Curcumin induces ferroptosis in hepatocellular carcinoma by regulating the P62-KEAP1-NRF2-signaling pathway

Abstract

Context

Hepatocellular carcinoma (HCC) remains a major global health challenge, with limited treatment options owing to chemotherapy resistance and severe systemic toxicity. Curcumin demonstrates broad-spectrum antitumor activity against HCC and various other malignancies. Ferroptosis, a regulated form of cell death driven by iron overload, glutathione depletion, and lipid peroxidation, has recently gained attention as a potential therapeutic strategy in cancer treatment. Among the regulatory networks of ferroptosis, the P62-KEAP1-NRF2-signaling pathway plays a pivotal role.

Objective

To assess whether curcumin induces ferroptosis in HCC cells through modulation of the P62-KEAP1-NRF2-signaling pathway.

Materials and methods

A Hepa1-6 xenograft mouse model was developed to examine curcumin-mediated effects on tumor growth, ferroptosis markers, and the expression profiles of P62, KEAP1, and NRF2. Complementaryin vitro experiments were performed using HepG2 cells treated with a ferroptosis inhibitor (ferrostatin-1) or subjected to P62 overexpression, followed by assessment of cell viability and ferroptosis-associated parameters.

Results

Curcumin administration (100 mg/kg for 15 days) markedly suppressed tumor growth, reduced glutathione levels in tumor tissues, and enhanced the accumulation of reactive oxygen species, malondialdehyde, and Fe²⁺.In vitro, curcumin inhibited HepG2 cell proliferation, promoted ferroptotic cell death, downregulated P62 and NRF2 expression, and upregulated KEAP1 expression. These effects were reversed by ferrostatin-1 treatment. Moreover, P62 overexpression significantly attenuated the ability of curcumin to regulate the P62-KEAP1-NRF2-signaling pathway and induce ferroptosis.

Discussion and conclusion

Our findings demonstrate that curcumin suppresses the P62-KEAP1-NRF2-signaling pathway to induce ferroptosis, a key mechanism underlying its anti-tumor effects. This study not only provides a novel scientific basis for the application of curcumin but also reveals potential therapeutic targets for hepatocellular carcinoma.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-025-15307-1.

Introduction

Hepatocellular carcinoma (HCC) is one of the most prevalent malignant tumors worldwide and a leading cause of cancer-related mortality. Epidemiological data indicate that approximately 906,000 individuals are diagnosed with HCC each year, with this subtype representing the predominant form of primary liver cancer [1]; Vogel, Meyer, Sapisochin, Salem, & Saborowski [2]. These alarming figures highlight the urgent need to identify and develop effective therapeutic strategies to address this lethal disease.

Ferroptosis, a recently characterized form of regulated cell death distinct from apoptosis and necrosis, is defined by abnormal iron accumulation and impaired redox homeostasis, resulting in reduced antioxidant capacity. Morphologically, ferroptotic cells exhibit intact nuclei, nonaggregated chromatin, unruptured cytoplasmic membranes, and shrunken or absent mitochondria with ruptured outer membranes and increased inner membrane density (Gautheron, Gores, & Rodrigues [3]). A key upstream driver of ferroptosis is the cystine/glutamate antiporter system (system xc⁻), which provides cysteine for glutathione (GSH) biosynthesis (Hayano, Yang, Corn, Pagano, & Stockwell [4]; Lee et al., [5, 6]). Cysteine depletion results in diminished GSH synthesis, thereby initiating ferroptosis [7]. Glutathione peroxidase 4 (GPX4), a critical regulator of ferroptosis [8], converts GSH to oxidized glutathione GSH and reduces toxic lipid peroxides (phospholipid hydroperoxides) into nontoxic lipid alcohols. Inhibition of GPX4 promotes the excessive accumulation of reactive oxygen species (ROS) and lipid peroxidation products, ultimately triggering ferroptotic cell death (Friedmann Angeli, Krysko, & Conrad [9]). In addition, the acetylation-deficient p53 mutant (p53 3KR) suppresses xCT expression, resulting in GSH depletion and enhanced ferroptosis [10]. Other studies have reported that ceruloplasmin inhibits ferroptosis in HCC cells by regulating iron homeostasis, whereas inhibition of ceruloplasmin increases intracellular Fe²⁺ and ROS levels, sensitizing cells to erastin- and RSL3-induced ferroptosis [11]. Furthermore, GPX4-dependent ferroptotic hepatocyte death has been linked to tumor-suppressive immune activation, and combining ferroptosis inducers improves survival in murine liver tumor models [12]. Collectively, these findings underscore the therapeutic promise of ferroptosis as a strategy for managing primary liver tumors and metastases, particularly in HCC.

The P62-KEAP1-NRF2-signaling pathway has emerged as a crucial regulator of ferroptosis in HCC [13]. Nuclear factor erythroid 2-related factor 2 (NRF2) is a central transcription factor that governs antioxidant responses and cytoprotective mechanisms (Capelletti, Manceau, Puy, & Peoc’h [14]). Under basal conditions, Kelch-like ECH-associated protein 1 (KEAP1) promotes NRF2 ubiquitination and proteasomal degradation, thereby maintaining low NRF2 activity (Hirayama, Miki, & Nagasawa [15]; R. Yang et al., [16]). During oxidative stress, NRF2 dissociates from KEAP1, accumulates in the nucleus, and activates downstream antioxidant genes. This process is modulated by P62, an autophagy receptor that stabilizes NRF2. P62 knockdown sensitizes HCC cells to ferroptosis induced by erastin and sorafenib, whereas NRF2 inhibition enhances the anticancer activity of these agents in vitro and in xenograft models [13]. In H22 cells, ATP2B3 reduces erastin-induced ferroptosis by regulating the P62-KEAP1-NRF2-HO-1 pathway [17]. Similarly, arenobufagin induces autophagy-dependent ferroptosis in HepG2 cells through modulation of this pathway [18].

Curcumin, a naturally occurring polyphenolic compound derived from turmeric rhizomes, is widely used in traditional medicine and as a dietary component in Eastern cuisine. Preclinical studies have demonstrated its broad pharmacological activity, including potential anticancer properties (Prasad, Gupta, Tyagi, & Aggarwal [19]). In HCC, curcumin exerts antiproliferative and pro-oxidative effects, suppressing tumor progression by modulating transcription factors, inflammatory cytokines, growth factors, kinases, and metabolic enzymes. It inhibits tumor cell proliferation by inducing cell cycle arrest, apoptosis, and ferroptosis [20]. Curcumin-mediated inhibition of NF-κB significantly reduces HCC cell growth [21]. More recently, studies confirmed that curcumin can trigger ferroptosis in cancer cells, contributing to its tumor-suppressive activity [22]. However, the precise molecular link between curcumin-induced ferroptosis and its antitumor effects in HCC remains incompletely defined.

The present study demonstrates that curcumin regulates ferroptosis through the P62-KEAP1-NRF2-signaling pathway, altering the antioxidant capacity of HCC cells and suppressing their growth. These findings clarify the protective and mechanistic roles of curcumin in HCC and provide a scientific foundation for its application as a therapeutic candidate, thereby strengthening the rationale for integrating traditional Chinese medicine into modern strategies for HCC treatment.

Materials and methods

Chemicals and Reagents

Fetal bovine serum (cat. no. D1220D), DMEM medium (cat. no. MA0212), Cell Counting Kit-8 (cat. no. MA0218), MTT reagent (cat. no. 298-93-1), Annexin V-FITC/7-AAD cell apoptosis-detection kit (cat. no. MA0428), EDU-555 cell proliferation-detection kit (cat. no. CX003), PBS (cat. no. MA0015), and curcumin (cat. no. MB2147) were all purchased from Meilun Biotechnology Co., Ltd. (Dalian, China). The malondialdehyde (MDA) detection kit (cat. no. S0131S) and Caspase inhibitor Z-VAD-FMK (cat. no. C1202) were purchased from Beyotime Biotechnology Co., Ltd. (Nantong, China). Ferrostatin-1 (cat. no. 347174-05-4) was purchased from MedChemExpress (MCE; New Jersey, USA). The reduced glutathione (GSH) detection kit (cat. no. A006-2-1) was purchased from Nanjing Jiancheng Biotechnology (Nanjing, China). The ferrous ion (Fe²⁺) detection kit (cat. no. BC5410) and dimethyl sulfoxide (DMSO; cat. no. D8370) were both purchased from Solarbio Technology Co., Ltd. (Beijing, China). The DAPI staining solution kit (cat. no. EE0011) and Sorafenib (cat. no. SJ-MX0071A) were purchased from SparkJade Biotechnology Co., Ltd. (Shandong, China). For western blotting, antibodies against P62 (species: rabbit, dilution: 1:5000, cat. no. ab109012) were purchased from Abcam (Cambridge, UK). Antibodies against KEAP1 (species: rabbit, dilution: 1:1000, cat. no. 8047), GPX4 (species: rabbit, dilution: 1:1000, cat. no. 59735), xCT/SLC7A11 (species: rabbit, dilution: 1:1000, cat. no. 12691), Histone H3 (species: rabbit, dilution: 1:1000, cat. no. 4499) and P53 (species: mouse, dilution: 1:1000, cat. no. 2524) were purchased from CST (USA). The antibody against β-actin (species: mouse, dilution: 1:4000, cat. no. GB15001) was purchased from Servicebio (Wuhan, China). The antibody against HO-1 (species: rabbit, dilution: 1:10000, cat. no. 10701-1-AP) and NQO1 (species: rabbit, dilution: 1:4000, cat. no. 11451-1-AP)were purchased from Proteintech (Wuhan, China). The antibody against xCT (species: rabbit, dilution: 1:1000, cat. no. DF12509) was purchased from Affinity Biosciences (USA). The antibody against NRF2 (species: rabbit, dilution: 1:1000, cat. no. A0674) was purchased from Abclonal (USA). Goat anti-mouse IgG (cat. no. ZB-2305) and goat anti-rabbit IgG (cat. no. ZB-2301) were provided by Zhongshan Golden Bridge Biotechnology Co., Ltd.

Animal model construction

C57BL/6JNifdc mice [specific pathogen-free (SPF) grade; Female; Weight, 17–19 g; Age, 6–7 weeks; Beijing Vital River Laboratory Animal Technology Co., Ltd.). The mice were housed under a SPF grade condition of Shandong University of Traditional Chinese Medicine (Jinan, China). The animal experimental protocol was conducted in accordance with the “National Institutes of Health Guidelines for the Care and Use of Laboratory Animals” and approved by the Institutional Animal Care and Research Advisory Committee of the Shandong University of Traditional Chinese Medicine (Approval No. SDUTCM20241113002; Jinan, China). A total of 30 mice were subjected to a 12-h light/dark cycle at 25.0 ± 2.0 °C under 55.0 ± 5.0% relative humidity and could freely access food and water. After 1 week of adaptation, 1 × 10⁶ Hepa1-6 cells in 0.2 mL of physiological saline were injected from the right axilla of the mice to form subcutaneous xenograft tumors. When Hepa1-6 cell tumors were observed with the naked eye, the mice were assigned into the following three groups of 10 each: (i) Model group (injected with cancer cells, intraperitoneal injection of an equal volume of physiological saline as a negative control); (ii) Curcumin group (injection of cancer cells, daily intraperitoneal injection of 100 mg/kg curcumin); (iii) Chemotherapy group (injection of cancer cells, daily intraperitoneal injection of 60 mg/kg Sorafenib; positive control). The regimens of Sorafenib and curcumin were based on their clinical applications and past studies, respectively [23]. The weight and tumor volume of each mouse were measured every 3 days over a period of 15 days. The tumor volume was calculated by using the following equation: tumor volume = length x width x width/2. When a tumor diameter of 1500 mm³ was recorded or severe disease symptoms (such as breathing difficulties or paralysis) were observed, the animals were euthanized. On the 15th day, 30 mice (n = 10/group) were euthanized via intraperitoneal injection of excessive pentobarbital sodium (200 mg/kg). The duration between injection and final tumor growth measurement was 20 days. The tumor tissues were removed, weighed, and stored immediately at −80 °C until further analysis.

Cell culture

The human HCC cell line (HepG2, CVCL_0027, cat. no.CL-0103) and the mouse-derived HCC cell line (Hepa1-6, CVCL_0327, cat. no.HTX1770) were both purchased from Meilun Biotechnology Co., Ltd. The cells were cultured in DMEM medium supplemented with 10% fetal bovine serum, 100 UI/mL penicillin, and 100 ug/mL streptomycin (cat. no. MA0110, Meilun Biotechnology Co., Ltd.). These cells were cultured in a humidified incubator at 37 °C under 5% carbon dioxide atmosphere (Thermo Scientific, USA).

H&E staining

The tumor tissues were embedded in paraffin, sectioned, and stained as per the standard operating procedures. After dehydration and transparency treatment, the sections were mounted. Whole-slide scanning was performed using a digital slide scanning system for image acquisition (SQS-600P, China).

Cell viability assay

Cell viability was assessed by the CCK-8 and MTT assays. For the CCK-8 assay, the experimental cells were seeded in a 96-well plate at a density of 5,000 cells/well. After treating the HepG2 cells with curcumin (5, 10, 20, 30, 40, 50, 60, 70, 80, 90 µM) for 12 h, 10 µL of the CCK-8 reagent was added to each well. The plates were incubated at 37 °C for 1 h, and the absorbance (OD) of each well was measured at 450 nm as per the manufacturer’s instructions. Similarly, for the MTT assay, the cells were seeded in a 96-well plate at a density of 5,000 cells/well. Hepatoma cells were treated with curcumin (0, 5, 20, 50 µM) for 12 h, followed by the addition of MTT reagent (10% or 20 µL/well). After 4 h, the culture was terminated, and the supernatant in each well was carefully aspirated. DMSO (150 µL per well) was added to each well, and the plates were shaken on an orbital shaker (Qilinbeier, China) for approximately 10 min to dissolve the formed crystals. The absorbance was then measured at a wavelength of 490 nm by using a microplate reader (BioTek, USA).

Migration assay

HepG2 cells were added to the upper chamber for culture and stained with H&E. HepG2 cells that migrated to their lower face were enumerated in 3 views per membrane under a microscope (ZEISS, Germany).

Flow cytometry

A total of 1 × 10⁵ cells/well in a 6-well plate, pretreated with ferrostatin-1 (Fer-1) and Z-VAD-FMK, and then treated with curcumin for 12 h. The cells were washed twice and incubated with a fluorescent dye. Subsequently, flow cytometry (Beckman, USA) was performed to scan and record the fluorescence signals emitted by FITC and 7-AAD.

Cell proliferation assay

The cells were seeded into a 6-well plate at a density of 1 × 10⁵ cells/well, treated with curcumin for 12 h, and then incubated with the EDU working solution, followed by the addition of a fixative, a permeabilization solution, and a reaction solution. Finally, 1X Hoechst 33,342 solution was added for nuclear staining. After washing, fluorescence detection was performed immediately under light-protected conditions.

Transmission electron microscopy

The processed cells were pre-embedded and fixed. After infiltration and embedding, the mixture was polymerized into blocks, and ultrathin sections were sliced at 60–80 nm. Negative staining was performed to enhance the contrast. Images were acquired by using a transmission electron microscope (Hitachi, Japan).

Cell transfection

During transfection, the cells were seeded into a 6-well plate and cultured to 80–90% confluency before performing plasmid transfection with the Homo sapiens sequestosome 1 (SQSTM1) (cat. no. HH20240517GX-PC02, Hanheng Biotechnology, Shanghai, China). When the cells seeded in the 6-well plates reached 30–50% confluency, siRNA transfection was performed using LipoFlyer 3.0 (cat. no. HB-LF3-1000, Hanheng Biotechnology) transfection reagent.

Determination of intracellular reactive oxygen species

ROS production in the cells was detected using DCFH-DA (cat. no. S0033S, Beyotime Biotechnology Co., Ltd.). After oxidation in the cells, the fluorescence of the labeled probes increased significantly. HepG2 cells were cultured with curcumin for 12 h, incubated with 10 µM DCFH-DA for 30 min, washed thrice with PBS, and immediately observed and imaged under a fluorescence microscope (ZEISS).

Measurement of MDA, total Iron, and GSH content

Commercial assay kits were used to measure the levels of MDA, total iron, and GSH in the cells and animal serum. The MDA levels were expressed in nmol/mg protein and related to the cellular protein concentration. The GSH levels were expressed as µmol/gprot protein and correlated with the cellular protein concentration.

BODIPY-C11 581/591 staining

Lipid peroxides were examined by BODIPY-C11 581/591 staining (BODIPY) (cat. no. 217075-36-0, SparkJade Biotechnology Co., Ltd.). After processing, the cells were incubated with BODIPY in the dark at 37℃ and a final concentration of 5 µM for 30 min. Subsequently, the washed cells were observed and photographed under a fluorescence microscope (ZEISS).

Cellular Immunofluorescence

The cells were washed twice with PBS and fixed with 4% paraformaldehyde (cat. no. EE0001, SparkJade Biotechnology Co., Ltd.) for 20 min, followed by permeabilization with Triton X-100 solution (cat. no. MB2486, Meilunbio Biotechnology Co., Ltd.) for 30 min and blocking with 5% BSA (cat. no. ED0017-B, SparkJade Biotechnology Co., Ltd.) solution at 37 °C for 30 min. Finally, the cells were treated with the primary antibody overnight at 4 °C, followed by incubation with FITC-labeled secondary antibody (cat. no. P0186, Beyotime Biotechnology Co., Ltd.) for 1 h at room temperature away from light. The cells were washed thrice with PBS, followed by counterstaining the nuclei with the DAPI solution for 30 min. After another wash with PBS, an anti-fade mounting medium was added, and the fluorescence intensity was observed under a fluorescence microscope (ZEISS).

Western blotting

The cells were collected, and HepG2 cells or tumor tissues were processed and placed in centrifuge tubes. A cell lysis buffer containing a mixture of 10 µg/mL phosphatase inhibitors (cat. no. GRF102, Elegzyme, China) and 10 µg/mL protease inhibitors (cat. no. ST506, Beyotime Biotechnology Co., Ltd.) was used to lyse the cell membranes and release proteins. The mixture was centrifuged to remove cell debris and collect the supernatant. Next, measure the protein concentration in the supernatant using a BCA protein assay kit (cat. no. ZJ102, Elegzyme) to ensure the appropriate amount of protein sample is loaded. Load the protein (10–30 µg) and marker (cat. no. 26616, Thermo Scientific) into the sample wells containing SDS-PAGE gel (cat. no. PG212, Elegzyme) electrophoresis buffer. SDS-PAGE gel electrophoresis was performed to separate the proteins, and the separated proteins were transferred onto a PVDF membrane (cat. no. IPVH00010, Merck KGaA, Germany) using a transfer buffer for blotting. The transferred membrane was incubated overnight at 60 rpm with anti-P62, anti-KEAP1, anti-NRF2, anti-GPX4, anti-xCT, anti-HO-1, anti-NQO1, anti-Histone H3, and anti-P53 antibodies. After washing five times for 5 min each time, the membrane was incubated with a suitable horseradish peroxidase (HRP)-conjugated secondary antibody at room temperature for 1 h. An ECL detection kit (cat. no. ED0016-A, SparkJade Biotechnology Co., Ltd.) was used for color development and exposure.

Statistical analysis

All experiments were performed independently at least three times, with the data presented as the mean ± standard error of the mean (SEM). The means were compared using either one-way analysis of variance (ANOVA; for multiple groups) or t-tests (for two groups) via the Prism 8 software program. The differences were considered to be statistically significant at p < 0.05. GraphPad Prism 8.0 (GraphPad Software Inc., La Jolla, CA) was used to visualize the data.

Results

Curcumin suppresses HCC growth

To evaluate the therapeutic potential of curcumin in HCC, we established ectopic xenografts derived from Hepa1-6 cells in mice. After 15 days of treatment, curcumin significantly suppressed tumor growth, demonstrating efficacy that was second only to the first-line drug sorafenib. This was reflected in significant reductions in tumor volume (45.9% vs. model group, p < 0.01) and tumor weight (45.9% vs. model group, p < 0.01) (Figs. 1A–C). Histopathological analysis revealed extensive tumor necrosis and decreased cellular density in both the curcumin- and sorafenib-treated groups (Fig. 1D), consistent with the observed suppression of tumor growth. In addition, curcumin treatment induced hallmark features of ferroptosis compared with the model group. These included a 1.6-fold reduction in GSH levels (p < 0.05), a 1.3-fold increase in MDA accumulation (p < 0.01), and a 2.2-fold elevation in Fe²⁺ content (p < 0.05) (Figs. 1E–G). Western blotting further demonstrated that curcumin and sorafenib decreased the expression of ferroptosis-associated proteins xCT and GPX4, as well as the pathway regulators P62 and NRF2. In contrast, both treatments increased KEAP1 and P53 expression (Fig. 1H–I), indicating simultaneous suppression of KEAP1/NRF2-mediated antioxidant defenses and activation of P53-dependent ferroptotic signaling.

Fig. 1.

Curcumin can inhibit tumor growth in vivo. (A) Images of tumor xenograft mice and tumor resection mice from each group. The effects of curcumin on (B) body weight and (C) tumor volume were measured every 3 days. (D) Hematoxylin and eosin staining of tumor tissues (magnification, ×100; scale bar, 200 μm). The levels of GSH (E), Fe²⁺ (F), and MDA (G) in mouse serum were detected using western blotting (H), and the expressions of pathway proteins P62, KEAP1, NRF2, and ferroptosis-related proteins GPX4, xCT, and P53 in mouse tumor tissues were analyzed (I) and quantified. Data are presented as the mean ± standard deviation. n = 3, *P < 0.05, **P < 0.01, and ***P < 0.001

Curcumin suppresses HepG2 HCC proliferation and invasion

To assess the antitumor activity of curcumin in HCC, we first examined its effects on HepG2 cell proliferation. HepG2 cells were treated with curcumin (0, 5, 20, and 50 µM) for 12 h, and MTT assays revealed a dose-dependent reduction in cell viability (Fig. 2A). The half-maximal inhibitory concentration was calculated as 50.75 µM (Fig. 2B), which was subsequently selected for further experiments. EDU staining confirmed a pronounced decrease in DNA synthesis activity in curcumin-treated cells compared with controls (Fig. 2C–D). In addition, Transwell assays demonstrated that curcumin significantly impaired the invasive capacity of HepG2 cells (Fig. 2E–F). Flow cytometry analysis further revealed that curcumin treatment substantially increased nonapoptotic cell death (D. Chen, Eyupoglu, & Savaskan [24]), (Figs. 2G–H). Importantly, this effect was not reversed by the pan-caspase inhibitor Z-VAD-FMK (curcumin + Z-VAD-FMK vs. curcumin alone: p >0.05), but was significantly rescued by the ferroptosis inhibitor (Fer-1) (curcumin + Fer-1 vs. curcumin alone: p < 0.01). Taken together, these results indicate that curcumin exerts its antiproliferative and anti-invasive effects on HepG2 cells predominantly through the induction of ferroptosis.

Fig. 2.

Curcumin can inhibit tumor activity. (A) MTT assay was performed to detect the cell proliferation ability; (B) CCK-8 method was used to screen concentrations; (C–D) EDU assay was performed to detect the proliferation ability of HepG2 cells (magnification ×10; scale bar, 100 μm); (E–F) Transwell assay was used to detect the cell invasion ability (magnification ×20; scale bar, 50 μm); (G–H) Flow cytometry was performed to determine the type of cell death. Data are presented as the mean ± SD, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Curcumin induces ferroptosis in HepG2 cells

To confirm that curcumin induces ferroptosis in HepG2 cells, we first examined cellular morphology by transmission electron microscopy (Fig. 3A). Compared with the control group, curcumin-treated cells displayed irregular nuclei and mitochondria that were reduced in size, mostly rounded, with fewer and thickened cristae, increased membrane electron density, and intact outer membranes. These ultrastructural alterations are consistent with the morphological hallmarks of ferroptosis. We next evaluated intracellular Fe²⁺ accumulation, lipid peroxidation (ROS and MDA levels), and antioxidant capacity (GSH) (Figs. 3B–F). Curcumin treatment significantly increased Fe²⁺ levels, elevated ROS and MDA levels, and markedly reduced GSH levels compared with the control group. To further confirm lipid peroxidation, we performed BODIPY 581/591-C11 staining. As shown in Figs. 3G, curcumin treatment enhanced green fluorescence intensity, indicating elevated lipid ROS accumulation relative to controls. Western blotting and immunofluorescence analyses demonstrated that curcumin reduced the expression of the ferroptosis-associated proteins GPX4 and xCT, while increasing P53 expression (Figs. 3H–J). In addition, curcumin decreased the levels of pathway proteins P62 and NRF2 but upregulated KEAP1. Together, these results provide strong evidence that curcumin induces ferroptosis in HepG2 cells by modulating both ferroptosis markers and the P62-KEAP1-NRF2-signaling pathway.

Fig. 3.

Curcumin induced ferroptosis in HepG2 cells. A The morphology and structure of HepG2 cells were examined by transmission electron microscopy. B-C The intracellular ROS levels were measured using the ROS assay kit. D The cellular GSH levels were measured by using a GSH detection kit. E Changes in the intracellular MDA content were assessed by using an MDA-detection kit. F The intracellular Fe²⁺ levels were determined by using an Fe²⁺ detection kit. G Detection of intracellular lipid peroxidation using BODIPY staining. H Changes in the ferroptosis-related protein P53 were detected by cellular immunofluorescence (magnification, ×20; scale bar, 50 μm). I-J The expressions of ferroptosis-related proteins xCT, GPX4, and pathway proteins P62, KEAP1, and NRF2 in HepG2 cells were analyzed. Data are presented as the mean ± SD, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fer-1 suppresses Curcumin-induced ferroptosis in HepG2 cells

To further confirm that curcumin induces ferroptosis in HepG2 cells, we performed early intervention with Fer-1. As shown in Fig. 4A, intracellular ROS levels were lower in the curcumin + Fer-1 group than in the curcumin-only group. Consistently, intracellular MDA and Fe²⁺ levels decreased, while GSH levels markedly increased in the curcumin + Fer-1 group compared with the curcumin group (Figs. 4B–D). These findings suggest that curcumin promotes the generation of lipid peroxides, as demonstrated by the reduced green fluorescence intensity of BODIPY staining following Fer-1 treatment (Figs. 4E). Moreover, immunofluorescence and Western blotting analyses (Figs. 4F–H) showed that the combination of curcumin and Fer-1 suppressed the expression of the ferroptosis-related proteins GPX4, xCT, and P53 compared with curcumin alone, further supporting the role of curcumin in inducing ferroptosis in HepG2 cells. In addition, treatment with curcumin alone reduced the expression of the pathway proteins P62 and NRF2, while increasing KEAP1 expression. By contrast, Fer-1 intervention reversed these changes, elevating P62 and NRF2 expression and reducing KEAP1 expression. The P62-KEAP1-NRF2-signaling pathway is a critical regulator of ferroptosis [13]. Based on these observations, we propose that curcumin induces ferroptosis in HepG2 cells by modulating the P62-KEAP1-NRF2-signaling pathway, thereby contributing to its antitumor effects.

Fig. 4.

Ferrostatin-1 suppresses curcumin-induced ferroptosis in HepG2 cells. (A) the detection of the intracellular ROS levels; (B) Changes in the intracellular MDA content detected by using an MDA detection kit; (C) The measurement of the intracellular Fe²⁺ levels by using an Fe²⁺ detection kit; (D) The measurement of the intracellular GSH levels by using a GSH detection kit; (E) Detection of intracellular lipid peroxidation using BODIPY staining; (F) The detection of P53 expression in cells using immunofluorescence (magnification, ×20; scale bar, 50 μm); (G) Expressions of pathway proteins P62, KEAP1, NRF2, and key ferroptosis proteins xCT and GPX4 in HepG2 cells. (H) Immunoblotting and quantitative analysis. Data are presented as the mean ± SD, n = 3, *P < 0.05, **P < 0.01, *** P < 0.001

Curcumin induces ferroptosis by regulating the P62-KEAP1-NRF2-Signaling pathway

To further verify that curcumin induces ferroptosis in HepG2 cells through regulation of the P62-KEAP1-NRF2-signaling pathway, we constructed a P62 overexpression plasmid (OE-P62) (Fig. 5A) and used it to increase P62 protein expression (Fig. 5B). Compared with the curcumin-only group, OE-P62 attenuated the increase in intracellular ROS levels (Fig. 5C). In addition, OE-P62 reduced lipid peroxide accumulation, as evidenced by the enhanced green fluorescence intensity of BODIPY staining following curcumin treatment (Fig. 5D). Immunofluorescence analysis demonstrated that curcumin significantly reduced intracellular P62 (Fig. 5E) and NRF2 (Fig. 5G) levels while increasing KEAP1 expression(Fig. 5F), compared with the control group. Importantly, OE-P62 reversed these curcumin-induced alterations in pathway proteins. Furthermore, Western blotting analysis revealed that OE-P62 counteracted the curcumin-induced NRF2 and P62 downregulation and KEAP1 upregulation. OE-P62 also suppressed NRF2 nuclear accumulation and decreased the expression of downstream targets, including HO-1, NQO1, GPX4, and xCT. However, there was no significant change in the expression of LC3B, a key autophagy marker. (Fig. 5H-I). Taken together, these findings suggest that curcumin promotes ferroptosis in HepG2 cells by modulating the P62-KEAP1-NRF2-signaling pathway (Fig. 6).

Fig. 5.

Curcumin induces ferroptosis by modulating the P62/KEAP1/NRF2-signaling pathway. (A-B) Western blotting and quantitative analysis of P62 overexpression. (C) A detection kit used for measuring the intracellular ROS levels. (D) Detection of intracellular lipid peroxidation using BODIPY staining. Changes in the levels of intracellular pathway proteins P62 (E), KEAP1 (F), and NRF2 (G) were detected by cellular immunofluorescence. (H) Protein expressions of pathway proteins P62, KEAP1, NRF2, and key ferroptosis proteins xCT, GPX4 and key autophagy protein LC3B were analyzed using immunoblotting and (I) quantitatively assessed. The data is expressed as the mean ± standard deviation (n = 3) *P < 0.05 and ** P < 0.01. *** P < 0.001

Fig. 6.

The graphical representation of this study: An in-depth understanding of the mechanism through which curcumin promotes ferroptosis in hepatocytes by regulating the P62-KEAP1-NRF2-signaling pathway

Discussion

Despite continuous advances in treatment strategies for HCC in recent years, chemotherapy resistance and systemic toxicity remain major challenges that limit patient outcomes (L. Q. Cao, Xie, Fleishman, Liu, & Chen [25]; Ikeda et al., [26]). This underscores the urgent need for novel therapeutic approaches that are both effective and safe. Natural products offer promising avenues for anticancer drug discovery due to their structural diversity, multitarget mechanisms of action, and relatively low toxicity (Quintero-Rincón, Caballero-Gallardo, & Olivero-Verbel [27]). Among them, curcumin, a natural polyphenolic compound with well-documented biological activities, has attracted considerable attention. Recognized as a “multifunctional molecule,” curcumin exhibits a broad range of pharmacological effects, including anticancer, anti-inflammatory, and antioxidant properties, making it a sustained focus of therapeutic research (Joshi, Verma, Kumar Semwal, Dwivedi, & Sharma [28]). Ferroptosis is a regulated, nonapoptotic form of cell death defined by the iron-dependent accumulation of lethal lipid peroxides within cellular membranes. As an intrinsic tumor-suppressive process, it plays a pivotal role in cancer biology. Importantly, mesenchymal and dedifferentiated tumor cells, often resistant to apoptosis and conventional therapies, display heightened sensitivity to ferroptosis, highlighting its therapeutic promise, particularly for refractory malignancies [29, 30]. In this study, we demonstrated that inhibiting the P62-KEAP1-NRF2-signaling pathway induces ferroptosis, which is an important mechanism by which curcumin exerts anti-tumor effects.

Curcumin and its derivatives can promote iron accumulation by increasing intracellular Fe²⁺ levels in diverse tumor cell types [31]; Zhang, Yu, Peng, & Peng [32]. Moreover, curcumin-induced overactivation of HO-1 markedly elevates trivalent iron, MDA, and ROS, resulting in excessive iron release and subsequent disruption of iron homeostasis [33]; Liu, Ma, & Lai [34]. The dual ability of curcumin to chelate iron and regulate redox balance further underscores its role in iron metabolism [35, 36]. Consistent with these findings, our in vitro and in vivo results revealed that curcumin treatment leads to significant Fe²⁺ accumulation in HepG2 cells, elevated lipid peroxidation, depletion of GSH, altered expression of ferroptosis-related proteins (GPX4, xCT, and P53), and distinct mitochondrial morphological changes. These effects were reversed by Fer-1, confirming that the form of cell death induced by curcumin is ferroptosis. Furthermore, animal experiments corroborated these results, reinforcing the conclusion that curcumin induces ferroptosis as a mechanism underlying its antitumor activity.

P62 is a multifunctional adaptor protein whose role depends heavily on the cellular context and its interacting partners. Beyond serving as a key autophagy receptor, P62 acts as a central hub for stress signaling. Through its KEAP1-interacting region (KIR domain), P62 binds KEAP1 and specifically activates the NRF2 antioxidant pathway [37]. Previous studies have demonstrated that the P62-KEAP1-NRF2 pathway can be synergistically activated during ferroptosis, thereby enabling cancer cells to resist ferroptotic stress [13]. For example, silencing CISD2 enhances autophagic flux, suppresses the P62-KEAP1-NRF2 pathway, and accelerates ferroptotic cell death [38]. In the present study, electron microscopy revealed marked ultrastructural changes consistent with ferroptosis in curcumin-treated cells, although no autophagosomes were observed. Importantly, the Fer-1 markedly reversed the ferroptotic alterations induced by curcumin, suggesting that P62 primarily mediates cell death through regulation of the ferroptotic pathway. P62 protects against ferroptosis by stabilizing NRF2, preventing its degradation by KEAP1, and promoting its nuclear translocation to activate downstream antioxidant response element-dependent genes. Many of these genes regulate redox homeostasis and iron metabolism, serving as critical suppressors of ferroptotic initiation. Consequently, the transcriptional activation of ROS- and iron-related genes through the P62-KEAP1-NRF2 signaling cascade is considered a major negative regulator of ferroptosis in HCC cells. Consistent with this, inhibition of the P62-KEAP1-NRF2 pathway enhances the anticancer effects of erastin and sorafenib in vitro and in vivo [38, 39]. Thus, targeting the P62-KEAP1-NRF2-signaling pathway represents a potential therapeutic strategy for triggering lipid peroxidation and promoting ferroptosis. Until now, no study has reported a direct link between curcumin and ferroptosis induction in HCC via inhibition of the P62-KEAP1-NRF2-signaling pathway. Our findings demonstrate that curcumin suppresses this pathway, leading to GSH depletion, increased intracellular Fe²⁺ and MDA accumulation, and enhanced lipid peroxidation in vitro and in vivo. Moreover, OE-P62 reversed the antitumor effects of curcumin on HCC by restoring NRF2 and its downstream effectors, including HO-1 and GPX4, indicating that P62 plays a key role in modulating ferroptosis. Collectively, these findings implicate the regulation of the P62-KEAP1-NRF2-signaling pathway as an important mechanism for curcumin-induced ferroptosis in HCC cells.

Notably, recent work has revealed a positive correlation between autophagy and ferroptosis sensitivity. In KRAS-mutant cells and tumor models, loss of USP13 activated the P62-KEAP1-NRF2-signaling pathway, shifting autophagy toward ferroptosis. This finding suggests that combining autophagy modulation with ferroptosis induction enhances antitumor efficacy [40]. However, in the present study, autophagosomes were not clearly detected by electron microscopy, and Western blot analysis revealed no significant alterations in the levels of the autophagy marker LC3B-II/LC3B-I. Although a secondary involvement of autophagy cannot be entirely ruled out, it is unlikely to constitute the primary mechanism. Thus, we propose that ferroptosis serves as the principal mechanism underlying the anti-tumor effects of curcumin. The potential crosstalk between autophagy and ferroptosis remains unexplored in this study. Future work should aim to elucidate the underlying mechanistic links between these two processes. Another aspect worth highlighting is the role of HO-1 in ferroptosis, which remains controversial. On the one hand, HO-1–derived metabolites such as free iron may exacerbate ferroptosis, while on the other hand, bilirubin has antioxidant properties that counteract lipid peroxidation (Chiang, Chen, & Chang [41]). In our experiments, HO-1 downregulation coincided with ferroptosis promotion, suggesting a net protective effect against tumor survival. However, further studies using genetic manipulation are needed to fully clarify its role.

Conclusions

In conclusion, our study reveals a novel antitumor mechanism by which curcumin induces ferroptosis and underscores the therapeutic relevance of the P62-KEAP1-NRF2-signaling pathway in HCC. Targeting this pathway to selectively weaken the antioxidant defenses of cancer cells, thereby driving ferroptosis, represents a promising strategy for liver cancer treatment.

Abbreviations

HCC

Hepatocellular carcinoma

HepG2

Human Hepatoblastoma G2

Hepa1-6

Hepatoma 1–6

Fer-1

Ferrostatin-1

GSH

Glutathione

GPX4

Glutathione peroxidase 4

ROS

Reactive oxygen species

NRF2

Nuclear factor erythroid 2-related factor 2

KEAP1

Kelch-like ECH-associated protein 1

MDA

Malondialdehyde

Fe²⁺

Ferrous ion

OE-P62

P62 overexpression plasmid

MTT

3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide

V-FITC

Viability-Fluorescein Isothiocyanate

7-AAD

7-Aminoactinomycin D

PBS

Phosphate buffered saline

CCK-8

Cell Counting Kit-8

DMEM

Dulbecco’s Modified Eagle

DMSO

Dimethyl Sulfoxide

EDU

5-Ethynyl-2’-deoxyuridine

siRNA

Small interfering RNA

BODIPY

BODIPY-C11 581/591 staining

DCFH-DA

2’,7’-Dichlorodihydrofluorescein diacetate

BSA

Bovine Serum Albumin

FITC

Fluorescein isothiocyanate

DAPI

4’,6-Diamidino-2-Phenylindole

BCA

Bicinchoninic Acid

SDS-PAGE

Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis

HRP

Horseradish peroxidase

PVDF

Polyvinylidene Fluoride

ECL

Enhanced Chemiluminescence

Authors’ contributions

Jinlan Deng and Zhijuan Wu: Writing - original draft, Data curation. Shangkun Ning: Conceptualization. Xu Chang: Visualization, Formal analysis. Jibing Liu: Supervision. Yangli Yu: Validation. Min Zhang: Writing - review &editing, Data Curation, Resources. Lin Zhang: Writing - review &editing, Data Curation, Funding acquisition. All authors reviewed the manuscript.

Funding

This research was funded by Shandong Traditional Chinese Medicine Science and Technology Project (Q-2022110); Shandong Provincial Natural Science Foundation for Youth (ZR2023QH448); Shandong of The Outstanding Youth Innovation Team of Shandon Institutions of Higher Learning (2023KJ189). Youth Fund Project of Zhongshan Hospital Affiliated to Dalian University (2025ONJJ14).

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

The animal experimental protocol was conducted in accordance with the “National Institutes of Health Guidelines for the Care and Use of Laboratory Animals” and approved by the Institutional Animal Care and Research Advisory Committee of the Shandong University of Traditional Chinese Medicine (Approval No. SDUTCM20241113002; Jinan, China).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.