Ferrostatin-1 (Fer-1; cat. no. S7243), liproxstatin-1 (cat. no. S7699), carbobenzoxy-valyl-alanyl-aspartyl-[O-m ethyl]-fluoromethylketone (Z-VAD-FMK; cat. no S7023), necrostatin-1 (cat. no. S8037), 3-methyladenine (3-MA; cat. no. S2767), SB203580 (cat. no. S1076), SP600125 (cat. no. S1460), and SCH772984 (cat. no. S7101) were obtained from Selleck Chemicals (Houston, TX, USA). N-acetyl-cysteine (NAC; cat. no. A0737), deferoxamine (DFO; cat. no. D9533), FeCl₃ (cat. no. 157740), buthionine-sulfoximine (BSO; cat. no. 19176), Iron Stain kit (cat. no. HT20), GSH (cat. no. G6013), DNase I (cat. no. AMPD1), hyaluronidase type III (cat. no. H3506), collagenase I (cat. no. C0130), insulin-transferrin-selenium (ITS; cat. no. 00031626), epidermal growth factor (EGF; cat. no. E4127), testosterone (cat. no. T6147), follicle stimulating hormone (FSH; cat. no. I9149), ponasterone (cat. no. P3490), neopterin (cat. no. N3386), diphenyleneiodonium chloride (DPI; cat. no. D2926), diethyldithiocarbamate (DETC; cat. no. 228680), 3-amino-1,2,4-triazole (ATZ; cat. no. A8056), Lipid Peroxidation malondialdehyde (MDA) Assay kit (cat. no. MAK085), Glutathione Peroxidase Cellular Activity Assay kit (cat. no. CGP1), Quantification kit for oxidized and reduced GSH (cat. no. 38185), Iron Assay kit (cat. no. MAK 025), and Caspase 3 Assay kit (cat. no. CASP3C) were obtained from Sigma Aldrich (Merck KGaA, Darmstadt, Germany) The following reagents were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA): Ferroportin-1 short interfering (si)RNA (cat. no. sc-60634), XCT siRNA (cat. no. sc-76934), GPx-4 siRNA (cat. no. sc-63302), control siRNA (cat. no. sc-37007), siRNA Reagent System (cat. no. sc-45064), ferroportin-1 CRISPR Activation Plasmid (cat. no. sc-424774-ACT), GPx-4 CRISPR Activation Plasmid (cat. no. sc-436949-ACT), GPx-1 CRISPR Activation Plasmid (cat. no. sc-420662-ACT), anti-rabbit/mouse IgG horseradish peroxidase conjugate (HRP; cat. no. sc-2379 and sc-2380) and antibodies against transferrin (Tf; cat. no. sc-393595), transferrin receptor 1/cluster of differentiation 71 (TFR1; cat. no. sc-32272), divalent metal transporter 1/natural resistance-associated macrophage proteins (DMT1; cat. no. sc-166884), ferritin (cat. no. sc-25617), GPX4 (cat. no. sc-166120), ferroportin-1 (Fpn; cat. no. sc-49668), and β-actin (cat. no. sc-58671). Antibodies against p38 mitogen-activated protein kinase (MAPK; cat. no. 9212), extracellular signal-regulated kinase1/2 (ERK1/2; cat. no. 9102), c-Jun NH2-terminal kinase (JNK; cat. no. 9252), phosphorylated (p)-p38 MAPK (cat. no. 4511), p-ERK1/2 (cat. no. 9101), p-JNK (cat. no. 9255), and XCT (SLC7A11; cat. no. 12691) were purchased from Cell Signaling Technology (Danvers, MA, USA). TRIzol reagent was obtained from Invitrogen (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The reverse transcription (RT) system and One Step SYBR PrimeScript RT-quantitative polymerase chain reaction (PCR) kit (cat. no. RR066) were purchased from Takara Bio, Inc. (Otsu, Japan). CellTiter 96 AQueous Non-Radioactive Cell Proliferation Assay kit (G cat. no. 5421) was obtained from Promega Corp. (Madison, WI, USA).

TM4 mouse Sertoli cells (cat. no. CM-1912; American Type Culture Collection, Manassas, VA, USA) were cultured in Dulbecco's modified Eagle medium/nutrient mixture F-12 (DMEM/F-12; Gibco; Thermo Fisher Scientific, Inc.) medium supplemented with 5% fetal bovine serum, 0.1% ITS, 1 µg/ml EFG, 0.1 µM testosterone, 25 U/l FSH, 100 units/ml penicillin, and 100 µg/ml streptomycin at 37°C in an atmosphere containing 5% CO₂ for 72 h. The I/R cell model was established as previously described (27,28). TM4 cells were cultured in airtight bottles with ischemic buffer (5.37 mM KCl, 136.89 mM NaCl, 0.44 mM KH₂PO₄, 0.338 mM Na₂HPO₄, 4.166 mM NaHCO₃, and 5 mM D-glucose; pH 6.8) at 37°C in a hypoxic incubator with 5% CO₂, 1% O₂ and 94% N₂ for 2 h. The medium was subsequently removed, normal routine culture was resumed and reoxygenation was initiated, as described above. The I/R cell model was the OGD/R model of TM4 cells. The cells of OGD/R model were set as OGD/R group and normal TM4 cells were set as control group.

Ferroportin-1 and GPx-4 knockdown was achieved by transfection with siRNA targeting ferroportin-1 and GPx-4 (si-ferroportin-1 and si-GPx-4; Santa Cruz Biotechnology, Inc.) using an siRNA Transfection Reagent (Santa Cruz Biotechnology, Inc.) with a control siRNA as the control (Santa Cruz Biotechnology, Inc.). Overexpression of ferroportin-1 and GPx-4 were achieved by transfection with ferroportin-1, GPx-4 and GPx-1 overexpression vectors (ferroportin-1, GPx-4, and GPx-1 CRISPR Activation Plasmid; Santa Cruz Biotechnology, Inc.) using Plasmid Transfection Medium (Santa Cruz Biotechnology, Inc.), with the CRISPR/Cas9 Plasmid used as the control. At 7 h post-transfection, the medium was changed to whole medium and cells were cultivated for an additional 24 h. The changes in ferroportin-1, GPx-4, and GPx-1 were measured using RT-qPCR, as described below.

Cell viability was analyzed using the CellTiter 96 AQueous Non-Radioactive Cell Proliferation Assay kit according to the manufacturer's protocol. TM4 cells were seeded onto 96-well plates (1×10³ cells/well), and 20 µl MTS/phenazine methosulfate was pipetted into each well. The plate was incubated for 2 h at 37°C, and then the absorbance was measured at 490 nm using a Spectramax M2 Multi-Mode Microplate Reader (Molecular Devices, LLC, Sunnyvale, CA, USA) to calculate the percentage of surviving cells.

Caspase-3 activity in TM4 cells was assessed using the Caspase-3 Activity Assay kit according to the manufacturer's protocol.

TM4 cells were lysed with MDA lysis buffer from the Lipid Peroxidation malondialdehyde (MDA) Assay kit and centrifuged at 8,000 × g for 10 min at 4°C to remove insoluble material. The level of MDA in TM4 cell lysates was assessed using a Lipid Peroxidation MDA Assay kit according to the manufacturer's protocol. The absorbance was measured at 532 nm using a spectrophotometer.

The expression of reduced and oxidized forms of GSH was analyzed using a quantification kit for oxidized and reduced GSH according to the manufacturer's protocol. Cultured TM4 cells were washed three times with PBS, 5 min each time mixed with assay buffer and substrate solution from the Quantification kit for oxidized and reduced GSH, and incubated at room temperature in a fluorometer. Absorbance was assessed at 412 nm and the results were expressed as a percent of the control.

GPX activity in the cell lysates was measured using the Glutathione Peroxidase Cellular Activity Assay kit according to the manufacturer's protocol. The absorbance was measured at 405 or 415 nm using a microplate reader. The values were expressed as a percentage of non-treated cells.

The intracellular iron concentration was measured using an Iron Assay kit according to the manufacturer's protocol. An iron reducer was added to cell lysates to reduce ferric iron (Fe³⁺) to ferrous iron (Fe²⁺) and incubated for 30 min at room temperature in the dark. The iron reaction produced a colored complex and absorbance was measured at 593 nm.

Total cellular RNA was isolated from the cells in the control and OGD/R groups using TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The RNA was reverse transcribed (42°C for 10 min, 95°C for 5 min, 4°C for 3 min, 5 cycles) into cDNA using an RT system. qPCR was performed using a One Step SYBR Green PrimeScript RT-qPCR kit with denaturation at 95°C for 15 min, 50 cycles at 95°C for 15 sec, 72°C for 15 sec, 95°C for 30 sec and 60°C for 15 sec. PCR was performed in triplicate and normalized using the ABI 7500 Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The primers used were as follows: Fpn, forward 5′-AGCCAAGATGGCACTAAGCAC-3′ and reverse 5′-TCTATGTTATCGAACAGACAT-3′; GPX4, forward 5′-AGCCAAGACCGAAGTAAACTACAC-3′ and reverse 5′-GGATCTTCATCCAGTTCCACAG-3′; GAPDH, forward 5′-CAAGGTCATCCATCCATGACAACTTTG-3′ and reverse 5′-GTCCACCACCCTGTTGCTGTAG-3′. GAPDH was used as the reference gene. Relative gene expression was analyzed by 2^−ΔΔCq method (28).

Cells in the control and OGD/R groups were lysed in cell lysis solution containing 1 mM phenyl-methane sulfonyl-fluoride. The total protein concentration was determined with bicinchoninic acid assay kit (Thermo Fisher Scientific, Inc.). Protein lysates (40 µg/lane) were separated by 10–12% SDS-PAGE and transferred onto 0.22 µm polyvinyli-dene difluoride membranes (EMD Millipore, Billerica, MA, USA). The membranes were blocked with 3% bovine serum albumin (Sigma-Aldrich; Merck KGaA) in Tris-buffered saline-0.05% Tween (TBST) for 1 h at 37°C. Membranes were subsequently incubated with antibodies against Fpn (1:100), GPX4 (1:200), p38 (1:500), p-p38 MAPK (1:500), ERK1/2 (1:500), p-ERK1/2 (1:500), JNK (1:500), p-JNK (1:500), Tf (1:200), TFR1 (1:200), DMT1 (1:200), ferritin (1:100), SLC7A11 (1:100) and β-actin (1:1,000) at 4°C overnight. The membranes were washed three times with TBST and incubated with horseradish peroxidase-conjugated goat anti-rabbit/mouse IgG for 1 h at room temperature the following morning. The blots were visualized using an enhanced chemiluminescence detection kit (Thermo Fisher Scientific, Inc.). The band density of target proteins was quantified and normalized to that of β-actin using Quantity One software (version 4.6.2, Bio-Rad Corporation, Hercules, CA, USA).

All data were processed with the using SPSS 19.0 (IBM Corp., Armonk, NY, USA). Data are presented as the mean ± standard deviation from three independent experiments. One-way analysis of variance with Tukey's multiple comparison post hoc test was employed for comparison among groups. P<0.05 was considered to indicate a statistically significant difference.

Testicular I/R injury has been reported to induce cell death in germ cells, but not in Sertoli cells (19,28,29). It was therefore examined whether I/R was able to induce cell death in Sertoli cells. TM4 cells in control and OGD/R groups were analyzed and a notable increase in cell death was observed in the OGD/R group compared with the control (P<0.01; Fig. 1A). I/R stress activates various types of cell death, including apoptosis, necrosis, autophagy, and ferroptosis-mediated cell death (30). Recent studies have indicated that I/R induces ferroptosis, but not other forms of cell death, in nephrotic tubular cells and hepatocytes (8,31,32). Cells were incubated with the cell death inhibitor Z-VAD-FMK, ferroptosis inhibitor Fer-1, necrosis inhibitor necrostatin-1 and autophagy inhibitor 3-MA and cell death was evaluated. The results indicated that OGD/R-induced cell death was ameliorated by Fer-1 compared with the control group (P<0.01; Fig. 1A); however, no significant effect was observed with inhibitors of apoptosis, necrosis or autophagy. OGD/R did not significantly promote the activity of apoptotic marker caspase 3 (Fig. 1B). The results indicate that ferroptosis serves a role in the accumulation of lipid peroxidation and iron; lipid ROS levels were increased in the OGD/R group compared with the control, and this effect was ameliorated by Fer-1 (P<0.01; Fig. 1C). To evaluate the effect of iron on OGD/R-induced cell death, the iron chelator DFO was used to treat Sertoli cells; the results demonstrated that cell death and iron levels were reduced (P<0.01; Fig. 1A and D). Taken together, these results suggest that ferroptosis contributes to OGD/R-induced growth inhibition in Sertoli cells, whereas cell death, necroptosis and autophagy do not.

Prior studies have reported that ferroptosis is characterized by an increase in lethal lipid peroxidation products, which exhibit cytotoxicity (4,33). Lipid ROS levels were assessed and the results revealed that lipid ROS were significantly increased in the OGD/R group compared with the control group (P<0.01; Fig. 2A). Lipid ROS generation was inhibited by liproxstatin-1 and NAC. Liproxstatin-1 and NAC both significantly reduced ROS generation (P<0.01; Fig. 2A). It was also observed that NAC and liproxstatin-1 effectively inhibited cell death (P<0.01; Fig. 2B). These results suggest that lipid ROS is required for OGD/R-induced ferroptotic cell death.

The production of lipid ROS is required for ferroptosis and is associated with multiple processes, including NADPH-mediated lipid peroxidation, iron-dependent ROS generation through the Fenton reaction and GSH depletion (4,23,25). To ascertain whether NADPH oxidases contribute to lipid ROS production, TM4 cells were treated with NADPH oxidase inhibitors, neopterin and DPI; lipid ROS levels were unaffected. However, treatment with the iron chelator DFO effectively blocked OGD/R-induced lipid ROS production (P<0.01; Fig. 3A) compared with untreated cells, suggesting that lipid ROS generation is associated with iron content. Recent studies have demonstrated that iron overload injures Sertoli cells (34). Cells were subsequently treated with exogenous Fe (FeCl₃) to confirm that iron affected OGD/R-induced ferroptosis and the results revealed that lipid ROS levels and cell death were further increased (P<0.01; Fig. 3A and B). A reduction in lipid ROS levels and cell death was also observed in cells cultured with DFO compared with the vehicle (P<0.01; Fig. 3A and B). These data suggest that ROS production results from iron accumulation and that iron may serve a crucial role in OGD/R-induced ferroptosis.

To explore whether lipid ROS generation partially reflects GSH depletion, GSH levels in cells were assessed. GSH content was significantly reduced in the OGD/R group compared with the control (P<0.01; Fig. 4). Cells were treated with GSH or the GSH precursor NAC and cell viability and lipid ROS levels were assessed. The results revealed that GSH or NAC treatment significantly prevented OGD/R-induced lipid ROS generation (P<0.01), whereas GSH had no significant effect on OGD/R-induced cell death (Fig. 4B and C). It was also observed that the protein level of XCT (SLC7A11), a cysteine transport receptor, was reduced following OGD/R injury (P<0.01; Fig. 4D). Furthermore, siRNA-mediated knockdown of XCT, the upstream activator of GSH, reduced the GSH content and markedly increased cell death and lipid ROS levels (P<0.01; Fig. 4A–C). These data indicate that GSH depletion is consistent with OGD/R-induced production of lipid ROS, resulting in ferroptosis. These findings suggest that ROS generation results from iron accumulation and GSH depletion.

Increased iron uptake and reduced iron output through iron metabolism proteins both lead to iron overload during the ferroptotic cell death process (4,35). First, it was determined which iron regulatory proteins were expressed in Sertoli cells using immunocytochemistry and western blotting. As presented in Fig. 5A, the following factors were expressed as normal in Sertoli cells: Tf, which binds ferric iron (Fe³⁺); TFR1, which imports iron into cells; DMT1, which mediates ferrous iron (Fe²⁺) release into the cytoplasm; cytosolic ferritin, which stores excess iron; and the membrane protein Fpn, which mediates iron export (35). The expression of Fpn was markedly reduced in the OGD/R group (P<0.01), and no alterations were observed in the expression of Tf, TFR1, DMT1 or ferritin. These results indicate that OGD/R may reduce iron export to induce ferroptosis by decreasing Fpn expression. Previous studies have revealed that Fpn mutation may lead to iron overload and subsequent diseases accompanied by normal or low transferrin expression (4,36). To further demonstrate whether Fpn inhibits OGD/R-induced lipid ROS and ferroptosis, Sertoli cells were transfected with the Ferroportin-1 CRISPR Activation Plasmid, treated with the Fpn activator ponasterone and subsequently analyzed for ROS and iron levels. Cells transfected with the Ferroportin-1 CRISPR activation plasmid or treated with ponasterone exhibited increased protein and mRNA expression of Fpn and blocked accumulation of lipid ROS and iron (P<0.01; Fig. 5). Cell death was also significantly decreased (P<0.01; Fig. 5F). By contrast, knockdown of Fpn with ferroportin-1 siRNA increased the OGD/R-induced lipid ROS level and iron content, but not cell death (P<0.01; Fig. 5D–F). Taken together, these results indicate that Fpn overexpression inhibited OGD/R-induced ferroptosis in Sertoli cells by reducing the content of iron and generation of lipid ROS and that the Fpn inducer ponasterone served the same role.

Considering that GSH depletion is crucial for OGD/R-induced ROS generation, it was examined how OGD/R-induced GSH depletion may lead to ferroptosis in Sertoli cells. The cells were treated with the GSH synthesis inhibitor BSO and antioxidant inhibitors, including the superoxide dismutase (SOD) inhibitor DETC and the catalase inhibitor ATZ, and subsequently, examined for cell death and GSH levels. The results revealed that inhibitors induced cell death and that the GSH content was not significantly reduced by these antioxidant inhibitors compared with BSO (Fig. 6A and B). The results indicated that altered downstream GSH depletion was likely responsible for the induction of ferroptotic cell death. It was considered that GPX4 inactivation due to GSH depletion induces ferroptosis by accumulating lipid ROS from lipid peroxidation. Previous studies have demonstrated that GPX4 regulates ferroptosis via indirect or direct targeting of lipid peroxidation and iron metabolism (4). Furthermore, inactivation of GPX4 triggers ferroptotic cell death in renal tubular epithelial cells via lipid peroxidation (24,37,38). To determine whether GPX4 expression is decreased in the cell model of OGD/R-induced testicular damage, the total activity of GPXs was examined as well as GPX4 expression using Western blotting. Notable decreases in GPXs activity and GPX4 expression were observed (P<0.01; Fig. 6C and D). Furthermore, treatment with GSH or NAC revealed that these molecules may activate GPXs (P<0.01; Fig. 6C). These results further indicate that GSH depletion inactivates GPXs with low GPX4 protein expression, resulting in the production of lipid ROS.

GPX4 has been described as the major regulator of erastin-induced cancer cell ferroptosis and OGD/R-induced ferroptotic renal tubular epithelial cell death (24,37,39–41). A previous study demonstrated that GPX4 is a prerequisite for sperm development (42). It was therefore investigated whether GPX4 regulates OGD/R-induced ferroptotic cell death in Sertoli cells. The results revealed that knocking down GPX4 with GPX-4 siRNA reduced the mRNA and protein levels of GPX4 and increased cell death as well as lipid ROS compared with a vector (P<0.01; Fig. 7A–D). GPX4 expression was subsequently upregulated in cells using the GPx-4 CRISPR Activation Plasmid and it was observed that GPX4 mRNA and protein levels were increased and OGD/R-induced ferroptosis and lipid ROS were decreased (P<0.01; Fig. 7A–D). The role of GPX1, a soluble hydroperoxide reducer, was also examined. GPX1 overexpression partially inhibited OGD/R-induced cell death without affecting lipid ROS generation (Fig. 7E and F), indicating that GPX1 may not scavenge additional ROS species. Taken together, these data indicate that GPX4 is inactivated following testicular OGD/R injury via GSH depletion and blocks OGD/R-induced ferroptosis by reducing lipid ROS.

Previous studies have reported that MAPK signaling is associated with erastin-induced ferroptosis in cancer cells and OGD/R-induced testicular injury (4,43,44). However, whether the MAPK signaling pathway serves a role in OGD/R-induced ferroptosis in Sertoli cells remains unclear. Three major MAPKs, p38 MAPK, ERK1/2 and JNK, which are involved in I/R-induced ferroptosis, were investigated. OGD/R promoted the phosphorylation of p38, but not JNK and ERK1/2 (P<0.01; Fig. 8A). In addition, NAC prominently depressed OGD/R-induced p38 MAPK phosphorylation (P<0.01; Fig. 8B). Cells were treated with the p38 activation inhibitor SB203580, JNK phosphorylation inhibitor SP600125 or ERK1/2 upstream activator inhibitor SCH772984 for 1 h and subsequently examined. OGD/R-induced cell death was blocked by SB203580 (P<0.01; Fig. 8C), indicating that p38 MAPK may be associated with OGD/R-induced ferroptosis in Sertoli cells. By contrast, SP600125 and SCH772984 had no effect on OGD/R-induced cell death (Fig. 8C). The OGD/R-induced p38 phosphorylation was notably inhibited by SB203580 (P<0.01; Fig. 8B). Furthermore, knocking down a major isoform of p38 using p38α siRNA reversed OGD/R-induced ferroptosis in Sertoli cells (P<0.01; Fig. 8C). Collectively, these results suggest that p38 MAPK participates in OGD/R-induced ferroptotic cell death.