The NRK-52E cell line, a proximal tubular epithelial cell line derived from rats, was provided by Prof. Sang-Ho Lee from Kyung Hee University College of Medicine. The cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 mg/ml penicillin, and 100 mg/ml streptomycin under a 5% CO₂ and 95% O₂ environment. The culture medium was replaced with fresh medium every 2 days. When the cells reached 70% confluence, they were subcultured after treatment with trypsin.

To confirm the reproducibility of the experimental results obtained from NRK-52E cells, primary cultures of rat proximal tubular epithelial cells were also established and used for experiments. Sprague Dawley rats (approximately 10 weeks old) were euthanized using carbon dioxide (CO₂) inhalation in accordance with the protocol approved by the Institutional Animal Care and Use Committee of Jeju National University (protocol no. 2024–0088; approval no. 2022–0014). The animals were placed in a dedicated sealed chamber, and CO₂ was introduced at a displacement rate of 30% of the chamber volume per minute until the concentration reached 60–70%. After the concentration was raised, the chamber was gently shaken up and down to maintain an appropriate CO₂ level. Following euthanasia, death was confirmed by checking for the cessation of heartbeat, after which the kidneys were removed. The kidneys were washed with Hanks' Balanced Salt Solution (HBSS), the renal capsule was removed, and the kidneys were dissected into four pieces, isolating only the cortical tissue. The HBSS was then removed, and the tissue was washed with cold phosphate-buffered saline (PBS). The washed tissue was transferred to a trypsin-treated flask, and 20 ml of a solution containing DNase I (1 mg/ml) and Collagenase Type I (2 mg/ml) in HBSS was added. The tissue was then sequentially passed through sieves of different pore sizes to remove distal tubule and glomerular cells. The proximal tubule cell suspension in HBSS was centrifuged at 228 × g (1,000 rpm) for 10 min, and the collected cells were cultured in rat tail collagen-coated plastic ware. After 1 week, the cells were harvested for intracellular l-[¹⁴C]cystine uptake assays, intracellular GSH content measurement, and lactate dehydrogenase assays.

NRK-52E cells (3×10⁵ cells/well) or an equivalent number of isolated RTECs were seeded into 24-well cell culture plates and cultured in DMEM medium containing 10% FBS for 2 days. After removing the culture medium, the cells were washed with 500 µl of extracellular fluid (ECF) buffer, pre-warmed to 37°C, and composed of 122 mM NaCl, 25 mM NaHCO₃, 3 mM KCl, 1.4 mM CaCl₂, 1.2 mM MgSO₄, 0.4 mM K₂HPO₄, 10 mM d-glucose, and 10 mM HEPES (pH 7.4). The cells were then stimulated for 4 h with dRib, BM, ML385, or brusatol. During the final 1 h of stimulation, 500 µl of 37°C ECF buffer containing 0.1 µCi l-[¹⁴C]cystine (1.7 µM) was added to each well to induce intracellular uptake of l-[¹⁴C]cystine. After incubation, the buffer was completely removed, and the cells were washed with cold, radioisotope-free ECF buffer to terminate the uptake process. The cells were lysed with 750 µl of 1% Triton X-100 in Dulbecco's phosphate-buffered saline, after which 500 µl of the lysate was mixed with 5 ml of scintillation cocktail for radioactivity measurement using the Wallac MicroBeta TriLux 1450 Liquid Scintillation and Luminescence Counter (PerkinElmer, San Juan, PR, USA). The remaining cell lysate was used to determine protein concentration using bicinchoninic acid protein assay kit (Pierce, Rockford, IL, USA). Intracellular l-[¹⁴C]cystine uptake was expressed as counts per minute (cpm) per microgram of protein, normalized to total protein concentration.

Intracellular GSH levels were measured using a GSH Assay Kit (Cayman, Ann Arbor, MI, USA) based on the enzymatic recycling method using glutathione reductase. Briefly, NRK-52E cells (3×10⁵ cells/well) or an equivalent number of isolated RTECs were seeded into 24-well cell culture plates and cultured in DMEM medium containing 10% FBS for 2 days. The cells were then stimulated for 6 h with dRib, BM, ML385, or brusatol. Following treatment, the cells were lysed by sonication, and the lysates were centrifuged to obtain the supernatant. The GSH concentration in the supernatant was measured according to the manufacturer's instructions. Intracellular GSH levels were normalized to total protein concentration and expressed as nmol/mg protein.

Intracellular iron levels were measured using the Iron Assay Kit (Sigma-Aldrich, St. Louis, MO, USA), a colorimetric assay. NRK-52E cells were seeded at a density of 1×10⁶ cells/well in a six-well culture plate and maintained in DMEM supplemented with 10% FBS. The cells were treated simultaneously with dRib, BM, ML385, or brusatol and incubated for 6 h. Subsequently, the cells were lysed using sonication, and the supernatant was collected after centrifugation. Total iron levels were measured according to the manufacturer's protocol. Intracellular iron levels were normalized to the total protein concentration of the cells and expressed as nmol/mg·protein.

Cytotoxicity was first assessed using a lactate dehydrogenase assay. NRK-52E cells were seeded at a density of 1×10⁵ cells/well in a 96-well cell culture plate, whereas isolated RTECs were plated at an equal number of cells per well in the same type of plate. The cells were maintained in DMEM supplemented with 10% FBS and treated simultaneously with dRib, BM, ML385, or brusatol for 6 h. Cytotoxicity was then measured using the Cytotoxicity Detection KitPLUS (Roche, Mannheim, Germany), and calculated using the following formula: (((experimental value - background control)-(low control - background control))/((high control - background control)-(low control - background control))) × 100. Finally, cell viability (% control) was expressed as: Cell viability (% control)}=100 - cytotoxicity (%).

We selected MDA as a marker to evaluate intracellular lipid peroxidation. NRK-52E cells were seeded in a six-well culture plate at a density of 1×10⁶ cells/well and cultured in DMEM medium containing 10% FBS for 48 h. The cells were then stimulated with dRib, BM, ML385, or brusatol simultaneously for 6 h. Subsequently, MDA levels were measured from the cell lysate using the EZ-Lipid Peroxidation Assay Kit (DoGenBio, Seoul, South Korea). The intracellular MDA levels were normalized to the total protein concentration of the cells and expressed as nmol/mg·protein.

Intracellular lipid peroxide levels were measured using flow cytometry with C11-BODIPY dye (Molecular Probes, Eugene, OR, USA). NRK-52E cells were seeded in a six-well culture plate at a density of 1×10⁶ cells/well and cultured in DMEM medium containing 10% FBS. The cells were then stimulated simultaneously with dRib, BM, ML385, or brusatol and incubated for 6 h. During the last 30 min, the cells were treated with 4 µM C11-BODIPY, followed by harvesting using 0.05% trypsin. The collected cells were centrifuged, resuspended in PBS, and analyzed for intracellular lipid ROS levels using a FACScan instrument (BD Bioscience, San Jose, CA, USA). A total of 10,000 cells per sample were analyzed, and the results were expressed as the ratio (fold) of the mean fluorescence intensity relative to the unstimulated control group.

All RT-qPCR experiments were performed in accordance with the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines (29). Total RNA was extracted from cells using TRIzol^® (Gibco Invitrogen, Grand Island, NY, USA) according to the manufacturer's instructions. Reverse transcription was performed using 2 µg of RNA with MMLV reverse transcriptase (MGmed Corporation, Seoul, Korea), oligo (dT)15 primer, dNTP (10 mM), and 40 units/µl RNase inhibitor (MGmed Corporation). RT-qPCR was conducted using the KAPA SYBR^® FAST qPCR Master Mix (KAPA Biosystems, MA, USA) and an iQTM 5 Multicolor Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Synthesized cDNA was amplified for 40 cycles at 95°C for 3 sec, 60°C for 30 sec, and 72°C for 30 sec, with an initial cycle at 95°C for 1 min, using specific primers. All primers were custom-synthesized by Bioneer Corporation (Daejeon, Korea), and their sequences are listed in Table I. Amplification specificity was confirmed by melt curve analysis, which showed a single distinct peak for each primer set. Standard curves generated from 5-fold serial dilutions of cDNA templates were used to determine amplification efficiency. All primer pairs showed amplification efficiencies ranging 92–105%, with R² values of >0.99. β-actin was used as a reference gene, and its stability was validated across experimental conditions using NormFinder analysis. All qPCR reactions were performed in triplicate, and results represent the mean of three independent biological replicates.

Cells were lysed using a lysis buffer, and equal amounts of protein from each group were separated using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. The membrane was incubated in Tris-buffered saline containing 0.1% Tween and 3% bovine serum albumin to block non-specific antibody binding. It was then incubated with primary antibodies specific to Nrf2 (#ab92946, Abcam), SLC7A11 (#175186, Abcam), ACSL4 (#MA5-42523, Invitrogen), CHAC1 (#217808, Abcam), and PTGS2 (#12282, Cell Signaling) at 4°C for 2 h. β-Actin (#A2228, Sigma) was used as a loading control. Afterward, the membrane was incubated with horseradish peroxidase-conjugated secondary anti-rabbit antibody (#PI-1000-1, Vector Laboratories) or anti-mouse antibody (#PI-2000-1, Vector Laboratories) for 2 h. Bands were visualized using Western Lightning™ Plus-Enhanced Chemiluminescence (PerkinElmer).

IP was performed using Dynabeads Protein G (#10003D, Thermo Fisher Scientific Inc., Waltham, MA, USA). NRK-52E cells were stimulated and cultured under the specified conditions, then lysed using ultrasonication. After centrifugation, the supernatant containing the cell lysate was collected. Dynabeads were conjugated with anti-Nrf2 antibody (#ab92946, Abcam) and anti-rabbit IgG antibody (#2729s, Cell Signaling) to prepare a bead-antibody mixture. This mixture was incubated with the cell lysate overnight at 4°C under rotation. The bead-antibody-antigen complex was washed four times with Tris-buffered saline with Tween 20. Lithium dodecyl sulfate sample buffer was then added to the beads, followed by incubation at 92°C for 5 min to elute the immunoprecipitant. The eluted immunoprecipitant was analyzed and quantified using western blot using anti-Keap1 and anti-Nrf2 antibodies.

NRK-52E cells were cultured on coverslips and divided into three groups: an untreated control group, a group treated with 50 mM dRib for 6 h, and a group treated with 0.2 µM BM and 50 mM dRib for 6 h. The cells were fixed with 4% paraformaldehyde at room temperature for 20 min and then permeabilized using a 0.1% Triton X-100 solution. Subsequently, blocking was performed with 5% bovine serum albumin at room temperature for 45 min. The cells were then incubated overnight at 4°C with primary antibodies: anti-Nrf2 antibody (1:100, #ab92946, Abcam) and anti-Keap1 antibody (1:100, #D6B12, Cell Signaling). After washing, the cells were incubated at room temperature for 1 h with the secondary antibody, donkey anti-rabbit AlexaFluor 488 (1:300, #ab150073, Abcam). Nuclear counterstaining was performed using 4′,6-diamidino-2-phenylindole at a 1:5,000 dilution. The stained cells were visualized using a STELLARIS 5 confocal microscope (Leica, Deerfield, IL, USA). To determine the ratio of Nrf2 fluorescence expressed in the nucleus and cytoplasm, images were analyzed using ImageJ software and displayed the values as a bar graph.

All data are presented as mean ± standard deviation. Statistical analyses were performed using SPSS software (version 14.0; SPSS Inc.), and P<0.05 was considered to indicate a statistically significant difference. Homogeneity of variance was assessed using Levene's test. When the assumption of equal variances was met (P>0.05), one-way ANOVA followed by Tukey's post hoc test was used. When the assumption was violated (P<0.05), Welch's ANOVA followed by Dunnett's T3 post hoc test was applied. For comparisons between two groups, an unpaired Student's t-test was used.

When NRK-52E cells were stimulated with dRib, there was a significant, dose-dependent decrease in intracellular cystine uptake, GSH content, and cell viability. However, co-treatment with BM significantly prevented these reductions (Fig. 1). Similarly, in primary cultured rat RTECs, dRib led to decreased cystine uptake, GSH levels and cell viability. These impairments were effectively restored by BM treatment (Fig. S1), indicating its protective effect against dRib-induced cellular damage. This shows that BM prevents cell death by increasing the intracellular transport of cystine in RTECs, thereby restoring GSH levels. Furthermore, intracellular iron levels, a key mediator of ferroptosis, were increased by dRib but were reduced to control levels by BM (Fig. 2C). Similarly, intracellular MDA levels and C11-BODIPY fluorescence, indicators of lipid peroxidation, were elevated by dRib but were nearly restored to control levels by BM (Figs. 2D, 2E and S2). Therefore, BM prevents dRib-induced ferroptosis in RTECs by counteracting the reduction in cystine uptake.

To determine whether the ferroptosis-inhibitory effect of BM depends on Nrf2 activation, we used the specific Nrf2 inhibitors, ML385 and brusatol. In isolated rat RTECs, ML385 and brusatol significantly attenuated the effects of BM in restoring cystine uptake, GSH content, and viability, which were reduced by dRib (Fig. S1). The same results were observed in NRK-52E cells (Fig. 2A, B and F). In addition, the ability of BM to suppress intracellular iron, MDA, and C11-BODIPY fluorescence levels, which were elevated by dRib, was also counteracted by ML385 and brusatol (Figs. 2C-E and S2). These findings indicate that the ferroptosis-inhibitory effect of BM is dependent on Nrf2 activation.

Nrf2 protein expression was increased in a dose-dependent manner by dRib and was further enhanced by the addition of BM (Fig. 3A and B). Although Nrf2 mRNA expression was also increased dose-dependently by dRib, the addition of BM significantly suppressed the dRib-induced increase (Fig. 4A). Meanwhile, SLC7A11 protein, which is known to be degraded by dRib (7), showed a dose-dependent decrease upon dRib treatment, but this decrease was significantly prevented by BM (Fig. 3A and C).

We further examined the expression of Nrf2 target genes, including SLC7A11, HO-1, NQO1, the catalytic subunit of glutamate-cysteine ligase (GCLC), and the modifier subunit of glutamate-cysteine ligase (GCLM). These genes were upregulated by dRib and showed a further increase upon BM treatment. The BM-induced increase in these Nrf2 target genes was significantly attenuated by ML385 and brusatol (Figs. 4B and 5).

Next, we investigated the effect of BM on the expression of ferroptosis biomarkers. The mRNA and protein levels of ACSL4, ChaC glutathione-specific gamma-glutamylcyclotransferase 1 (CHAC1), and prostaglandin-endoperoxide synthase 2 (PTGS2) were all significantly increased by dRib in a dose-dependent manner. However, treatment with BM markedly reduced the expression of these ferroptosis-related markers (Figs. 3 and 4C-E). These regulatory effects on SLC7A11, ACSL4, CHAC1, and PTGS2 gene expression were also consistently observed in primary cultured rat RTECs (Fig. S3). These findings further support the notion that BM prevents ferroptosis by activating Nrf2 pathway.

To elucidate the mechanism through which BM activates the Nrf2 pathway, we investigated the interaction between Nrf2 and Keap1 and their intracellular localization.

In the co-IP assay assessing their interaction, Keap1 was pulled down by Nrf2 in both the control and dRib-treated groups, indicating a physical association. However, when BM was added, the pull-down of Keap1 by Nrf2 was reduced (Fig. 6A and B). This indicates that although the physical interaction between Nrf2 and Keap1 remained intact under dRib treatment alone, the addition of BM weakened their association, implying that BM disrupts the Nrf2-Keap1 interaction.

To examine changes in the intracellular localization of Nrf2 and Keap1 upon dRib and BM treatment, we performed IF staining and observed the cells using confocal microscopy. In the control group, Nrf2 was evenly distributed between the cytoplasm and nucleus, whereas in the dRib-treated group, nuclear translocation of Nrf2 was increased. Notably, in the BM-treated group, nuclear translocation of Nrf2 was markedly enhanced, with minimal Nrf2 observed in the cytoplasm (Fig. 6C and D). However, Keap1 localization remained unchanged, as it was consistently retained in the cytoplasm regardless of dRib or BM treatment (Fig. S4). Results from co-IP and confocal microscopy indicate that BM disrupts the interaction between Nrf2-Keap1, thereby facilitating the nuclear translocation of Nrf2.