Clinical osteosarcoma and tumor-adjacent normal samples from 80 patients with osteosarcoma were obtained from the Department of Orthopedics, The Second Xiangya Hospital, Central South University (Changsha, China). Informed consent was obtained from all patients, and the study was approved by the ethics committee of the Second Xiangya Hospital. The patients had not received any treatment prior to tissue collection. The tissue histology classification was demonstrated using the World Health Organization classification (19). The characteristics of the patients are shown in Table I. After tissue collection, 20 patients with T1/T2 stage osteosarcoma were treated with cis-platinum (30 mg once daily; Sigma-Aldrich, St. Louis, MO, USA) and the samples of osteosarcoma and the tissues were collected at all three stages prior to the second and fourth chemotherapeutic treatment.

MG-63 and SAOS-2 osteosarcoma cell lines (The Second Xiangya Hospital, Central South University) were cultured at 37°C in an incubator with 5% CO₂ in Dulbecco's modified Eagle's medium (DMEM; Gibco, Thermo Fisher Scientific, Inc., Waltham, MA, USA) with 10% fetal bovine serum (FBS; Gibco, Thermo Fisher Scientific, Inc.) and penicillin/streptomycin (100 μg/ml, Sigma-Aldrich).

Antibodies against SerpinE2 (rabbit polyclonal; cat. no. ab75348), CyclinD1 (rabbit monoclonal; cat. no. ab134175), CDK4 (rabbit monoclonal; cat. no. 108357) and GAPDH (rabbit polyclonal; cat. no. ab9485) were purchased from Abcam (Cambridge, UK), and anti-Annexin V-APC was purchased from Invitrogen (Thermo Fisher Scientific, Inc.). Bortezomib was purchased from Toronto Research Chemicals Inc. (North York, ON, Canada) and doxorubicin was obtained from Sigma-Aldrich (St. Louis, MO, USA).

TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Inc.) was used to extract the total RNAs (2 µg). RT was performed using a PrimeScript RT Reagent Kit with a gDNA Eraser (Takara Bio, Inc., Dailan, China) qPCR was performed on an ABI 7500 Real Time PCR system (Thermo Fisher Scientific, Inc.) using SYBR premix Ex Taq II (Takara Bio Inc.). The primers were designed by Primer Express version 3.0 software (Applied Biosystems; Thermo Fisher Scientific, Inc.) and the sequences were as follows: Forward: 5′-TCTCATTGCAAGATCATCGCC-3′ and reverse: 5′-CCCCATGAATAACACAGCACC-3′ for SerpinE2; forward: 5′-CCTAGTACTGCAATTCGGGAAATT-3′ and reverse: 5′-CCTGGAATCCTGCATAAGCAC-3′ for cyclin-dependent kinase (CDK)1; forward: 5′-CCAGGAGTTACTTCTATGCCTGA-3′ and reverse: 5′-AATCCGCTTGTTAGGGTCGTA-3′ for CDK2; forward: 5′-CACAGTTCGTGAGGTGGCTTTA-3′ and reverse: 5′-TGTCCTTAGGTCCTGGTCTACATG-3′ for CDK4; forward: 5′-TGCACAGTGTCACGAACAGA-3′ and reverse: 5′-ACCTCGGAGAAGCTGAAACA-3′ for CDK6; forward: 5′-CTCTTAACCGCGATCCTCCAG-3′ and reverse: 5′-CAATAAAAGATCCAGGGTACATGATTG-3′ for Cyclin A; forward: 5′-AAAGGCGTAACTCGAATGGA-3′ and reverse: 5′-CCGACCTTTTATTGAAGAGCA-3′ for Cyclin B; forward: 5′-TCGCTGGAGCCCGTGAA-3′ and reverse: 5′-CCGCCTCTGGCATTTTGG-3′ for Cyclin D1; forward: 5′-ATACAGACCCACAGAGACAG-3′ and reverse: 5′-TGCCATCCACAGAAATACTT-3′ for Cyclin E; and forward: 5′-CGCTCTCTGCTCCTCCTGTT-3′ and reverse: 5′-CCATGGTGTCTGAGCGATGT-3′ for glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The reaction conditions were as follows: 1 cycle at 95°C for 10 sec, 40 cycles at 95°C for 10 sec and 60°C 30 sec. The relative expression levels were normalized to GAPDH expression using the 2^−ΔΔCq method (20).

The DNA fragment of SerpinE2 was obtained from Invitrogen, Thermo Fisher Scientific, Inc. and cloned into the pEF-BOS-EX expression vector, as previously described (21). The vectors were then transfected into MG-63 and SAOS-2 cells using Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific, Inc.). The wild-type controls of MG-63 and SAOS-2 cells were not transfected with vectors. Briefly, 1 µg SerpinE2 over-expressing (OE) vector was added to cells in the OE groups. There were five replicates for each group. After 24 h transfection according to the manufacturer's instructions, the cells were used for the following experiments.

The cells were counted using a hemocytometer (cat. no. Z359629; Sigma-Aldrich) under an inverted microscope (Axiovert 200 M; Zeiss, Thornwood, NY, USA). The dead cells were determined by 0.4% Trypan Blue (Sigma-Aldrich, St. Louis, MO, USA) staining. A minimum of 100 cells were counted.

Clonogenic growth was assayed by seeding 1×10³ cells in 0.5 ml DMEM containing 10% FBS and 0.33% agar. The cells were cultured for 1 week, the medium was removed and the cells were fixed in 75% methanol and 25% acetic acid for 15 min. The cells were then stained using 0.5% crystal violet (Shanghai Sangon Biological Engineering Technology Co., Ltd., Shanghai, China) in methanol for 20 min. Then the colonies were counted. The clonogenicity was indicated as % colonies relative to MG-63 WT cells. The clonogenic analysis was performed in triplicate.

To analyze cell differentiation, the cells (1×10⁵ cells) were collected using TrypLE (Invitrogen, Thermo Fisher Scientific, Inc.) with 100 U/ml DNase (Invitrogen, Thermo Fisher Scientific, Inc.). The cells were fixed in cold methanol, blocked using 4% FBS and washed with phosphate-buffered saline with Tween 20 (EMD Millipore, Billerica, MA, USA) and then incubated with Annexin V-APC antibody (cat. no. A35110; Invitrogen, Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Finally, the cells were stained with propidium iodide (Invitrogen; Thermo Fisher Scientific, Inc.) and were analyzed using an FC500 machine and CXP version 2.1 software (both purchased from Beckman Coulter, Pasadena, CA, USA).

The cells were treated with bortezomib (5 nM) and doxorubicin (50 nM) as pervious reported (21). After 48 h treatment, the cells were analyzed by a clonogenicity assay.

The proteins were extracted using radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology, Wuhan, China). After determining the protein concentration using a Bradford assay method (Beyotime Institute of Biotechnology), 15 µg protein for each sample was electrophoresed using 15% sodium dodecyl sulfate-polyacrylamide gel and transferred to a polyvinylidene fluoride membrane (Millipore, Billerica, MA, USA). The membranes were incubated with primary antibodies (dilution SerpinE2 1:1,000; Cyclin D1 1:500; CDK4 1:1,000; GAPDH 1:2,000) overnight at 4°C. After three (5 min/time) washes in phosphate-buffered saline with Tween-20, the secondary antibodies (dilution 1:200) were incubated with the membrane for 30 min at room temperature. Finally, the protein signals were visualized with an enhanced chemiluminescent detection system (EMD).

The survival data of osteosarcoma patients were indicated using a Kaplan-Meier curve and analyzed using log-rank test by SPSS v 17.0 (SPSS Inc, Chicago, IL, USA) using the data of the 80 patients. The difference between paired groups was analyzed by paired t-test and the differences among multiple groups (>2) was analyzed by one-way analysis of variance followed by Tukey's test. Gene expression correlations were analyzed using the Spearman's rank correlation coefficient test. P<0.05 was considered to indicate a statistically significant difference.

Significantly higher expression of SerpinE2 was found in osteosarcoma tissues compared with the tumor-adjacent normal tissues from 80 patients (Fig. 1A). In addition, higher SerpinE2 expression in tissues that had metastasized compared with non-metastatic tissues (Fig. 1B). Notably, SerpinE2 expression in tumor-node-metastasis stage II–III was markedly higher than expression in stage I (Fig. 1C). In addition, the levels of SerpinE2 increased substantially at the three serial stages at diagnosis, prior to the second and fourth chemotherapeutic treatment (Fig. 1D).

To demonstrate the effects of SerpinE2 on osteosarcoma cells, MG-63 and SAOS-2 cell lines were transfected with SerpinE2 OE vectors. The increased expression of SerpinE2 in the OE cells compared with untransfected control cells (WT) were verified by western blotting (Fig. 2A). The growth of WT and OE cells was evaluated for 5 days and the results showed significantly increased cell growth of OE cells compared with WT cells (Fig. 2B). The colony formation of MG-63 and SAOS-2 cells was also significantly promoted by over-expression of SerpinE2 (Fig. 2C and D).

Colony formation assays were used to investigate the possibility that SerpinE2 promotes drug resistance in osteosarcoma. There were more colonies of SerpinE2 MG-63-OE cells generated following bortezomib (5 nM) and doxorubicin (50 nM) treatment compared with those of MG-63-WT cells (Fig. 3A). Annexin V as a marker of apoptotic cell death was detected by flow cytometry, which revealed that treatment of cells for 48 h with bortezomib (5 nM) and doxorubicin (50 nM) contributed to fewer apoptotic MG-63-OE cells compared with MG-63-WT cells (Fig. 3B).

To detect the mechanism of SerpinE2 promoting cell proliferation in osteosarcoma, the cell cycle-related genes, CDK1, CDK2, CDK4, CDK6, cyclin A, cyclin B, cyclin D1 and cyclin E were analyzed after transfection with SerpinE2 (Fig. 4A). Only CDK4, CDK6 and cyclin D1 were significantly correlated with SerpinE2 expression (P<0.05). According to the correlation coefficients, CDK4 (r=0.84) and cyclin D1 (r=0.88) were positively correlated with SerpinE2 expression, while no correlation was observed between CDK6 and SerpinE2 (r=0.24). In the cell lines, the protein levels also suggested the correlation between CDK4 and SerpinE2 as well as cyclin D1 and SerpinE2 expression (Fig. 4B). Notably, CDK4 and cyclin D1 mediate the development of tumors by driving cell-cycle progression through G1, which leads to cell proliferation (Fig. 4C). Thus, SerpinE2 can promote cell proliferation of osteosarcoma cells via promoting CDK4 and cyclin D1 expression (Fig. 4C).

The expression of SerpinE2 was determined by RT-qPCR. To confirm the correlation between SerpinE2 expression and survival rates, it was demonstrated that patients with osteosarcoma with high SerpinE2 expression [cut off: 30%, high (>30%) vs. low (<30%)] had the worst outcome compared with the patients with low SerpinE2 expression and medium SerpinE2 expression (P<0.05), indicating that patients with high SerpinE2 expression had poor overall survival (Fig. 5).