CYBB-Mediated Ferroptosis Drives Podocyte Injury in Lupus Nephritis and Represents a Therapeutic Target

Abstract

Objective

Lupus nephritis (LN) is a major cause of kidney failure in systemic lupus erythematosus, with podocyte injury being a key determinant of proteinuria and poor renal outcome. Ferroptosis, an iron-dependent form of regulated cell death, has been implicated in kidney diseases, but its role in LN remains unclear.

Methods

We integrated in vivo experiments using MRL/lpr mice, in vitro assays with immortalized podocytes (MPC5), and transcriptomic analysis of human glomerular datasets (GSE32591). Ferroptosis involvement was evaluated by ferrostatin-1 (Fer-1) treatment, measurement of ferroptosis markers, and assessment of podocyte proteins. Bioinformatic analyses (differential expression, WGCNA, LASSO regression, and FerrDb integration) identified candidate ferroptosis regulators, followed by functional validation of CYBB via siRNA knockdown and overexpression.

Results

Progressive nephritis in MRL/lpr mice showed iron overload, lipid peroxidation, glutathione depletion, and GPX4 downregulation, leading to podocyte loss and proteinuria. Fer-1 treatment markedly ameliorated renal pathology and preserved podocyte integrity. Human LN datasets identified CYBB as a ferroptosis-related hub gene upregulated in disease. CYBB expression correlated with renal dysfunction and oxidative injury, while in vitro assays confirmed that CYBB overexpression enhanced ROS generation and ferroptotic podocyte damage, whereas CYBB knockdown or Fer-1 reversed these effects.

Conclusion

These findings identify CYBB-mediated ferroptosis as a key driver of podocyte injury in LN. By promoting ROS generation and lipid peroxidation, CYBB serves as a mechanistic link between oxidative stress and ferroptotic cell death. Both pharmacological and genetic inhibition of CYBB mitigated ferroptosis, preserved podocyte integrity, and improved renal function, highlighting CYBB as a promising therapeutic target in lupus nephritis.

Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by the production of autoantibodies and the formation of immune complexes, leading to inflammation and damage in multiple organs.¹ Lupus nephritis (LN) remains one of the most severe manifestations, affecting up to 60% of SLE patients and representing a major cause of morbidity and mortality.^2,3 Despite advances in immunosuppressive therapies, a significant portion of patients fail to achieve complete remission and progress to end-stage renal disease (ESRD), highlighting the urgent need for novel therapeutic strategies based on a deeper understanding of LN pathogenesis.^4,5

Podocyte injury is central to the development of proteinuria, a hallmark of poor renal prognosis in LN.⁶ As terminally differentiated cells critical for maintaining the glomerular filtration barrier, podocyte depletion or dysfunction directly contributes to glomerulosclerosis and renal failure.⁷ While immune complex deposition, complement activation, and pro-inflammatory cytokines have been traditionally implicated in podocyte damage in LN.⁸ However, recent evidence suggests that regulated cell death (RCD) pathways may also play a pivotal role in podocyte loss, although their contribution to LN has not been fully elucidated.⁹

Indeed, increasing evidence supports a link between ferroptosis and lupus-related renal injury, with studies reporting iron overload, glutathione depletion, and elevated lipid peroxidation in LN patients and lupus-prone mice.^10,11 Furthermore, ferroptosis inhibition using ferrostatin-1 (Fer-1) has been shown to reduce kidney damage in lupus models.¹² However, the upstream regulatory mechanisms of ferroptosis, particularly in glomerular podocytes, remain poorly defined. It is unclear how oxidative stress signaling interfaces with ferroptosis to promote podocyte dysfunction during LN progression.

In this study, we investigated the contribution of ferroptosis to podocyte injury in LN using lupus-prone MRL/lpr mice and immortalized podocytes. By integrating transcriptomic profiling with experimental validation, we identified CYBB as a ferroptosis-associated gene markedly upregulated in LN. We further explored its mechanistic role and therapeutic relevance by modulating ferroptosis pharmacologically with Fer-1 and genetically through CYBB silencing or overexpression. These findings provide new mechanistic insight into ferroptosis-mediated podocyte injury and establish CYBB as a potential therapeutic target in lupus nephritis.

Materials and Methods

Animals and Experimental Design

Female MRL/lpr mice and age-matched C57BL/6 controls were purchased from Changzhou Cavens Laboratory Animal Co., Ltd (CAVENS) and maintained under specific pathogen–free conditions at the laboratory animal center of The First Affiliated Hospital of Nanchang University with controlled temperature, humidity, and a 12-h light/dark cycle, with free access to food and water. All experimental procedures were approved by the Institutional Animal Care and Use Committee of The First Affiliated Hospital of Nanchang University (approval no. CDYFY-IACUC-202509GR038) and were in accordance with the NIH Guide for the Care and Use of Laboratory Animals. To evaluate LN progression and therapeutic intervention, mice were assigned to four groups (n = 5 per group): 8-week-old MRL/lpr mice as pre-LN controls (SLE group), 16-week-old MRL/lpr mice with established nephritis as the LN group, and age-matched 16-week-old C57BL/6 mice as healthy controls. A subset of 16-week-old MRL/lpr mice (Fer-1 group) received daily intraperitoneal injections of ferrostatin-1 (Fer-1, 1 mg/kg, MedChemExpress, USA) for 4 weeks, while the remaining groups received vehicle (saline). At the end of treatment, mice were deeply anesthetized with an intraperitoneal injection of pentobarbital sodium (50 mg/kg, Sigma-Aldrich, USA). After confirming the loss of reflexes, blood and kidney samples were collected under anesthesia. Euthanasia was completed by administering an additional dose of pentobarbital sodium to ensure death. All procedures complied with the AVMA Guidelines for the Euthanasia of Animals (2020) and were approved by the Institutional Animal Care and Use Committee of The First Affiliated Hospital of Nanchang University.

Cell Culture and Treatments

The conditionally immortalized mouse podocyte line (MPC5) was obtained from Cellverse Co., Ltd. (Shanghai, China). Cells were maintained in DMEM (Cellverse, Shanghai, China) supplemented with 10% fetal bovine serum (FBS; Pricella, Wuhan, China) and 1% penicillin–streptomycin (Solarbio, Beijing, China) at 37 °C in a humidified incubator with 5% CO₂. Differentiated podocytes were used for subsequent experiments. To mimic the LN microenvironment, cells were exposed to 10% serum obtained from LN for 48 h, with or without pretreatment with ferrostatin-1 (Fer-1, 1 μmol/L) for 2 h. Serum was collected from multiple 16-week-old MRL/lpr mice and pooled to minimize inter-individual variability. For the control group, 10% serum from age-matched C57BL/6 mice was used. To explore the functional role of CYBB, cells were transfected with CYBB-targeting siRNA or overexpression plasmids (Genecreate, Wuhan, China) using Lipofectamine 3000 (Beyotime, Shanghai, China). After 6 h of transfection, the medium was replaced with DMEM containing 10% LN sera, and the cells were cultured for an additional 48 h. In designated groups, Fer-1 was added 4 hours after plasmid transfection and used to pretreat cells for 2 hours before replacing the medium with LN serum-containing medium. Samples were collected for biochemical and molecular analyses at the end of the treatment period.

Renal Function and Autoantibody Measurements

Urine samples were collected over 24 hours using metabolic cages. Protein concentration in urine was quantified using the Bradford method (Jiancheng, Nanjing, China). Serum creatinine (SCr) levels were measured using the sarcosine oxidase method (Jiancheng, Nanjing, China). Circulating anti-dsDNA antibody levels were determined using an ELISA kit (Wuhan Shenke Experimental Technology Co., Ltd., Wuhan, China).

Histology and Immunostaining

Kidney tissues were fixed in 4% paraformaldehyde, routinely processed, embedded in paraffin, and sectioned at 3–4 μm thickness. Paraffin sections were stained with hematoxylin and eosin (H&E), periodic acid–Schiff (PAS), and Masson’s trichrome (all from Solarbio, Beijing, China) to evaluate renal morphology and fibrosis. To assess oxidative stress and protein expression, immunohistochemistry (IHC) was performed using antibodies against 4-hydroxynonenal (4-HNE; 1:200 dilution; Bioss, Beijing, China) and CYBB (1:500 dilution; Proteintech, Wuhan, China). After antigen retrieval, sections were blocked with 5% normal goat serum, incubated with primary antibodies overnight at 4 °C, and then treated with HRP-conjugated secondary antibodies. Signals were visualized using DAB substrate. In parallel, immunofluorescence staining was performed on kidney sections to detect podocin (1:1000 dilution; Proteintech) and GPX4 (1:200 dilution; Bioss) expression. For colocalization analysis, sections were co-incubated with anti-podocin and anti-GPX4 primary antibodies, followed by Alexa Fluor–conjugated secondary antibodies (Proteintech). Nuclei were counterstained with DAPI (Solarbio, Beijing, China). Images were captured using a confocal microscope (Leica, Wetzlar, Germany) and a digital slide scanner (3DHISTECH, Budapest, Hungary), and quantitative analysis was conducted using ImageJ software (NIH, Bethesda, MD, USA). Renal histological evaluation was performed in a blinded manner by an experienced pathologist. Renal histopathological changes were quantitated as described previously.¹³ Glomerular alterations were semi-quantitatively scored on a 0–4 scale: 0, no lesion; 1, mild (<25% of glomeruli involved); 2, moderate (25–50%); 3, extensive (50–75%); and 4, severe (>75%). For immunohistochemical analysis of 4-HNE, five randomly selected high-power fields (×400) per kidney section were analyzed using ImageJ. The average optical density was calculated as integrated optical density divided by the positively stained area. The mean AOD per sample was used for intergroup comparisons.

Biochemical Assays

Renal tissue homogenates and podocyte lysates were determined for iron, malondialdehyde (MDA), and glutathione (GSH) using commercial colorimetric assay kits (Jiancheng, Nanjing, China) according to the manufacturer’s instructions. Intracellular reactive oxygen species (ROS) were quantified with the DCFH-DA fluorescent probe (Beyotime, China).

Protein Extraction and Western Blotting

Tissue and cell lysates were prepared in RIPA lysis buffer (Solarbio, Beijing, China) freshly supplemented with PMSF (Solarbio). Protein concentration was determined by the BCA method (Yeasen, Shanghai, China) prior to gel loading. Equal amounts of protein (20μg) were resolved on SDS-PAGE gels and transferred onto PVDF membranes (Millipore, USA). Membranes were blocked in 5% skim milk and incubated overnight at 4 °C with primary antibodies against nephrin (1:500 dilution; Affinity, USA), podocin (1:500 dilution; Proteintech), CYBB (1:500 dilution; Proteintech), GPX4 (1:500 dilution; Bioss), SLC7A11 (1:1000 dilution; Affinity), and β-actin (1:20000 dilution; Proteintech). After incubation with HRP-conjugated secondary antibodies (Proteintech), signals were detected by enhanced chemiluminescence (Thermo Scientific, USA) and visualized using a chemiluminescence imaging system (Jena, Germany). All Western blot analyses were performed with three independent biological replicates. Quantification was performed with ImageJ software (NIH).

RNA Extraction, qPCR, and Transfection

Total RNA was extracted from podocytes using TRIzol reagent (Vazyme, Nanjing, China) and immediately reverse-transcribed into cDNA. cDNA was synthesized with a reverse transcription kit (Transgen, Beijing, China). Quantitative real-time PCR (qPCR) was performed with SYBR Green Master Mix (TransGen, Beijing, China) on a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The cycling program was 94 °C for 30s, followed by 40 cycles of 94 °C for 5s and 60 °C for 30s, with melting curve analysis. GAPDH was used as the internal control, and relative expression was calculated using the 2^−ΔΔCt method. Primer sequences for CYBB and GAPDH are provided in Supplementary Table 1.

For gene silencing, three siRNAs targeting murine CYBB were designed and synthesized by Genecreate. The sequences are shown in Supplementary Table 2, and the most effective siRNA was used in subsequent experiments. For overexpression, a plasmid encoding murine CYBB was used, with empty vector as control. The plasmid map is provided in Supplementary Figure 1. Transfections were carried out using Lipofectamine 3000 (Beyotime, Shanghai, China) according to the manufacturer’s instructions.

Bioinformatic Analysis

Human glomerular transcriptome data were obtained from the GEO database (GSE32591), comprising 32 biopsy samples from patients with LN and 14 samples from healthy living donors as controls. Raw data were preprocessed in R (version 4.3.1) using standard methods, including background correction, log₂ transformation, and quantile normalization. Principal component analysis (PCA) was performed to assess potential batch effects, and the ComBat function from the “sva” package was used for correction when necessary. Differentially expressed genes (DEGs) were identified using the limma package, with thresholds of absolute log₂ fold change (|log₂FC|) > 1 and adjusted P < 0.05.

Functional enrichment analysis of DEGs was conducted for Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways using the clusterProfiler R package. The results were visualized with the enrichplot package. Weighted gene co-expression network analysis (WGCNA) was performed to identify gene modules associated with the LN phenotype. Least absolute shrinkage and selection operator (LASSO) regression was subsequently conducted using the glmnet package to further refine candidate genes.

A curated list of ferroptosis driver genes was obtained from FerrDb V2, and the overlap with WGCNA- and LASSO-derived candidate genes was assessed using Venn diagram analysis. Differences in the expression of hub genes between LN and control groups were evaluated using the Wilcoxon rank-sum test in GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA), with P values < 0.05 considered statistically significant.

Statistical Analysis

Data are expressed as mean ± SEM with individual data points shown in graphs. Comparisons between groups were made using one-way ANOVA with Tukey’s post hoc test or unpaired Student’s t-test where appropriate. Nonparametric data were analyzed by the Wilcoxon rank-sum test. Correlations were assessed by Pearson’s or Spearman correlation coefficients depending on data distribution. A P value < 0.05 was considered statistically significant.

Result

MRL/Lpr Mice Exhibit Significant Nephritis Characteristics at 16 weeks

Serial assessments of renal function indicated progressive deterioration in MRL/lpr mice. Compared with 8-week-old MRL/lpr mice, 12-week-old mice showed mild increases in 24h-UP, SCr, and anti-dsDNA antibody levels, with the differences in 24h-UP and SCr not reaching statistical significance. By 16 weeks, 24h-UP, SCr, and anti-dsDNA antibody levels were significantly elevated, indicating pronounced renal impairment in these mice (Figure 1a–c).

Figure 1.

Progressive renal dysfunction in MRL/lpr mice and its attenuation by Fer-1. (a) 24h-UP, (b) serum anti-dsDNA antibody, and (c) SCr levels in MRL/lpr mice at 8, 12, and 16 weeks of age. Comparison of (d) 24h-UP, (e) anti-dsDNA antibody, and (f) SCr levels among Con, SLE, LN, and Fer-1 treated groups. Data are presented as mean ± SEM; ns: no significance; *P<0.05; **P<0.01; ***P<0.001.

For subsequent analyses, 8-week-old MRL/lpr mice were designated as the SLE group, representing an early lupus stage without overt renal damage. Sixteen-week-old MRL/lpr mice were defined as the LN group, and age-matched C57BL/6 mice served as the control (Con) group. Additionally, a cohort of 16-week-old MRL/lpr mice treated with the ferroptosis inhibitor Fer-1 was included as the Fer-1 group. Assessment of 24h-UP, SCr, and circulating anti-dsDNA antibody levels across groups revealed that SLE mice exhibited mild increases in 24h-UP and anti-dsDNA antibodies without significant changes in SCr. In contrast, LN mice showed pronounced elevations in all three parameters compared with controls, confirming the establishment of nephritis (Figure 1d–f).

Ferroptosis is Activated During the Progression of Nephritis and is Suppressed by Fer-1

Ferroptosis activity in the kidney was assessed by measuring key biochemical markers, including iron content, MDA, and GSH. Compared with control and SLE groups, LN mice showed significantly elevated renal iron content and MDA levels, along with markedly reduced GSH levels (Figure 2a–c). Western blot analysis further revealed a significant reduction in the ferroptosis-associated antioxidant proteins GPX4 and SLC7A11 in LN kidneys (Figure 2d and e). Treatment with Fer-1 partially restored GSH levels, reduced MDA accumulation, normalized renal iron content, and increased GPX4 and SLC7A11 expression.

Figure 2.

Evaluation of ferroptosis markers in kidneys of MRL/lpr mice. (a) Renal iron content, (b) GSH levels, and (c) MDA levels in Con, SLE, LN, and Fer-1 treated groups. (d) Representative Western blots showing the expression of GPX4, SLC7A11, podocin, and nephrin in kidney tissues. (e) Quantitative analysis of GPX4, SLC7A11, podocin, and nephrin protein expression levels. Data are presented as mean ± SEM; ns: no significance; *P<0.05; **P<0.01; ***P<0.001.

Fer-1 Treatment Ameliorates Renal Injury and Preserves Kidney Function in LN Mice

Histological analysis with H&E, PAS, and Masson’s trichrome staining showed marked glomerulosclerosis, tubular epithelial necrosis, and interstitial inflammation in LN mice, whereas renal architecture was largely preserved in control and SLE groups. These pathological alterations were significantly attenuated in the Fer-1 group, which exhibited improved glomerular and tubular morphology and reduced interstitial injury (Figure 3a). Semi-quantitative scoring showed significantly elevated mesangial proliferation, endocapillary hypercellularity, and crescent formation in LN mice compared with controls, while Fer-1 treatment markedly reduced these scores (Figure 3b). Fer-1 treatment also improved renal function and autoantibody levels, as indicated by reduced proteinuria, decreased SCr, and lower anti-dsDNA antibody titers compared with untreated LN mice (Figure 1d–f). Immunohistochemistry for 4-HNE, a marker of lipid peroxidation, was performed to assess oxidative lipid damage. LN kidneys exhibited intense 4-HNE accumulation in both glomerular and tubular compartments, and this accumulation was significantly reduced after Fer-1 treatment. Quantitative analysis of 4-HNE staining scores confirmed these observations, showing significantly elevated levels in the LN group compared with Con and SLE groups, while Fer-1 treatment markedly decreased the scores (Figure 3a and c).

Figure 3.

Histopathological and oxidative damage assessment in kidneys of MRL/lpr mice. (a) Representative images of kidney sections from Con, SLE, LN, and Fer-1 treated mice stained with H&E, PAS, and Masson’s trichrome, showing glomerular, tubular, and interstitial lesions. Immunohistochemical staining of 4-HNE was used to assess lipid peroxidation. (b) Semi-quantitative histopathological scoring of mesangial proliferation, endocapillary hypercellularity, and crescent formation based on H&E, PAS, and Masson’s staining. (c) Quantification of 4-HNE immunostaining was performed using average optical density (AOD) analysis. Scale bar: 100 μm. Data are presented as mean ± SEM. ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001.

Podocyte Injury Correlates with Ferroptosis Activity in LN

Since podocyte injury is a central driver of proteinuria in LN, podocyte integrity was evaluated in each experimental group. Western blot analysis showed that nephrin and podocin expression was maintained in the control and SLE groups but markedly decreased in LN mice, indicating severe podocyte injury (Figure 2d and e). Consistently, immunofluorescence staining revealed marked podocin loss in glomeruli of LN mice, whereas Fer-1 treatment significantly restored podocin expression and maintained podocyte structure (Supplementary Figure 2). To further investigate the relationship between ferroptosis and podocyte injury, co-localization analysis of GPX4 and podocin was conducted. LN mice showed reduced GPX4 expression in regions of podocyte loss, whereas Fer-1 treatment restored both signals, indicating that ferroptosis inhibition was associated with preserved podocyte integrity (Figure 4a and b).

Figure 4.

Co-localization analysis of GPX4 and podocin reveals ferroptosis-associated podocyte injury with quantitative assessment. (a) Immunofluorescence staining of GPX4 (red), podocin (green), and DAPI (blue) was performed on kidney sections from Con, SLE, LN, and LN + Fer-1 treated mice. LN mice exhibited marked reductions in GPX4 and podocin expression, with diminished co-localization, indicating ferroptosis activation and podocyte injury. Fer-1 treatment partially restored GPX4 and podocin levels. (b) Quantitative analysis of fluorescence intensity confirmed significant downregulation of both markers in LN mice, which was attenuated by Fer-1 administration. Scale bar: 100 μm. Data are presented as mean ± SEM; *P < 0.05; ***P < 0.001.

LN Serum Exposure Promotes Ferroptosis-Associated Podocyte Injury in vitro

To mimic the pathological environment of LN in vitro, differentiated MPC5 cells were treated with serum from LN mice. Compared with cells treated with control serum, podocytes exposed to 10% LN serum for 48 hours showed significant downregulation of nephrin and podocin, indicating structural and functional damage. Moreover, intracellular iron accumulation, ROS production, and lipid peroxidation (MDA) were markedly elevated, while antioxidant defenses, such as GSH content, GPX4, and SLC7A11 expression, were significantly reduced (Figure 5a–d). These alterations indicate activation of ferroptotic processes in podocytes exposed to LN serum.

Figure 5.

LN serum induces ferroptosis-associated podocyte injury in vitro, alleviated by Fer-1 treatment. (a) Western blot of GPX4, SLC7A11, nephrin and podocin in podocytes treated with control serum, LN serum, or LN serum plus Fer-1; (b) Quantification of protein expression shown in (a); (c) Intracellular iron, GSH, and MDA levels in podocytes under different treatments; (d) Flow cytometry analysis and quantification of ROS levels. Data are presented as mean ± SEM; *P<0.05; **P<0.01; ***P<0.001.

To investigate the role of ferroptosis, podocytes were pretreated with the ferroptosis inhibitor Fer-1 before LN serum exposure. Fer-1 markedly attenuated iron overload, ROS generation, and lipid peroxidation, while restoring GSH levels and upregulating GPX4 and SLC7A11 expression. Fer-1 pretreatment also preserved nephrin and podocin expression (Figure 5a–d), indicating alleviation of podocyte injury. These results demonstrate that serum from LN mice directly induces ferroptosis-associated injury in podocytes, and that pharmacological inhibition of ferroptosis provides cytoprotective effects against this injury.

Integrative Transcriptomic Analysis Identifies CYBB as a Ferroptosis-Associated Hub Gene in Human Lupus Nephritis Glomeruli

To characterize disease-associated transcriptional changes, we analyzed the glomerular expression dataset GSE32591, comprising 32 LN and 14 control glomerular samples. Differential expression analysis (limma, P < 0.05, |log₂ FC| > 1) identified 299 DEGs, consisting of 235 upregulated and 64 downregulated transcripts (Figure 6a). Heatmap visualization of the top 50 variable genes revealed distinct separation between LN and control groups, reflecting robust disease-associated transcriptional reprogramming (Figure 6b).

Figure 6.

Identification and functional enrichment analysis of differentially expressed genes in the glomeruli of patients with lupus nephritis. (a) The volcano plot illustrates the DEGs identified. (b) The heatmap displays the expression patterns of the top 50 differentially expressed genes. KEGG enrichment analysis of differentially (c) upregulated gene and differentially (d) downregulated gene; GO enrichment analysis of differentially (e) upregulated gene and differentially (f) downregulated gene.

Functional enrichment analysis revealed that upregulated DEGs were strongly associated with immune activation, enriched in pathways including NF-κB and Rap1 signaling, and involved in biological processes such as cytokine responses and leukocyte activation. In contrast, downregulated DEGs were predominantly enriched in metabolic pathways, including amino acid and arginine biosynthesis, as well as in extracellular stress response pathways (Figure 6c–f).

Next, network-based approaches were applied to identify LN-relevant genes. WGCNA identified two major gene modules, among which the blue module showed the strongest correlation with LN status (correlation coefficient r = 0.85). From this module, 81 hub genes were selected for subsequent analysis (Figure 7a–d and Supplementary Excel sheet 1). In parallel, LASSO regression reduced the DEG set to 13 candidates with nonzero coefficients, representing potential predictive markers for LN (Figure 7e, f and Supplementary Excel sheet 1).

Figure 7.

Identification of key LN-associated genes using WGCNA and LASSO regression analysis. (a) Analysis of scale-free topology for network construction; (b) Gene clustering dendrogram showing module assignment; (c) Module eigengene clustering dendrogram; (d) Heatmap showing correlations between gene modules and LN phenotype; (e) LASSO coefficient profiles of candidate genes; (f) Tenfold cross-validation for optimal λ selection in the LASSO model.

Integration with 207 ferroptosis driver genes curated in FerrDb V2 (Supplementary Excel sheet 1) identified CYBB as the sole overlapping candidate (Supplementary Figure 3). Further analysis confirmed that CYBB was significantly upregulated in LN glomeruli compared with controls (P < 0.001). These findings identify CYBB as a ferroptosis-associated hub gene with potential relevance to glomerular injury in LN.

Upregulation of CYBB Expression in LN Exacerbate Podocyte Ferroptosis

Based on bioinformatic predictions and in vivo ferroptosis phenotypes, we next investigated CYBB expression and its functional role in LN podocytes. Immunohistochemical analysis demonstrated that renal CYBB expression was significantly upregulated in early-stage (8-week) and established LN (16-week) MRL/lpr mice compared with age-matched controls (Figure 8a).

Figure 8.

CYBB promotes ferroptosis-associated podocyte injury in lupus nephritis. (a) Immunohistochemical staining showing increased CYBB expression in kidneys from MRL/lpr mice at Con, SLE and LN; Scale bar: 100 μm. (b and c) Intracellular ROS, (d) MDA, and (e) GSH levels in MPC5 podocytes exposed to LN serum under four conditions: untreated (LN), CYBB silencing (siRNA), CYBB overexpression (CYBB-OE), and CYBB-OE with Fer-1 pretreatment; (f and g) Western blots and quantification of GPX4, nephrin, and podocin expression under the indicated conditions. Data are presented as mean ± SEM; ns, not significant; *P<0.05; **P<0.01; ***P<0.001.

Since bioinformatic analysis identified CYBB as a key ferroptosis-associated gene in LN, and ferroptosis contributes to podocyte injury in LN mice, we further examined the functional role and underlying mechanism of CYBB in podocytes in vitro. Podocytes were transfected with CYBB-targeting siRNA or overexpression plasmids, and the efficiency of transfection was confirmed by qPCR and Western blotting (Supplementary Figure 4). We compared GSH and MDA levels among LN, NC-siRNA+LN (negative control siRNA), and OE-NC+LN (empty vector control) groups, and found no significant differences (Supplementary Figure 5). Therefore, we selected LN-treated cells as the control group for subsequent Western blot assays.

CYBB silencing significantly attenuated ferroptotic responses induced by LN serum, as evidenced by decreased ROS and MDA levels, accompanied by restoration of GSH and GPX4 expression. Nephrin and podocin expression was correspondingly preserved. Conversely, CYBB overexpression enhanced oxidative stress, ferroptotic phenotypes, and podocyte structural protein loss (Figure 8b–g).

An additional group of podocytes was pretreated with Fer-1 for 2 h following CYBB overexpression and then cultured in medium containing 10% LN serum for 48 h. As shown in Figure 8b–g, compared with the CYBB overexpression group exposed to LN serum, Fer-1 pretreatment restored GPX4, nephrin, and podocin expression, increased GSH content, and reduced ROS and MDA levels. These results indicate that the exacerbating effect of CYBB on LN podocyte injury is at least partially mediated through ferroptosis.

Discussion

In this study, we demonstrate that ferroptosis is a central mechanism of podocyte injury in LN and identify CYBB as a key upstream regulator linking oxidative stress to ferroptotic cell death. Using lupus-prone MRL/lpr mice and podocyte cultures, we observed that lupus nephritis was characterized by iron accumulation, lipid peroxidation, glutathione depletion, and downregulation of key antioxidant defenses such as GPX4 and SLC7A11. Pharmacological inhibition with Fer-1 preserved podocyte integrity, reduced proteinuria, and improved renal pathology. Furthermore, CYBB emerged as a hub gene consistently upregulated in human LN glomeruli, and its functional modulation directly influenced ferroptosis susceptibility in podocytes. These findings support the role of ferroptosis as a potential contributor to LN pathogenesis and suggest CYBB as a candidate regulator warranting further investigation.

Transcriptomic analyses identified CYBB as a ferroptosis associated gene consistently upregulated in glomeruli from lupus nephritis patients. Functional experiments confirmed that CYBB expression was increased in both lupus mouse kidneys and podocytes exposed to lupus serum. Mechanistically, CYBB overexpression enhanced reactive oxygen species production, intensified lipid peroxidation, and aggravated ferroptotic injury in podocytes, while CYBB knockdown or Fer-1 treatment reversed these effects. These findings position CYBB as a critical driver of ferroptosis mediated podocyte injury and suggest that CYBB-mediated oxidative stress constitutes an important mechanism of lupus nephritis progression. Our results are consistent with recent reports showing that ferroptosis plays a central role in the pathogenesis of lupus nephritis.^14,15 Previous studies have demonstrated increased lipid peroxidation products such as 4-hydroxynonenal in patient biopsies and lupus-prone mice, alongside reduced antioxidant activity in renal tissue.^16,17 Moreover, experimental inhibition of ferroptosis has been shown to reduce proteinuria and histopathological damage in lupus models.¹⁸ Our data extend these observations by focusing specifically on podocytes, the key determinants of proteinuria, and by identifying CYBB as a gene that links oxidative stress to ferroptotic injury in these cells.

The CYBB gene encodes NOX2, the catalytic subunit of the NADPH oxidase complex responsible for generating superoxide during the respiratory burst.¹⁹ Under physiological conditions, CYBB-derived ROS serve crucial roles in antimicrobial defense and immune regulation, particularly in neutrophils and macrophages.²⁰ Basal ROS production is also important for maintaining immune homeostasis, as highlighted by patients with chronic granulomatous disease who lack functional NOX2 and often develop dysregulated immunity.^21,22 However, excessive or sustained activation of CYBB shifts ROS production from protective to pathogenic, amplifying oxidative stress and promoting tissue injury.²³ In addition to its role in ROS generation, CYBB/NOX2 may also integrate redox and inflammatory signals by activating pathways such as NF-κB, thereby exacerbating local inflammation and enhancing ferroptotic susceptibility in podocytes.²⁴

LN is characterized by a pronounced state of oxidative stress within the kidney, which contributes to glomerular injury and disease progression.²⁵ NOX2 is expressed in glomerular cells including podocytes, and studies have observed that its level is elevated in lupus-affected renal tissues.^26–28 Infiltrating immune cells such as neutrophils and macrophages are also potent sources of ROS due to their NOX2 activity.^29,30 Persistent activation of these infiltrating immune cells establishes a sustained pro-oxidant milieu within the glomeruli, promoting lipid peroxidation and oxidative injury.³¹ Such immune cell derived oxidative stress can amplify lipid peroxidation in adjacent podocytes, thereby exacerbating ferroptotic injury. This cross-talk between infiltrating leukocytes and resident renal cells suggests a convergence of inflammatory and oxidative pathways that exacerbate podocyte dysfunction and renal injury in lupus nephritis.³¹ Recent insights into ferroptosis have further highlighted its role in shaping B cell fate in LN. Oxidative stress and ferroptotic signaling promote aberrant plasma cell differentiation and enhance autoantibody production.³² Beyond B cells themselves, their crosstalk with other immune cells under oxidative stress is crucial. For instance, neutrophil derived IL-6 can upregulate SLC7A11 in renal infiltrating B cells, conferring resistance to ferroptosis and sustaining pathogenic persistence.³³ These findings provide a mechanistic basis for considering ferroptosis modulation as a complement to B cell targeted therapies. While current agents like rituximab and belimumab primarily function through apoptotic or immune-mediated B cell depletion, they may also indirectly influence ferroptosis susceptibility—eg, via BAFF deprivation weakening cellular redox defenses.³⁴ Notably, recent clinical updates have proposed B cell–targeted agents such as rituximab as first-line options in pediatric LN under select conditions.³⁵ Thus, combining ferroptosis based strategies with B cell immunotherapy may help overcome therapeutic resistance and improve immune regulation in LN.³² Although our study focused on podocyte intrinsic mechanisms, future work should systematically dissect the contributions of ferroptosis across renal and immune compartments, and evaluate whether CYBB/NOX2 modulation can jointly attenuate both structural and immunologic drivers of lupus nephritis.

This study has several important limitations that should be acknowledged. First, all experiments were performed in lupus-prone mice and immortalized podocytes, which may not fully recapitulate the complexity of human lupus nephritis. Validation in patient biopsies and primary human podocytes is therefore warranted. Second, although we identified CYBB as a critical regulator of ferroptosis in podocytes, its broader roles in other renal compartments and infiltrating immune cells were not systematically addressed in this study. Given the physiological role of CYBB/NOX2 in host immune defense, particularly in neutrophil mediated pathogen clearance, therapeutic strategies targeting CYBB must carefully balance efficacy with potential immunosuppressive risks. Future work should investigate cell type specific contributions of ferroptosis to lupus nephritis, including the interplay between podocytes, tubular epithelial cells, and immune infiltrates. In addition, clinical studies are required to determine whether ferroptosis related markers such as CYBB can be developed into reliable biomarkers for disease activity or therapeutic response. Ultimately, a deeper understanding of ferroptosis in the renal and immune compartments may pave the way for precision therapies that selectively target pathogenic pathways while preserving protective functions.

In conclusion, our findings suggest that ferroptosis may be a key contributor to podocyte injury in lupus nephritis, with CYBB potentially acting as an upstream regulator linking oxidative stress to ferroptotic cell death. While these results advance our understanding of ferroptotic pathways in LN, they are primarily based on murine models and podocyte cultures. Further in vivo validation and mechanistic exploration are needed to establish whether targeting CYBB or ferroptosis could offer viable therapeutic benefits in human disease.

Funding Statement

This study was supported by the National Natural Science Foundation of China (82260898), the Jiangxi Provincial Key Research and Development Program (20243BBI91028), and the Jiangxi Provincial Natural Science Foundation, General Program (20242BAB25486).

Data Sharing Statement

The datasets used during this study can be available from the corresponding author on reasonable request.

Ethics Statement

This study used publicly available, fully de-identified human single-cell sequencing data, which is exempt from ethical review under Article 32 of the Ethical Review of Life Science and Medical Research Involving Human Subjects (China, 2023). All procedures involving animals were reviewed and approved by the Institutional Animal Care and Use Committee of The First Affiliated Hospital of Nanchang University (Approval No. CDYFY-IACUC-202509GR038). All animal care and experimental protocols complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the ARRIVE guidelines.

Author Contributions

Bihua Wang and Houhui Jiang contributed equally to this work and share first authorship. Bihua Wang and Houhui Jiang: Conceptualization, Methodology, Investigation, Formal analysis, Writing original draft. Mengxia Liu and Haili Chen: Investigation, Methodology, Data curation. Dengfeng Wu: Formal analysis, Visualization. Sheng Zhao and Yuxin Yan: Investigation, Formal analysis. Xia Fang and Sanyun Li: Writing review & editing. Rui Wu: Conceptualization, Methodology, Supervision, Funding acquisition, Writing review & editing. All authors took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare no competing interests.