Brain specimens from male patients aged 43–62 years were used in the present study. Table I summarizes the clinical characteristics of the 6 patients with glioma included in the present study. The donors' brain samples were obtained from the Chinese Brain Bank Center (CBBC) at the South-Central University for Nationalities (SCUN) and exhibited no signs of clinical or post-mortem neurological disease. The brain tissue samples were collected in June 2018. The human donation plan implemented by the Wuhan Red Cross Society passed the autopsy. Consent was obtained for brain autopsy and use of the brain material according to the protocol of CBBC and the human body donation program, and medical records for research purposes were provided by the donors themselves or their relatives and approved by the Biomedical Research Ethics Committee of SCUN (approval no. 2017-SCUEC-MEC-004). Human glioma tissues were obtained at the time of surgery (n=6) at the Department of Neurosurgery, Renmin Hospital of Wuhan University (Wuhan, China). The samples were collected between September 2018 and November 2018. Pathological diagnosis was confirmed independently by three neuropathologists. Procurement of tissues for the present study was approved by the Institutional Ethics Committee of the Faculty of Medicine at Renmin Hospital of Wuhan University (approval no. 2018K-C017) and written informed consent was obtained from each patient before they succumbed to the disease. In the present study, glioma tissue samples were collected from 6 glioma patients as the glioma group, and healthy brain tissue samples were collected from two healthy donors as the control group. The healthy donors were both male, with an age of 38 and 49 years. Written informed consent was obtained from these individuals. Brain tissue samples from the healthy donors were from the CBBC at the SCUN in June 2018.

The human glial cell line HEB was purchased from Shanghai Bioleaf Biotech Co., Ltd. The human T98G glioma cell line was purchased from the American Type Culture Collection (ATCC), and the human U87 MG and U251 glioblastoma cell lines were obtained from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. The U87 MG cells (cat. no. TCHu138) used in the present study were originally from the ATCC; therefore, they are glioblastoma cells of unknown origin. Cells were cultured in DMEM supplemented with 10% FBS (both from Gibco: Thermo Fisher Scientific, Inc.), 100 µg/ml penicillin and 100 µg/ml streptomycin (Sigma-Aldrich; Merck KGaA), and were incubated at 37°C with 5% CO₂.

Western blot analysis was performed as described previously (24). Briefly, brain tissues (glioma tissue and healthy brain tissue) and glioma cells were lysed in ice-cold RIPA buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% Triton X-100, 0.25% deoxycholate, 1.5 mM MgCl₂, 1 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 10 mM ZnAF, 10 mM pervanadate, 10 µg/ml leupeptin and 10 µg/ml aprotinin) and incubated for 30 min. The protein concentration was determined using the BCA method. The samples lysate was heated at 100°C for 10 min after being mixed with sample loading buffer. The 10 µg protein samples were equally loaded on 10% SDS-PAGE and then transferred to PVDF membranes (EMD Millipore). Following blocking with 5% non-fat milk in Tris-buffer at room temperature for 2 h, the PVDF membranes were incubated with the following primary antibodies overnight at 4°C: ACSL4 mouse monoclonal antibody (1:1,000; cat. no. sc-365230; Santa Cruz Biotechnology, Inc.), GPx4 mouse monoclonal antibody (1:1,000; cat. no. sc-166570; Santa Cruz Biotechnology, Inc.) and GAPDH rabbit monoclonal antibody (1:3,000; cat. no. ab8245; Abcam). The membranes were subsequently incubated with a horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibody (1:10,000; cat. no. ab205719; Abcam) at room temperature for 1 h and signals were visualized using super-signal West Femto (Pierce; Thermo Fisher Scientific, Inc.). Densitometry analysis was performed using ImageJ software (version 1.46; National Institutes of Health).

Fresh frozen human glioma tissue samples and healthy brain tissue samples, which were stored at −80°C, were fixed in 4% paraformaldehyde at 4°C for 24 h and then transferred into 30% sucrose solution in 100 mol/ml PBS at 4°C for 72 h. Subsequently, the glioma tissue samples were kept in 4% paraformaldehyde solution at 4°C overnight. The tissues were then cut into 16-µm sections using a Leica VT1000S vibratome (Leica Microsystems GmbH). Brain sections were stained with primary mouse anti-ACSL4 (1:100; cat. no. sc-365230; Santa Cruz Biotechnology, Inc.) or mouse anti-GPx4 (1:100; cat. no. sc-166570; Santa Cruz Biotechnology, Inc.) antibodies overnight at 4°C. Goat anti-mouse IgG1 cross-adsorbed secondary antibody, Alexa Fluor 488 (1:1,000; cat. no. A-21121; Molecular Probes; Thermo Fisher Scientific, Inc.) was incubated with the tissues at 37°C in the dark for 2 h as the secondary antibody. Tissues were counterstained with Hoechst nuclear dye (Thermo Fisher Scientific, Inc.) at room temperature for 1 min and sections were images were captured using a fluorescence microscope (magnification, ×40). Immunofluorescence analysis was performed using ImageJ software (version 1.46; National Institutes of Health).

pcDNA3.1-ACSL4 plasmid (Shanghai GeneChem Co., Ltd.). negative control pcDNA3.1 vector plasmid (Shanghai GeneChem Co., Ltd.), small interfering (si)-ACSL4 (cat. no. sc-60619; Santa Cruz Biotechnology, Inc.) and negative control siRNA (cat. no. sc-37007; Santa Cruz Biotechnology, Inc.) were used for transfection. The cells were transfected with 1 µg/µl pcDNA3.1-ACSL4 plasmid, 1 µg/µl negative control pcDNA3.1 plasmid, 3 µmol/l si-ACSL4 or 3 µmol/l negative control siRNA using Lipofectamine™ 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Prior to transfection, cells were plated in 6- or 96-well plates and grown to 40–50% confluence. Cells were transfected for 48 h before adding the ferroptosis inhibitor or agonist.

5-HETE, an ACSL4-mediated product of AA oxidation in lysates of intestines and cells (25), was assessed using a 5-HETE ELISA kit (cat. no. CED739Ge; Uscn Life Sciences, Inc.), according to the manufacturer's protocol.

12/15-HETE levels were assessed using 12/15-HETE ELISA kits (cat. nos. ab133034 and ab133035; Abcam), according to the manufacturer's protocols. Each 96-well plate, contained control, blank, standard and sample wells. Each sample was assayed with a minimum of two replicates. Initially, 100 µl of the appropriate diluent and 50 µl assay buffer was added to the non-specific binding (NSB) wells and 100 µl of the appropriate diluent was added to the B0 (0 pg/ml standard) wells. Subsequently, 50 µl 12/15-HETE alkaline phosphatase conjugate was added to NSB, B0, standard and sample wells and 50 µl 12/15-HETE antibody was added to the B0, standard and sample wells. The plates were incubated on a plate shaker for 2 h at 56 × g at room temperature, after which, the contents of the wells were washed three times. Following washing, 5 µl 12/15-HETE alkaline phosphatase conjugate was added to the total activity wells and 200 µl pNpp substrate solution was added to each well. Plates were incubated at 3°C for 3 h without shaking. To stop the reactions, 50 µl stop solution was added to each well. Absorbance was measured at 405 nm using a microplate reader and the mean net absorbance measurement for each well was calculated using the manufacturer-supplied formula to calculate the levels of 12/15-HETE.

LDH is a cytoplasmic enzyme that is retained by viable cells with an intact plasma membrane and released from cells with damaged membranes (26). LDH release was measured using a colorimetric CytoTox 96 Cytotoxicity kit (Promega Corporation) according to the manufacturer's protocol. Maximum LDH release levels were measured by treating the cultures with 10× lysis solution to measure complete lysine levels in the cells. Absorbance was measured at 490 nm using a 96-well plate reader (Molecular Devices, LLC). LDH release, as a percentage, was determined by calculating the ratio of experimental LDH release to maximal LDH release according to the manufacturer's protocol (27).

Cell viability was determined using a CCK-8 assay (Dojindo Molecular Technologies, Inc.) according to the manufacturer's protocol. Cells were seeded in 96-well plates at a density of 5,000/well. CCK-8 was added at 0, 24, 48 and 72 h and incubated at 37°C for 2 h for the measurement of absorbance. Absorbance was measured at 450 nm using a microplate reader (Bio-Rad Laboratories, Inc.).

T98G, U87 and U251 cells were treated with the ferroptosis inducers sorafenib (5 µM) or erastin (10 µM), or the ferroptosis inhibitors ferrostatin-1 (1 µM) or U0126 (5 µM), or an equivalent volume of DMSO for 3 days at 37°C with 5% CO₂. The cells were divided into six groups: i) Control group, untreated cells; ii) control + DMSO group, cells with equivalent volume of DMSO (5 µM); iii) sorafenib group, cells treated with sorafenib (5 µM); iv) siRNA negative control group, cells treated with 2 µmol/l siRNA negative control; v) si-ACSL4 group, cells with 2 µmol/l si-ACSL4; and vi) si-ACSL4 + sorafenib group, cells treated with 2 µmol/l si-ACSL4 and sorafenib (5 µM). Erastin was purchased from Hycultec GmbH. Sorafenib was purchased from LC Laboratories. Ferrostatin-1 and U0126 were purchased from Santa Cruz Biotechnology, Inc.

The data and statistical analysis in the present study was performed according to the recommendations on experimental design and analysis (28). Group sizes per experiment were based on a power analysis. Student's t-test or one-way ANOVA were used where appropriate to determine significance. Bonferroni's post-hoc test was used where appropriate. All results are presented as the mean ± standard error of the mean. P<0.05 was considered to indicate a statistically significant difference.

It has been demonstrated that ferroptosis is involved in the pathopoiesis of glioma (29). However, research in this area remains limited. Therefore, in the present study, the levels of ferroptosis in human glioma tissues and glioma cells were measured. GPx4, 12-HETE and 15-HETE are either the lipid peroxidation products or associated with the deposition of ferritin involved in ferroptosis. These molecules are increasingly recognized as markers of ferroptosis (11,13,24). Therefore, GPx4, 12-HETE and 15-HETE were selected as detection indicators for ferroptosis. The higher the level of GPx4 and the lower the level of 12/15-HETE, the less ferroptosis (11,13,24). In the present study, the expression of GPx4 was significantly increased in human glioma tissues compared with brain tissues from healthy patients as determined by western blotting (Fig. 1A). Significantly increased expression of GPx4 was also observed in glioma cells in vitro (Fig. 1B). Immunofluorescence staining was used to evaluate GPx4 expression in Hoechst-positive cells in human glioma tissues. The results demonstrated that GPx4 expression was significantly upregulated in glioma cells in vivo (Fig. 1C). ELISA was used to evaluate the levels of 12-HETE and 15-HETE. Both 12-HETE and 15-HETE levels were decreased in glioma cells compared with the controls (Fig. 1D and E). These results show a higher level of GPx4 and a lower level of 12/15-HETE in human glioma tissues and glioma cells compared with the healthy human tissues and glial cell. Therefore, it was suggested that ferroptosis is reduced in glioma both in vivo and in vitro.

ACSL4 was previously suggested to be a key mediator of ferroptosis (11). However, to the best of our knowledge, whether ACSL4 modulates ferroptosis in glioma has not been studied. A significant downregulation of ACSL4 was observed in human glioma tissues compared with the control (Fig. 2A). In addition, immunofluorescence staining showed that ACSL4 expression in Hoechst-positive cells in human glioma tissues was significantly downregulated in glioma cells in vivo (Fig. 2B). ACSL4 expression in glioma cells was also significantly reduced compared with in glioma cells (Fig. 2C). Therefore, it was hypothesized that the reduction in ACSL4 expression may be associated with ferroptosis disorder in glioma.

To examine whether ferroptosis affects proliferation, cytotoxicity (LDH) and cell survival (CCK-8) assays were performed to evaluate cell proliferation in cells treated with ferroptosis inducers or inhibitors. T98G, U87 and U251 glioma cells were used to observe the role of ferroptosis on proliferation. It is understood that the more LDH release, the more severe the cell damage, which results in a decrease in cell proliferation (26). Similarly, in a CCK-8 assay, the higher the indicator, the higher the cell viability (30). Treatment with the ferroptosis inducers sorafenib and erastin resulted in a significant increase in LDH release (Fig. 3A-C) and a significant decrease in the viability of glioma cells (Fig. 3G-I). This indicates that ferroptosis-induction reduces proliferation and viability. By contrast, treatment with the ferroptosis inhibitors Ferrostatin-1 and U0126 resulted in a significant decrease in LDH release (Fig. 3D-F) and a significant increase in viability (Fig. 3J-L). This suggests that ferroptosis-inhibition increases proliferation and viability. This is consistent with the expected results (31,32). These results suggest that ferroptosis attenuates viability in glioma cells.

ACSL4 has been proposed to serve a crucial role in ferroptosis and proliferation in other animal models. ACSL4 inhibition prior to reperfusion protects against ferroptosis and cell death in intestinal ischemia/reperfusion (33). However, whether ACSL4 modulates ferroptosis and proliferation in glioma cells remains unclear. pcDNA3.1-ACSL4 and si-ACSL4 were transfected into three different cultured glioma cell lines to elucidate the effects of ACSL4 on ferroptosis and proliferation in glioma. pcDNA3.1-ACSL4-mediated overexpression of ACSL4 significantly reduced the expression of GPx4 at the protein level (Fig. 4A, E and I) and significantly increased other markers of ferroptosis, including 5-HETE, 12-HETE, and 15-HETE (Fig. 4B-D, F-H and J-L). The lower the level of GPx4 and the higher the level of HETEs, the more ferroptosis (8). The results demonstrate that ACSL4 overexpression upregulates ferroptosis. As presented in Fig. 4M-R, ACSL4 overexpression significantly increased the release of LDH and significantly reduced the viability of glioma cells, suggesting that ACSL4 overexpression downregulated proliferation in glioma cells. siRNA-mediated knockdown of ACSL4 significantly increased GPx4 protein expression (Fig. 5A, E and I) and reduced the expression of 5-HETE, 12-HETE and 15-HETE (Fig. 5B-D, F-H and J-L). This indicated that siRNA-mediated knockdown of ACSL4 increased GPx4 expression, and reduced the markers of ferroptosis. The results demonstrated that downregulation of ACSL4 decreased ferroptosis. In addition, si-ACSL4 significantly reduced the release of LDH (Fig. 5M-O) and significantly increased the viability of glioma cells (Fig. 5P-R). These results demonstrated that downregulation of ACSL4 promoted proliferation in glioma cells. Therefore, it was concluded that ACSL4 serves a vital role in regulating ferroptosis and proliferation in glioma cells.

As ACSL4 was demonstrated to serve a role in regulating proliferation in glioma cells, the underlying mechanisms were then determined. si-ACSL4 and the ferroptosis inducer Sorafenib were used to determine the mechanism by which ACSL4 regulates proliferation in glioma cells. As presented in Fig. 3, sorafenib induced ferroptosis and reduced glioma cell proliferation. It was also demonstrated that siRNA-mediated knockdown of ACSL4 could inhibit ferroptosis and promote glioma cell proliferation (Fig. 5). On the basis of these results, further investigation was performed. As presented in Fig. 6A-C, LDH release was significantly increased in the sorafenib group compared with the control + DMSO group and this effect could be significantly reduced by siRNA-mediated knockdown of ACSL4 in the si-ACSL4 + sorafenib group. This demonstrated that knockdown of ACSL4 could reduce sorafenib-induced ferroptosis. Furthermore, Fig. 6D-F demonstrates the viability of glioma cells in the different groups at 72 h after glioma cells were seeded in 96-well plates in the CCK-8 assay. The results of the CCK-8 assay indicated that the sorafenib-induced reduction in cell viability was reversed by siRNA-mediated knockdown of ACSL4. Taken together, these results suggest that ferroptosis is involved in the regulation of proliferation by ACSL4 in glioma cells; ACSL4 can inhibit proliferation of glioma cells by activating ferroptosis.