The normal human liver cell line (L02) was purchased from The Cell Collection Center of Wuhan University. The Cell Counting Kit-8 (CCK-8) assay was purchased from Dojindo Chemical Technology Co., Ltd. GLY was purchased from Selleck Chemicals. The lactate dehydrogenase (LDH), malondialdehyde (MDA), glutathione (GSH) and ROS kits were purchased from Beyotime Institute of Biotechnology. The primary antibodies against glutathione peroxidase 4 (GPX4, cat. no. sc-166570) was purchased from Santa Cruz Biotechnology, Inc. GAPDH (cat. no. 10494-1-AP) were purchased from ProteinTech Group, Inc. The primary antibodies against high mobility group box 1 (HMGB1, cat. no. 3935), nuclear factor E2-related factor 2 (Nrf2, cat. no. 12721) and homooxygenase-1 (HO-1, cat. no. 43966) were purchased from Cell Signaling Technology, Inc. The goat anti-rabbit IRDye fluorescent secondary antibody IRDye800CW (cat. no. 926-32211) was purchased from LI-COR Biosciences. Cy3 (cat. no. BA1031 and BA1032) and FITC (cat. no. BA1101 and BA1105) fluorescently labeled rabbit anti-goat secondary antibodies were purchased from Wuhan Boster Biological Technology, Ltd. The iron ion detection kit was purchased from Abcam. D-GalN, LPS and TNF-α were purchased from Sigma-Aldrich; Merck KGaA.

L02 cells were cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.) containing 10% fetal bovine Serum (FBS, Gibco; Thermo Fisher Scientific, Inc.) The cells were cultured at 37°C in an incubator with 5% CO₂. When cells reached a confluency of 70–80%, cells were digested with trypsin and seeded into 6-(1.5×10⁶ cells/1.5 ml DMEM per plate) or 96-well (5×10³ cells/100 µl DMEM per plate) plates. The cells were divided into five groups as follows: The normal group, model group, GLY 0.5 mM intervention group, GLY 1 mM intervention group and GLY 2 mM intervention group. The GLY intervention groups were stimulated with the corresponding concentration of GLY for 2 h at 37°C in advance of further treatment. After 2 h, TNF-α (100 ng/ml) and D-GalN (44 µg/ml) was added to the cells in the model group and GLY intervention groups for 24 h. The cells and supernatant were collected for subsequent experimental testing.

In total, 40 male specific- pathogen-free C57BL/6 mice (Hubei Animal Experimental Center) 6–8 weeks old and weighing 20–25 g were housed in the animal experiment center in Renmin Hospital of Wuhan University with 12 h light/dark cycle, temperature of 25±2°C and relative humidity of 50±15%. All mice were provided with food and water ad libitum. The mice were randomly divided into 5 groups: The normal group, model group, 15 mg/kg GLY group, 30 mg/kg GLY group and 60 mg/kg GLY group. Except for the normal group, the other four groups of mice were injected intraperitoneally with D-GalN (400 mg/kg) and LPS (100 µg/kg) to induce the ALF model. According to a previous study on GLY gavage doses (13), three doses of GLY (15, 30 and 60 mg/kg/day) intervention groups were used. A total of 24 mice were divided into three groups. Mice received gavage with different doses of GLY for 3 days before induction of the ALF model. The ALF model group received the same amount of isotonic saline. After 24 h, the liver tissues were removed. The serum was collected to determine the levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and total bilirubin (TBIL) by using a fully automatic biochemical analyzer (ADVIA 2400, Siemens AG). The present study was approved by the Institutional Animal Care and Use Committee of Renmin Hospital of Wuhan University.

L02 cells were uniformly seeded in a 96-well plate at a density of 5×10³ cells/100 µl in DMEM with different concentrations of GLY (0, 0.5, 1 and 2 mM). According to the manufacturer's protocol, 10 µl CCK-8 reagent was added to each well after 24 h. After mixing, the optical density (OD) for each well was measured using a microplate reader. Cell viability was calculated according to the following formula: Cell viability=(OD value of experimental group-OD value of blank control group)/(OD value of normal group-OD value of blank control) ×100.

The LDH, MDA, Fe²⁺, ROS and GSH kits were performed according to the manufacturers' protocols. The values for the levels of LDH, MDA, Fe²⁺, ROS and GSH in the cells, cell supernatant and liver tissues were measured using a microplate reader at the absorption wavelengths of 490, 532, 593, 490 and 412 nm.

The cells or liver tissues were homogenized by radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime Institute of Biotechnology) on ice to extract total protein. Protein concentration was determined by bicinchoninic acid (BCA) protein assay reagent assay kit (Beyotime Institute of Biotechnology). Protein lysates (30 µg) were subjected to 12% SDS-PAGE. The proteins were transferred to PVDF membranes (EMD Millipore). After blocking with 20% non-fat milk at room temperature for 1 h, the membranes were incubated with primary antibodies against GPX4 (1:1,000), Nrf2 (1:1,000), HO-1 (1:1,000), HMGB1 (1:1,000) and GAPDH (1:1,000) overnight at 4°C. The IRDye800CW secondary antibody (1:10,000) was incubated at 37°C for 1 h. The protein bands were visualized using the Odyssey Infrared Imaging System (version 3.0, LI-COR Biosciences).

For immunofluorescence, circular slides were placed in 24-well plates and were seeded with 1×10⁴ L02 cells/400 µl DMEM per plate. After treatments, the slides were fixed with 4% paraformaldehyde at 37°C for 30 min, permeabilized with 0.2% Triton X-100 (Beyotime Institute of Biotechnology) at 37°C for 20 min, blocked with 5% bovine serum albumin (BSA, Beijing Solarbio Science & Technology Co., Ltd.) at 37°C for 30 min and incubated with primary antibodies against HMGB1 (1:100) and GPX4 (1:100) at 4°C overnight. Slides were then incubated with Cy3 and FITC-labeled fluorescent secondary antibodies (1:100) at 37°C for 1 h in dark room. Then the sections were stained with 5 µg/ml DAPI (Beyotime Institute of Biotechnology) at 37°C for 5 min in a dark room. The slides were observed under a fluorescence microscope at ×200 magnification.

Mouse liver tissues were fixed in 4% paraformaldehyde solution at 37°C for 24 h and embedded in paraffin, then sectioned at 5 µm thickness. The sections were stained with Harris hematoxylin (Beyotime Institute of Biotechnology) at 37°C for 5 min and eosin (95% ethanol preparation, Beyotime Institute of Biotechnology) at 37°C for 15 sec. After staining, the liver tissues were observed under a light microscope at ×200 magnification. For the immunofluorescence detection of liver tissues, paraffin sections were dewaxed and hydrated, and immersed in citrate buffer for antigen retrieval. Sections were blocked with 5% BSA at 37°C for 30 min, and incubated with GPX4 (1:200) and HMGB1 (1:200) antibodies at 4°C overnight. Slices were then incubated with the Cy3 and FITC-labeled secondary antibody at 37°C for 1 h in a dark room. DAPI (5 µg/ml) was used to stain nuclei at 37°C for 5 min in dark room. Finally, the slices were imaged under an inverted fluorescence microscope at ×200 magnification.

All data are presented as the mean ± SEM. Data were analyzed using SPSS 17.0 statistical software (SPSS, Inc.). All experiments were repeated three times. The differences among three or more groups were analyzed using one-way ANOVA followed by Bonferroni's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The destruction of the cell membrane structure induced by ferroptosis causes the release of LDH in the cytoplasm into the culture medium. Therefore, the degree of liver damage can be determined by detecting the levels of LDH in the supernatant of cultured cells. As shown in Fig. 1A and B, when compared with the normal group, the cell viability in the L02 model group induced by TNF-α/D-GalN was significantly reduced, while the LDH level was significantly increased (P<0.01). Treatment with 1 or 2 mM GLY increased cell viability and decreased the level of LDH compared with the model group (P<0.01).

ROS is the general term for a class of highly reactive compounds that are composed of oxygen, including oxygen free radicals and non-radical oxygenated products (14). ROS accumulate following adverse external stimuli and can exceed the scavenging ability of the peroxidase system, leading to oxidative stress, triggering lipid peroxidation, damage to the cell membrane and finally inducing ferroptosis (15). MDA is a product of lipid peroxidation. It has strong biological toxicity and its production can aggravate biofilm damage. Therefore, it is an index for membrane lipid peroxidation, which indirectly reflects the degree of damage to a cell (16). In order to validate the effect of GLY on the antioxidant capacity of a TNF-α/D-GalN-induced L02 cell injury model, ROS and MDA levels were detected. As shown in Fig. 1C and D, when compared with the normal group, ROS and MDA levels in the model group were significantly increased (P<0.01). Treatment with GLY reduced the levels of ROS, MDA and LDH.

GSH is an important non-enzymatic antioxidant. GSH plays an important role in restoring liver function through removing lipid peroxides, including O²⁻, H₂O₂ and LOOH (17). Therefore, the level of GSH is an important factor in determining antioxidant capacity. GSH and Fe²⁺ are key regulators of ROS in the process of ferroptosis (18). In the process of ferroptosis, Fe²⁺ can convert lipid peroxide into ROS, and GPX4 converts lipid peroxide into the corresponding alcohol with the help of GSH (19). As shown in Fig. 1E and F, when compared with the normal group, the level of GSH was significantly decreased (P<0.01) and the Fe²⁺ level in the model group was significantly increased (P<0.01). After treatment with 1 or 2 mM GLY, the levels of GSH were significantly increased (P<0.01) and the levels of Fe²⁺ were significantly decreased (P<0.01) compared with the model group. Therefore, it can be concluded that GLY downregulates the levels of ROS by promoting GSH and inhibiting Fe²⁺, thereby alleviating ferroptosis.

GPX4 competitively inhibits the production of ROS; when the activity of GPX4 is inhibited, the effect of ROS on ferroptosis is more subtle (20). As shown in Fig. 2A-C, when compared with the normal group, the protein levels of GPX4, Nrf2 and HO-1 in the model group were significantly reduced (P<0.01), while the protein level of HMGB1 in the model group was increased (P<0.01). After treatment with GLY, the levels of GPX4, Nrf2 and HO-1 were significantly increased, while the protein level of HMGB1 in the model group was significantly decreased. When compared with the normal group, the level of GPX4 in the model group was markedly reduced (Fig. 2D), while the level of HMGB1 in the model group was markedly increased (Fig. 2E), as determined by immunofluorescence analysis. After treatment with 1 or 2 mM GLY, the level of GPX4 was increased, while the level of HMGB1 in model group was decreased.

As shown by the H&E staining in Fig. 3A, the liver tissue in the normal group was clear and intact, with the liver plates neatly arranged. There were no signs of degeneration, necrosis or inflammatory cell infiltration. In the ALF model group, the structure of the liver tissue was disordered and the hepatocytes were necrotic. Inflammatory cell infiltration was also observed. The degree of liver cell damage in the GLY treatment groups was reduced compared with the model group. Inflammatory cell infiltration was also reduced. The degree of liver damage was lower following the higher doses of GLY. As shown in Fig. 3B-D the serum levels of ALT, AST and TBIL in the model group were significantly higher than those in the control group (P<0.01). Following treatment with GLY, the serum levels of ALT, AST and TBIL were significantly lower than those in the ALF model group; however, they were still higher than the normal group.

As shown in Fig. 4A-C, when compared with the normal group, the levels of LDH, ROS and MDA in the model group were increased. After treatment with GLY, the levels of LDH, ROS and MDA were decreased.

As shown in Fig. 4D and E, when compared with the normal group, the level of GSH was significantly decreased (P<0.01), while the level of Fe²⁺ in the model group was significantly increased (P<0.01). After treatment with GLY, the level of GSH level was increased (P<0.01), while the level of Fe²⁺ significantly decreased (P<0.01).

As shown in Fig. 5A-C, when compared with the normal group, the expression levels of GPX4, Nrf2 and HO-1 in the model group were significantly decreased (P<0.01), while the expression of HMGB1 in the model group was significantly increased (P<0.01). After treatment with GLY, the levels of GPX4, Nrf2 and HO-1 were increased (P<0.01), while the level of HMGB1 was reduced. When compared with the normal group, the protein level of GPX4 in the model group was reduced, while the level of HMGB1 was increased, as determined by immunofluorescence (Fig. 5D). After treatment with GLY, the level of GPX4 was increased and the level of HMGB1 was decreased.