In total, 10 human glioma cell lines, namely MGR1, MGR3, SF295, SKMG-1, SKMG-4, T98G, U251, U251/CP2, U373 and U87, were used in the present study. All cells were maintained in Dulbecco's modified Eagle's medium /F-12 (Gibco Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (GE Healthcare Life Sciences, Chalfont, UK) and 1.0% penicillin/streptomycin (Gibco Life Technologies) in a humidified incubator at 37°C with 5% CO₂. DHA (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich; 20 µg/ml) and stored at −20°C in the dark. TMZ (Sigma-Aldrich) was dissolved in DMSO, (stock concentration, 0.2 M) and stored at −20°C.

The glioma cells were seeded into 96-well plates at a density of 5×10³ cells/well and incubated overnight at 37°C. Next, the cells were treated with DHA at serial concentrations (0, 0.3125, 0.625, 1.25, 2.5, 5.0, 10.0 and 20.0 µg/ml) for 72 h. Cell viability was analyzed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. In order to detect the effect of DHA on TMZ efficiency, the cells were treated with 2 µM DHA in combination with various concentrations of TMZ (ranging between 31.25 and 2,000 µM). Following 72 h of incubation, the MTT assay was performed and the half maximal inhibitory concentration (IC₅₀) was calculated. Briefly, MTT (Sigma-Aldrich) was added to cells following treatment and incubated for 4 h at 37°C. DMSO was subsequently added and absorbance was measured by microplate reader (Multiskan MK3; Thermo Fisher Scientific, Waltham, MA, USA) after shaking for ~30 sec.

The cell lysates were collected using radioimmunoprecipitation assay lysis buffer (1% NP-40, 1% sodium deoxycholate, 0.1% SDS, 150 mM NaCl and 10 mM Na₂HPO₄; pH 7.2; Beyotime Institute of Biotechnology, Japan), supplemented with a protease inhibitor cocktail (Roche Diagnostics GmbH, Mannheim, Germany). Immunoblotting was performed as previously described (19). Rabbit anti-human antibodies against LC3-B (polyclonal; catalog no. ab48394; dilution, 1:1,000; Abcam, Cambridge, MA, USA), Beclin-1 (monoclonal; catalog no. ab51031; dilution, 1:750; Abcam) and Caspase-3 (polyclonal; catalog no. ab90437; dilution, 1:100; Abcam) were added and incubated overnight at 4°C. A mouse anti-human β-actin antibody (catalog no. AA128; dilution, 1:1,000; Beyotime Institute of Biotechnology) was loaded as the internal control. The membranes were then washed three times with phosphate-buffered saline (PBS). Horseradish peroxidase-conjugated goat anti-mouse (dilution, 1:1,000; catalog no. A0216; Beyotime Institute of Biotechnology) and goat anti-rabbit (dilution, 1:1,000; catalog no. 7074; Cell Signaling Technology Inc., Danvers, MA, USA) secondary antibodies were applied for 30 min at room temperature. The membranes were then washed three times with PBS and incubated with enhanced chemiluminescence reagent (Millipore, Billerica, MA, USA) according to the manufacturer's instructions. Next, the membranes were exposed to ECL sensitive films (Hyperfilm ECL; GE Healthcare Bio-Sciences, Pittsburgh, PA, USA), and developed using an OPTIMAX X-Ray film processor (Protec GmbH & Co. KG, Oberstenfeld, Germany). All protein bands were normalized to β-actin, which was used as the internal loading control.

The glioma SKMG-4 cells, which are sensitive to DHA, were seeded into a 30-mm culture dish and treated with 2 µg/ml DHA for 24 h. MDC (Sigma-Aldrich), at a concentration of 0.05 mM, was then added for 10 min. Next, the cells were fixed in 4% paraformaldehyde (Sigma-Aldrich) for 15 min and washed twice with phosphate-buffered saline. The cytoplasmic autophagic vacuoles were then observed under a fluorescence microscope (Olympus BX61; Olympus Corp., Tokyo, Japan).

The U251 cells were maintained in standard culture medium, adjusted to 5×10⁶ cells/ml. Next, 300 µl cell suspension was injected into the axillary region of the nude mice (age, 4–5 weeks; mean weight, 22.6 g). Mice were maintained in specific-pathogen-free grade room, with 6 mice per cage, at 18–29°C and with light from 7:30 am to 19:30 pm. Two weeks later, the tumor masses were resected, cut into sections measuring 2 mm³ and transplanted subcutaneously into the backs of a second set of experimental nude mice. These mice were randomly separated into four groups (n=10) and treated for 5 continuous days with either DMSO (control), 50 mg/kg DHA only, 50 mg/kg TMZ only, or 50 mg/kg DHA plus 50 mg/kg TMZ. The tumor sizes were measured using a vernier caliper every three days during the following month. The tumor volumes (TV) were calculated according to the following formula: TV = 1/2 × length × width² (20). The present study was approved by the ethics committee of Sun Yat-sen University Cancer Center (Guangzhou, China) according to Institutional Animal Care and Use Committee protocols. Chloral hydrate was applied for anesthesia (10%; 80 µl per mouse), and mice were sacrificed by cervical dislocation.

The data is expressed as the mean ± standard error of the mean. Student's t-test was used to determine any statistically significant differences. P-values are two sided, with P<0.05 considered to indicate a statistically significant difference. Statistical analysis was performed using SPSS 16.0 (SPSS, Inc., Chicago, IL, USA).

The IC₅₀ of ten glioma cell lines treated with serial concentrations DHA was calculated. Following treatment, the SKMG-4 cells appeared to be low in density and exhibit abnormal morphologies compared with the non-treated cells (Fig. 1A and B). The extent of DHA toxicity was mild and differed amongst the glioma cell lines (Fig. 1C). The IC₅₀ also varied between the ten human glioma cell lines, from 1.17±0.08 µg/ml in U251/CP2 cells to 23.57±0.80 µg/ml in U373 cells (Fig. 1D).

The cell lysates were collected from SKMG-4 cells following treatment with DHA for 72 h. Markers of apoptosis and autophagy were then detected by western blotting. There were no significant alterations to the expression of Caspase-3 following DHA treatment. However, the expression of the autophagy markers, Beclin-1 and LC3-B, increased upon DHA treatment in a dose-dependent manner (Fig. 2A). The autofluorescent reagent MDC, a specific autophagic vacuole marker, was also applied in order to monitor the process of autophagy. Following culture with DHA and MDC, the SKMG-4 cells were fixed and observed under a fluorescent microscope. It was revealed that MDC accumulated in DHA-treated cells, but not in untreated control cells (Fig. 2B and C). The data therefore indicated that autophagy was induced by DHA in glioma cell lines.

The present study aimed to investigate whether combined treatment with DHA and TMZ was able to increase cell sensitivity to TMZ. In total, four glioma cell lines, namely SKMG-4, U251, U373 and T98G, were used for the combination treatment due to the differences in their sensitivity to DHA and their genetic background. The 10% inhibition concentration of TMZ (IC₁₀) was calculated according to the results of the cytotoxicity assay (Fig. 1C). The concentration of DHA was fixed to the IC₁₀ of each cell line and then combined with serial concentrations of TMZ. The IC₅₀ of TMZ was calculated following co-treatment for 72 h. It was revealed that the application of DHA reduced the IC₅₀ of TMZ in all four cell lines; from 417.36±17.92 to 143.90±28.74 µM in SKMG-4 cells (P<0.01), 455.06±64.39 to 246.84±26.14 µM in U251 cells (P<0.01), 759.03±76.96 to 537.15±11.88 µM in U373 cells (P<0.01) and 1274.72±159.70 to 863.20±73.26 µM in T98G cells (P<0.05) (Fig. 3). These data indicated that DHA sensitized glioma cell lines to TMZ.

The glioma U251 cell line was used for the in vivo tumor formation and inhibition assay, due to its ability to form tumors in xenografts in vivo and its sensitivity to DHA (low IC₅₀) in vitro. The tumor size was monitored every 3 days, following continuous treatment for 5 days with DMSO, DHA, TMZ and DHA+TMZ. It was revealed that 50 mg/kg DHA alone had no effect on tumor growth in vivo compared with the DMSO-treated control cells. By contrast, TMZ significantly suppressed tumor growth compared with that of the DMSO-treated control cells. Co-treatment with DHA and TMZ significantly reduced the tumor size compared with the effects of TMZ alone (Fig. 4; P<0.01, P<0.05). This indicated that DHA was able to strengthen the tumor inhibitory effect of TMZ in vivo. Taken together, the data demonstrated that DHA inhibited tumor cell proliferation and sensitized glioma cells to TMZ in vitro and in vivo, potentially through the activation of autophagy.