNA (2 µg) was used for cDNA synthesis by reverse transcriptase-PCR. The mRNA expression was quantitatively determined using a real-time polymerase chain reaction (PCR) system (Bio-Rad, USA) using SYBR Green PCR Master Mix. GAPDH was used as the invariant control. Amplification was performed with the following protocol: denaturation at 94°C for 30 sec, annealing at 55°C for 30 sec, and extension at 72°C for 50 sec. The primer sequences used for real-time PCR are shown in Table II.

A549 cells were cultured alone or co-cultured with HS-5 cells, and treated with AdBMP9 for 3 days. Then the A549 cells were washed three times with cold PBS and lysed in RIPA lysis buffer, and the cell lysate was denatured after boiling. The protein concentration was measured by a BCA protein assay kit. Fifty micrograms of lysate was loaded onto 10% SDS-PAGE gels, and subsequently transferred onto PVDF membranes. After blocking with 5% bovine serum albumin (BSA; Solarbio) in TBST for 2 h at 37°C, the membranes were incubated with the following primary antibodies [polyclonal rabbit anti-BMP9 (Abcam, Cambridge, MA, USA); anti-caspase3, anti-Bax, anti-Bcl2, anti-MMP2, anti-PCNA, anti-IL-6, anti-IL-8 (Wanleibio, Shenyang, China); monoclonal rabbit anti-ERK1/2, anti-p-ERK1/2, anti-P38, anti-p-P38, anti-AKT, anti-p-AKT, anti-p-P65 (Cell Signaling Technology, Danvers, MA, USA); monoclonal mouse anti-β-actin (ZSGBBIO, Beijing, China); rabbit anti-histone (Abmart, Shanghai, China)] at 4°C overnight, washed with TBST for 3 times, followed by incubation with a secondary antibody conjugating with horseradish peroxidase (ZSGBBIO) for 1 h, and washed again. Protein levels were quantified with the SuperSignal West Pico Chemiluminescent Substrate kit.

The animal experiments were performed in accordance with the guidelines established by the Animal Care and Use Committee of Chongqing Medical University Laboratory Animal Research. The 4- to 6-week-old male NOD/SCID mice were randomly divided into 4 groups (n=4/group), respectively, and were injected subcutaneously with A549 cells (6×10⁶); a mixture of A549 (6×10⁶) and HS-5 cells (3×10⁶); A549/GFP (6×10⁶) and HS-5 cells (3×10⁶); A549/BMP9 (6×10⁶) and HS-5 cells (3×10⁶). Tumor volume was measured every week with Vernier calipers, and calculated in mm³ as ab²/2, where 'a' and 'b' represent the largest and smallest tumor diameter, respectively. The mice were sacrificed after 7 weeks, and tumor tissues were collected, embedded in paraffin, and cut into 4-µm sections. The sections were stained using H&E and immunohistochemical analyses were conducted.

Paraffin sections of tumor tissues were de-waxed, rehydrated and heat-treated for antigen retrieval with citric acid buffer. The sections were then blocked with normal goat serum for 30 min, incubated with polyclonal rabbit anti-Ki-67 (Wanleibio) at 4°C overnight, and were analyzed using an immunohistochemistry kit (PV-9001; ZSGBBIO) following the manufacturer's protocol. Staining procedures were performed under standardized conditions. The sections were counterstained with hematoxylin, mounted, and coverslipped. Staining intensity was independently assessed by the authors.

All values are expressed as the mean ± SEM. Statistical significance was determined using the Student's t-test in GraphPad Prism 5. A P-value of <0.05 was considered to indicate a statistically significant result.

We aimed to ascertain the effect of BMP9 on the biological behavior of NSCLC cells AdBMP9 was used to overexpress BMP9 in the lung adenocarcinoma A549 cells which have a low level of BMP9. The infection effciency of AdBMP9 in the A549 cells was observed under a fluorescence microscope (Fig. 1A). RT-PCR and western blot results showed that these recombinant cells were successfully established (P<0.001; Fig. 1B and C). Cell proliferation ability was assessed by MTT and colony-forming assays. The results showed that the proliferation of the A549/BMP9 cells was decreased significantly (Fig. 2A and B). The cell apoptosis rate of the A549/BMP9 cells was significantly increased compared with the rates noted in the blank and GFP groups by flow cytometry (P<0.01; Fig. 2C). Cell migration and invasion abilities were detected by wound-healing and Transwell invasion assays. The findings demonstrated that BMP9 decreased the wound-healing rate and the number of invasive A549 cells (Fig. 2D and E). Furthermore, we detected the protein levels of DNA replication factor (PCNA), cell apoptosis factors (pro-apoptosis factors caspase-3 and Bax and anti-apoptosis factor Bcl2), and migration-related factor (MMP-2) by western blotting. The results showed that PCNA, MMP-2 and Bcl2 were decreased, while caspase-3 and Bax were upregulated by BMP9 overexpression in the A549 cells (Fig. 2F).

When co-cultured with HS-5 cells, the cell viability of the A549 cells obviously increased as determined by MTT assay. Cell migration and invasion abilities were increased as determined by wound-healing and Transwell invasion assays (Fig. 3A–C). After AdBMP9 was added to the co-culture system, the cell proliferation rate, the wound-healing rate and the number of invading A549 cells in the co-culture/BMP9 group was found to be significantly decreased than these values in the co-culture/GFP group (Fig. 3A–C). To investigate the effects of HS-5 cells and BMP9 on the tumor growth of lung cancer cells in vivo, A549 or adenovirus-infected A549 cells and HS-5 cells were subcutaneously implanted into nude mice. The tumor size was monitored weekly and the tumors were dissected after xenografting for 7 weeks. Tumor volumes differed obviously after 4 weeks. HS-5 cells accelerated the growth of the A549 tumors, while BMP9 significantly inhibited the accelerative role of the HS-5 cells compared to the GFP group (Fig. 3D). These results were consistent with those in vitro. H&E staining showed no variation in heterogeneity between the single A549 and A549/HS-5 group, but there were many lymphocytic invasive cells in the tumor tissue of the BMP9 group (Fig. 3E). Ki-67 expression was detected by immunohistochemical staining. The results showed that the Ki-67-positive cell rate was increased in the A549/HS-5 group when compared with the A549 group, and BMP9 inhibited the expression of Ki-67 (Fig. 3F).

Studies have reported that expression levels of IL-6 and IL-8 are higher in lung cancer patients with metastasis than those without metastasis. Meanwhile, IL-6 and IL-8 could be secreted by bone marrow stromal cells, and they are known to influence osteoclast formation and bone resorption. BMP9 is the most effective bone formation factor of the BMPs. We investigated whether IL-6 and IL-8 are related to the pro-metastatic effect of A549 cells in the HS-5 cell-mediated tumor microenvironment. We detected the mRNA and protein levels of IL-6 and IL-8 in the A549 and HS-5 cells in the co-culture system by RT-PCR and western blotting. IL-6 and IL-8 were increased in the A549 (Fig. 4A and B) and HS-5 cells (Fig. 4C and D, but had no statistical significance) in the co-culture system, BMP9 inhibited the expression levels of IL-6 and IL-8 compared to the GFP group (Fig. 4A–C). When IL-6 and IL-8 were knocked down by siRNA, the interference efficiency of the three siRNAs targeting IL-6 and IL-8 was verified by RT-PCR and western blotting (Fig. 4E and F). IL-6-siRNA-3 and IL-8-siRNA-2 were the most highly functional one and were used in the following assays. The wound-healing rate and the number of invasive A549 cells were decreased in the co-culture system with knockdown of IL-6 or IL-8 (Fig. 4G and H).

NF-κB and MAPK are crucial signaling pathways responsible for gene induction in NSCLC cell proliferation (30,31). Activation of these pathways in lung cancer cells initiates signaling cascades and leads to overproduction of inflammatory cytokines and growth factors including IL-6, IL-8 and MMPs, which promote cancer progression and metastasis (32,33). Whether these pathways involved in lung cancer cells thrive within the bone environment remains unkown. As shown in Fig. 5A and B, the A549 cells had enhanced expression of p-ERK1/2 and p-P65, but not p-P38 and p-AKT after co-culture with the HS-5 cells for 3 days. When IL-6 and IL-8 were silenced by small interference RNA siIL-6 and siIL-8, or overexpression of BMP9, the expression of p-ERK1/2 and p-P65 was decreased. The same results were found with the MEK- and NF-κB-specific inhibitors U0126 (10 µM; Selleckchem, USA) and BAY11-7082 (5 µM; Beyotime, Shanghai, China), respectively (Fig. 5C and D). Meanwhile, U0126 and BAY11-7082 decreased the invasive ability of the A549 cells in the co-culture system (Fig. 5E).