HMOX1 and BNIP3 work together to control nerve cell death after spinal cord injury through mitochondria cleanup

We further examined the changes in HMOX1 expression following OGD/R treatment. Both the western blotting and qRT-PCR results demonstrated that HMOX1 protein and mRNA levels increased significantly, peaking at 6 h of deprivation followed by 24 h of reoxygenation (Fig. 2N–P). The immunofluorescence results were consistent with these findings (Fig. 2Q). These data indicate that OGD/R injury triggers ferroptosis in PC12 cells, accompanied by upregulation of HMOX1.

Next, we assessed mitophagy flux. P62 immunofluorescence and TOMM20/LC3 co-staining results aligned with the immunoblotting data (Fig. 5G–J). MMP measurements showed a substantial decrease following HMOX1 overexpression, whereas HMOX1 knockdown restored MMP (Fig. 5K, L). Finally, transmission electron microscopy confirmed that HMOX1 knockdown reduced autophagosome formation, while HMOX1 overexpression increased it (Fig. 5M, N). These findings support the hypothesis that HMOX1 and BNIP3 interact to trigger the activation of excessive mitophagy.

Next, we investigated the changes in ferroptosis. Western blotting revealed that the OGD/R + LV-shHMOX1+LV-BNIP3 group exhibited upregulated ACSL4 expression and downregulated FTH1 and GPX4 expression, which was counteracted by mdivi-1 treatment (Fig. 6H, I). Similarly, LV-shHMOX1+LV-BNIP3 treatment elevated MDA levels and reduced GSH levels, both of which were normalized by mdivi-1 administration (Fig. 6J). Mito-FerroGreen is a fluorescent probe designed specifically to detect Fe²⁺ in mitochondria. FerroOrange, Mito-FerroGreen, and C11 BODIPY staining were consistent with the above findings. Specifically, the overexpression of BNIP3 following HMOX1 knockdown increased mitochondrial Fe²⁺ levels and induced lipid peroxidation. This effect could be reversed by mdivi-1 (Fig. 6K–P). These findings demonstrate that BNIP3-mediated mitophagy contributes significantly to HMOX1-induced ferroptosis in neuronal cells.

Adult female Sprague-Dawley rats (200-250 g) were provided by the Experimental Animal Center of Lanzhou University. The study was approved and overseen by the Ethics Committee of Lanzhou University (approval number: D2024-289). All surgical procedures were performed under sterile conditions. Anesthesia was induced by intraperitoneal pentobarbital injection (40 mg/kg). First, the left femoral vein and right femoral artery were isolated and cannulated for fluid return, blood withdrawal, and blood pressure monitoring. Next, a 1.0 cm segment of the spinal cord centered at T10 was exposed through layered dissection. Hemorrhage was induced by withdrawing 30% of the total blood volume (duration: 10 min, rate: 0.5 ml/min). The blood volume withdrawn (ml) was calculated as the body mass (g)×7.4%×30%. Blood was collected and stored in a heparinized sterile syringe at 4°C. Resuscitation was initiated after 45 min of hemorrhage. Somatosensory evoked potential (SEP) and motor evoked potential (MEP) were monitored throughout the experimental procedure. Manual bladder expression was performed every 8 h postoperatively until reflex bladder emptying returned.

In the sham group, the wound was closed directly following inguinal vascular isolation and removal of the T10 spinous process and lamina to expose the dura mater. In the SCIRI + LV-shNC, SCIRI + LN-shHMOX1, and SCIRI + LN-shHMOX1+Z-VAD-FMK groups, the respective lentiviral vectors and Z-VAD-FMK (10 μM; Selleckchem) were administered intrathecally after SCIRI. Spinal cord tissues were harvested 48 h after surgery for subsequent experiments.

Rats were maintained on a 12-h light-dark cycle with ad libitum access to food and water. All rats were acclimated in the testing chamber or apparatus for 1 h before behavioral testing. Neurological function was scored using the Basso Mouse Scale (BMS) at predetermined time points, beginning on day 1 after injury.

Forty-eight hours after surgery, rats were anesthetized with 4% pentobarbital sodium and underwent systemic perfusion with phosphate-buffered saline (PBS) and 4% paraformaldehyde. The isolated spinal cord tissues were fixed in 4% paraformaldehyde for 24 h, followed by ethanol gradient dehydration and paraffin embedding. Sections (5 μm) were prepared for hematoxylin-eosin (HE) and luxol fast blue (LFB) staining (Solarbio).

PC12 cells (Shanghai Institute of Biochemistry and Cell Biology, Shanghai, China) were routinely cultured in high-glucose Dulbecco's modified Eagle's medium (DMEM; Meilun Biotech) supplemented with 10% fetal bovine serum in an incubator at 37°C with 5% CO₂. Cells were cultured for 24 h with or without the ferroptosis inducer Erastin (2 μM; PeproTech), the mitophagy inducer rapamycin (80 nM; Selleckchem), the mitophagy inhibitor mdivi-1 (40 μM; Selleckchem), and the iron chelating agent deferoxamine (DFO; 10 μM; Selleckchem) for 24 h. Cells were collected when confluence reached 70%–80%.

When the PC12 cell density reached approximately 30% in culture flasks, lentivirus (GeneChem) was transfected at a multiplicity of infection of 100. The medium was replaced 48 h after transfection. HMOX1 and BNIP3 expression levels were quantified using qRT-PCR and western blotting.

An OGD/R model was created to simulate in vitro SCIRI using established protocols [37]. PC12 culture medium was replaced with glucose-free DMEM, and the cells were subjected to OGD in a tri-gas incubator (94% N₂, 5% CO₂, 1% O₂; 37°C) for 2, 4, 6, or 8 h. Subsequently, the medium was replaced with high-glucose DMEM, and the cells were transferred to a regular incubator (5% CO₂; 37°C) for 24 h of reoxygenation. The control group cells were cultured in high-glucose in a regular incubator for the same duration.

Cell viability and the levels of malondialdehyde (MDA) and glutathione (GSH) in tissues or cell lysates were evaluated using a Cell Counting Kit-8 (CCK-8) assay (Dojindo), MDA assay kit (Beyotime), and GSH assay kit (Beyotime). The absorbance was measured using a BioTek microplate reader (Agilent Technologies).

Changes in mitochondrial membrane potential (MMP) changes were assessed using the JC-1 MMP Assay Kit (Beyotime). Cells were incubated with JC-1 staining solution at 37°C for 20 min in a cell culture incubator, rinsed twice with JC-1 staining buffer, and imaged using a fluorescence microscope (Olympus).

Intracellular and mitochondrial Fe²⁺ and lipid peroxidation levels in live cells were detected using FerroOrange (Dojindo), Mito-FerroGreen (Dojindo), and C11 BODIPY (Thermo Fisher Scientific) fluorescence probes. Cells were seeded on coverslips, treated, and stained according to the manufacturer's instructions. Subsequently, cells were imaged using a fluorescence microscope (Olympus).

Total RNA was extracted using TRIzol (Takara) and reverse-transcribed using the PrimeScript RT Master Mix Kit (Takara). HMOX1, ACSL4, FTH1, and GPX4 expression levels were quantified using the SYBR PrimeScript RT-PCR Kit (Takara), and target mRNA expression levels were normalized to GAPDH. Relative expression levels were calculated using the 2^−ΔΔCT method. The primers used for qRT-PCR are listed in Table S1.

Total proteins were extracted from PC12 cells and rat spinal cord tissues using RIPA lysis buffer (Beyotime). Protein concentrations were measured using a BCA Protein Assay Kit (Beyotime) and analyzed by SDS-PAGE. Antibodies used in this study were HMOX1 (1:2000, Abcam), ACSL4, FTH1, GPX4, LC3, COXIV, TOMM20, and cleaved-caspase 3 (all 1:1000, Abmart), BNIP3 (1:1000, Cell Signaling Technology), P62 (1:5000, Abmart), Flag-Tag (1:2000, Cell Signaling Technology), and β-actin (1:1000, Beyotime).

Paraffin sections were dewaxed, hydrated, and antigen retrieved. Cells were fixed in 4% paraformaldehyde for 30 min and permeabilized with 0.2% Triton X-100 for 15 min. The subsequent staining procedures were identical for both tissue and cell samples. After blocking with 10% goat serum at 37°C for 1 h, the samples were incubated with primary antibodies at 4°C overnight. After washing, the samples were incubated with fluorescent secondary antibodies at 37°C for 1 h. Nuclei were stained with DAPI, and imaged using a fluorescence microscope (Olympus).

PC12 cells were washed twice with PBS, harvested, and lysed with IP binding buffer and protease inhibitor (Beyotime). An aliquot was reserved for western blotting, while the remaining lysate was incubated with specific antibodies and protein A/G beads using a BeaverBeads Protein A/G Immunoprecipitation (IP) Kit (Beaver Bioscience), gently mixed to ensure homogeneous bead-antibody complexes, and incubated overnight at 4 °C on a rotator. Magnetic beads were separated using a magnetic stand. Thereafter, beads were boiled in sodium dodecyl sulfate loading buffer and analyzed by western blotting.

Total protein was extracted from PC12 cells, and IP was performed as described using specific antibodies and protein A/G-agarose beads, followed by MS (Thermo Fisher) analysis of the isolated immunoprecipitates.

Statistical analysis was conducted using GraphPad Prism (version 9.0), with all experiments performed in at least triplicate. The results are presented as the mean ± standard deviation. Student's t-test was employed for two-group comparisons, while one-way analysis of variance was used for multi-group analyses. P < 0.05 was considered statistically significant.