MINPP1 promotes ferroptosis in HBV-related hepatocellular carcinoma by regulating CTSB K33-linked deubiquitination via ZRANB1

Abstract

Background

See More

Hepatocellular carcinoma (HCC) is significantly influenced by hepatitis B virus (HBV) infection. However, the roles of ferroptosis and ubiquitination modifications in this context remain poorly understood.

Methods

In this study, we utilized immunoprecipitation, immunofluorescence, and analysis of ubiquitin modifications to explore the regulatory mechanisms of MINPP1 in ferroptosis and its effects on tumor progression. Further mechanistic studies revealed that ZRANB1 regulates the K33-linked ubiquitination of CTSB. Ultimately, the contribution of the MINPP1-CTSB axis to tumor progression was validated using in vivo experiments.

Results

Our study demonstrates that MINPP1 regulates ferroptosis in HBV-positive HCC cells via a glycolytic bypass mechanism. Bioinformatics analysis indicates that MINPP1 stabilizes CTSB, thereby participating in the regulation of ferroptosis. Specifically, MINPP1 modulates K33-linked deubiquitination of CTSB through ZRANB1, which stabilizes CTSB expression and identifies its deubiquitination site. In contrast, the MINPP1-ZRANB1-CTSB axis does not regulate ferroptosis in HBV-negative HCC cells. However, upon the introduction of HBV into these cells, the MINPP1-ZRANB1-CTSB axis becomes active and promotes ferroptosis. Finally, in vivo assays showed that MINPP1 regulates tumor progression by regulating K33-linked ubiquitination of CTSB, thereby affecting ferroptosis levels.

Conclusion

Our research showed outcomes suggest that the MINPP1-ZRANB1-CTSB axis promotes ferroptosis in HBV-positive HCC cells through glycolysis, emphasizing the function of MINPP1 in mediating ferroptosis in HBV-related HCC cells via CTSB deubiquitination modification. This provides valuable insights and a foundation for the treatment of HBV-associated HCC.

Introduction

Hepatitis B virus (HBV) infection is a major etiological factor contributing to the development of hepatocellular carcinoma (HCC) globally. HBV infection may progress to chronic hepatitis, eventually leading to cirrhosis and the development of HCC. The occurrence of HCC correlates strongly with the duration of HBV infection and the level of viral load [1]. Currently, the treatment approach for chronic HBV infection primarily involves antiviral drugs such as nucleotide analogs that effectively inhibit HBV replication, reduce hepatic inflammation and fibrosis, and mitigate the risk of HCC in patients [2]. However, there is currently no definitive cure for chronic HBV infection. Consequently, achieving breakthroughs in the clinical diagnosis and treatment strategies for HBV-related HCC remains a significant challenge.

Ferroptosis is a form of cell death triggered by excessive lipid peroxidation [2], characterized by mitochondrial abnormalities, iron accumulation, and lipid peroxidation leading to cell membrane rupture [3]. In recent years, ferroptosis has been widely utilized as an important modality for cancer treatment aimed at eliminating malignant tumor cells [4]. Recent research has shown that inducing ferroptosis can partially mitigate the progression of hepatocellular carcinoma associated with HBV [5]. It has been reported that the HSPA8 gene is upregulated in HBV infection, which inhibits the ferroptotic response of hepatoma cells by enhancing the expression of the SLC7A11/GPX4 axis and diminishing the accumulation of reactive oxygen species and Fe²⁺ levels in both in vivo and in vitro. This gene also stimulates HBV replication, thereby promoting liver cancer development [5]. Additionally, HBV reduces the level of ferroptotic response through SRSF2-mediated abnormal PCLAF splicing, thus enhancing resistance to sorafenib treatment for hepatocellular carcinoma [6]. However, considerable uncertainties persist concerning the therapeutic potential of ferroptosis in treating HBV-associated HCC.

In eukaryotic cells, protein degradation primarily occurs through two major pathways: the autophagy-lysosomal pathway and the ubiquitin-proteasome system (UPS). Among these, the ubiquitin-proteasome system is responsible for approximately 80–85% of all protein degradation processes [7]. Ubiquitin (Ub) is covalently conjugated to the lysine residues of target proteins via a three-step enzymatic cascade that involves the sequential action of an activating E1, a conjugating E2, and a ligase E3 [8]. Not only does ubiquitin play a role in ferroptosis by regulating ferroptosis-related substrates [9, 10], but it also appears to regulate hepatocellular carcinoma development. For instance, silencing MYH9 can inhibit HBx-induced ubiquitination and degradation of GSK3β, thereby attenuating tumor stemness in hepatocellular carcinoma [11], while hepatitis B virus X protein enhances its stability by inhibiting GRP78 ubiquitination through TRIM25 to promote MAN1B1 expression and thereby facilitate liver cancer progression [12]. The potential therapeutic approach of modulating ferroptosis-related substrates via ubiquitination modification for HBV-induced HCC remains largely unexplored.

To this day, the sole identified enzyme in the human genome with the ability to dephosphorylate InsP6 is Multiple Inositol Polyphosphate Phosphatase 1 (MINPP1) [13]. MINPP1 is a key regulator in multiple biological processes, such as apoptosis, endoplasmic reticulum stress, and the formation of bone and cartilage [14, 15]. Recent research has elucidated the link between MINPP1 and pontine cerebellar hypoplasia (PCH), a genetic disorder resulting from loss-of-function mutations in MINPP1 [16, 17]. Our previous investigations have also demonstrated that MINPP1 acts as a tumor suppressor gene implicated in the glycolytic bypass pathway. This pathway is subject to negative regulation by an upstream miRNA (miRNA-30b-5p), which consequently drives increased energy production specifically in HBV-positive HCC [18]. Furthermore, bioinformatics analysis has revealed that MINPP1 governs ferroptosis development. The mechanism underlying the sensitization to ferroptosis following glycolysis inhibition involves a metabolic shift towards oxidative phosphorylation (OXPHOS). This shift results in increased cellular reactive oxygen species (ROS) levels, which disrupt iron homeostasis and promote lipid peroxidation [19]. Tumor cells subjected to erastin or RSL3-induced ferroptosis display markedly diminished glycolytic activity [20]. while TUG1 prevents hepatic stellate cell ferroptosis by upregulating PDK4-mediated glycolysis [21]. However, the role of TUG1 may differ in HBV-related HCC cells. Recent studies have shown that TUG1 affects cell proliferation and apoptosis in HCC by modulating specific signaling pathways [22, 23], suggesting that TUG1 may influence the metabolism and ferroptosis process of HCC cells through different mechanisms.

In this study, we explored the molecular mechanisms by which MINPP1 contributes to the pathogenesis of HBV-positive HCC. Previous studies have reported downregulation of MINPP1 in HBV-positive HCC [18], and subsequent bioinformatics analysis suggested its potential involvement in the ferroptosis process specific to this type of cancer. Herein, we present initial evidence demonstrating that the regulation of glycolysis mediated by MINPP1 governs ferroptosis in HBV-positive hepatocellular carcinoma. Mechanistically, our research results indicate that MINPP1 governs the expression of ZRANB1 protein, and ZRANB1 alleviates the ubiquitination of CTSB (a ferroptosis driver) through K33-linked deubiquitination. Additionally, we identified the specific site where ZRANB1 exerts its deubiquitination modification effect, thereby influencing the susceptibility of HBV-positive HCC cells to ferroptosis. Our research elucidates a novel mechanism by which the MINPP1-ZRANB1-CTSB axis modulates ferroptosis through the glycolytic bypass pathway, thereby influencing the progression of HBV-associated HCC. These results indicate that targeting ferroptosis could represent a promising therapeutic avenue for the treatment of HBV-related HCC.

Methods

Reagents and antibodies

Anti-MINPP1 (PA5-101680), anti-HA (26183-HRP), anti-ACSL4 (PA5-87082), and anti-ZRANB1 (PA5-143420) antibodies were purchased from Thermo Fisher Scientific. Anti-Flag (AF519) antibody was purchased from Beyotime Biotechnology. Anti-GPX4 (67763-1-Ig) antibody was purchased from Proteintech Group. Anti-lactate dehydrogenase (ab52488), anti-Hexokinase II (ab209847), anti-LAMP1 (ab24170), anti-cathepsin B, anti-CTSB (ab125067) and anti-β-actin (ab179467) antibodies were purchased from Abcam. Anti-GLUT-1 antibody (07-1401) was purchased from Sigma-Aldrich. 2-deoxy-D-glucose (D8375), anti-Myc (B7554), and secondary antibodies were purchased from Sigma-Aldrich. Erastin (HY-15763), ferrostatin-1 (HY-100579), deferoxamine (HY-B1625), RSL3 (HY-100218 A), Nec-1s (HY-14622 A), MG132(HY-13259), Z-VAD-FMK (HY-16658B), cycloheximide (HY-12320) and polybrene (HY-112735) were obtained from MedChemExpress.

Tissue specimens

A total of 40 liver tissue samples from patients with HCC were collected from the First Affiliated Hospital of Zhejiang University School of Medicine. Among these, 20 samples were from HCC patients who were positive for HBV, while the remaining 20 samples were from HCC patients who were negative for HBV. All samples were obtained from patients who had been diagnosed and subsequently underwent surgical resection (Ethics approval number: 2024-238-01).

Cell viability assay

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) and trypan blue exclusion assay. A total of 5,000 cells were seeded into each well of a 96-well plate with 200 µL of culture medium. For the CCK-8 assay, Cells were incubated with 100 µL of reaction mixture containing 10 µL CCK-8 solution and 90 µL DMEM for 2 h. The optical density was measured at 450 nm. For the trypan blue exclusion assay, Cells were harvested and mixed with an equal volume of 0.4% trypan blue solution (Sigma-Aldrich). The mixture was then loaded onto a hemocytometer, and viable (unstained) and non-viable (stained) cells were counted under a light microscope. Cell death was calculated as the percentage of stained cells relative to the total cell count.

EdU assay

Stable transfected cell lines were seeded into 96-well plates (2 × 10⁴ cells per well) and cultured in DMEM for one day. Subsequently, the cells were incubated with 50 µmol/L EdU (RiboBio, C10310-2) at 37 ℃ for 2 h. The cells were then fixed with a 4% formaldehyde solution for 30 min, followed by permeabilization with 0.5% Triton X-100 for 10 min. After adding 400 µL of 1 × Apollo reaction mixture, 400 µL of Hoechst 33,342 was also added. The cells were rinsed three times with PBS before observing EdU-positive cells. Finally, images of the cells were captured using a microscope.

Enzyme-linked immunosorbent assay (ELISA)

The levels of hepatitis B surface antigen (HBsAg) and hepatitis B e antigen (HBeAg) in the supernatants of HBV-replicating cells were quantitatively determined using ELISA kits (Sangon Biotech). The experimental procedures were carried out following the guidelines provided in the kit instructions. The culture medium was changed one day before sample collection.

Western blot

Cells subjected to different treatments were harvested, washed with cold PBS, and lysed in RIPA buffer at 4 ℃ for 30 min to obtain total protein extracts. Protein concentrations were measured using the BCA assay. Protein samples of equivalent quantity were subjected to separation on SDS-PAGE gels through electrophoresis and then transferred onto PVDF membranes. After that, the PVDF membranes were treated with a blocking solution containing 5% bovine serum albumin. Next, these membranes were incubated with primary antibodies at 4 ℃ for an overnight period. Subsequently, the membranes were further incubated with secondary antibodies conjugated with horseradish peroxidase for 1–2 h. Protein bands were visualized using chemiluminescence detection.

Immunoprecipitation (IP)

Cells were initially lysed in RIPA buffer supplemented with protease inhibitors on ice for 30 min, followed by centrifugation at 4 ℃ for 15 min to remove cellular debris. The supernatant was then incubated with pre-washed Protein G Sepharose beads at 4 ℃ with rotation for 1–2 h to perform pre-clearing and minimize non-specific binding. Next, the specific antibody was added to the supernatant and incubated at 4 ℃ overnight with gentle agitation. The mixture was subsequently combined with Protein G Sepharose beads and further incubated at 4 ℃ for 2 h to facilitate the binding of antibody-protein complexes to the beads. After washing the beads five times with lysis buffer, the beads were boiled in 1 × SDS-PAGE sample buffer for 5 min to elute the bound proteins. The eluted proteins were then subjected to SDS-PAGE and analyzed by Western blotting.

Ubiquitination assay

Cells were first plated in 12-well plates and cultured in a 5% CO₂ incubator at 37 ℃ until they reached an appropriate density. Subsequently, the cells were co-transfected with the specified plasmid and an HA-tagged ubiquitin plasmid for 36 h, with a control group receiving an empty vector. The culture medium was changed 6 h after transfection. Cells were treated with 20 µM MG132 in the culture medium for 8 h between 24 and 48 h post-transfection. One-half of the harvested cells were directly used for immunoblotting, while the remainder were processed for denaturing co-precipitation. Cell lysis was performed using a denaturing buffer [24], and the lysates were incubated overnight at 4 ℃ with either Myc beads (Beyotime, P2183S) or HA beads (Beyotime, P2121). The beads were then extensively washed with the appropriate buffers. Finally, the samples were subjected to immunoblotting analysis using specific antibodies [25].

Immunofluorescence cell staining

Cells were initially seeded onto culture plates containing sterile coverslips and cultured in an incubator at 37 ℃ with 5% CO₂ until they reached an appropriate density. The cells were then rinsed gently 3 times with pre-chilled PBS. Next, cells were fixed with PBS containing 4% paraformaldehyde at room temperature for 20 min to stabilize cellular structures. After fixation, cells were washed 3 times with PBS, each for 5 min. To enhance membrane permeability and facilitate antibody penetration, cells were incubated with PBS containing 0.1%-0.3% Triton X-100 at room temperature for 10 min, then washed 3 times with PBS. The cells were then blocked with a solution containing 1% BSA, 10% normal goat serum, and 0.05% Tween-20 at room temperature for 1 h to reduce non-specific binding. Without washing, the blocking solution was discarded, and primary antibodies diluted in the blocking solution were added and incubated overnight in a humidified chamber at 4 °C. After incubation, cells were gently washed 3 times with PBS and then incubated with secondary antibodies for approximately 20 min at room temperature. Then, cells were washed 3 times with PBS in the dark, each for 5 min. Ultimately, the nuclei were labeled with DAPI, and fluorescence microscopy was employed to acquire the images. The secondary antibodies used were Alexa Fluor^® 555 (Abcam, ab150078) or Alexa Fluor^® 488 (Abcam, ab150113).

Determination of intracellular ROS, MDA, and GSH

To measure intracellular reactive oxygen species (ROS) levels, we employed the CellROX Green Reagent (Nanjing Warbio Biotechnology Co., Ltd, M10424). Typically, cells were seeded into culture dishes at a density of 1 × 10⁵. After 24 h of incubation, cells were subjected to various treatments. Following treatment, the culture medium was removed, and cells were gently washed twice with pre-warmed PBS. According to the manufacturer’s instructions, the CellROX™ Deep Green Reagent was diluted in serum-free medium to a working concentration of 5 µM and added to the culture plate. Cells were then incubated at 37 ℃ in the dark for 30 min to allow the reagent to fully penetrate the cells and react with ROS. After incubation, cells were washed three times with pre-warmed PBS to remove unbound reagent. Cells were then immediately observed under a fluorescence microscope to detect intracellular ROS levels. Levels of glutathione (GSH) and malondialdehyde (MDA) in cells were quantified using GSH and MDA assay kits purchased from Beyotime Biotechnology.

Mitochondrial superoxide measurements

Cells were incubated with MitoSOX Red (Beyotime, S0061S) at 37 ℃ in a 5% CO₂ incubator in the dark for 30 min. Subsequently, the cells were washed with PBS, and fluorescence images were acquired using a fluorescence microscope.

Mitochondrial membrane potential measurements

Cells were first plated into culture plates at an optimal density, and experimental and control groups were set up based on experimental requirements. Following treatment, the culture medium was carefully removed, and cells were gently rinsed 3 times with PBS pre-warmed to 37 ℃ to eliminate residual impurities. A working solution of TMRE diluted in serum-free medium was then added to cover the cells completely, which were incubated at 37 ℃ in a 5% CO₂ incubator in the dark for 20 min. After incubation, the TMRE (Beyotime, C2001S) solution was promptly discarded, and cells were quickly washed 3 times with pre-warmed PBS. Cells were subsequently examined using a fluorescence microscope. TMRE accumulates in the mitochondria of cells with intact mitochondrial membrane potential, producing red fluorescence.

Intracellular iron assay

Cells were plated into culture plates at an optimal density. Following treatment, the culture medium was carefully removed, and cells were rinsed gently 3 times with PBS pre-warmed to 37 ℃ to eliminate residual impurities. A diluted FerroOrange dye (Cell Signaling Technology, #36104) solution (1:2000, v/v) was then added to the cells, which were incubated at 37 ℃ for 30 min. Fluorescence images were subsequently captured using a confocal microscope.

Cell lines, cell cultures, and transfer

The HBV-positive Hep3B and HBV-negative Huh7 liver cell lines were acquired from the State Key Laboratory of Infectious Diseases at the Shenzhen People’s Hospital.

The HHL-5 cell line was purchased from Bluef (Shanghai) Biotechnology Development Co., Ltd. The LO2 cell line was purchased from Cellverse Co., Ltd. The PLC/PRF/5 cell line was acquired from Beyotime Biotechnology. PLC/PRF/5 and LO2 cell lines were cultured in RPMI-1640 supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin in a humidified incubator at 37 ℃ with 5% CO₂. The other cell lines were cultured in DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin in a humidified incubator at 37 ℃ with 5% CO₂. For transient plasmid expression in Huh7 cells, a PEI transfection solution was used [26]. Plasmid transfection into various cancer cell lines was performed using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. The transient transfection experiment was conducted by introducing the plasmid pHBV1.3, which contains a 1.3-fold HBV DNA fragment, and the GFP expression plasmid (pEGFP-C1, Sangon Biotech) into Huh7 cells. Detailed sequence information is provided in Supplementary Table 1.

Plasmid construction and lentiviral infection

Primers containing specific restriction enzyme sites were designed, and the target gene was amplified by PCR using cDNA as the template. The PCR product was purified following agarose gel electrophoresis and gel extraction. The pGLV3/H1/GFP vector (GenePharma, Shanghai, China) was double-digested with the same restriction enzymes, purified, and ligated to the target gene. The ligation product was transformed into Escherichia coli DH5α, and single colonies were selected for identification and sequencing. The constructed plasmid was co-transfected with packaging and envelope plasmids into HEK 293T cells. Supernatants were collected 48–72 h post-transfection, filtered, concentrated, and counted. For infection, target cells with a confluence of 30%-50% were treated with the virus and polybrene. The medium was replaced after 12–24 h, and GFP expression was assessed at 72 h to evaluate transduction efficiency. Lentiviral vectors were used to establish stable cell lines expressing MINPP1, CSTB-WT, and CSTB-2KR.

Quantitative real time-PCR

RNA extraction was performed on the samples using Trizol reagent (Invitrogen, 15596018CN) following the standard protocol to ensure high integrity and purity of the RNA. A suitable quantity of the extracted RNA was then reverse-transcribed into cDNA using a reverse transcription kit (Bio-Bio Engineering Co., Ltd.), adhering strictly to the reaction conditions outlined in the kit instructions. Then, subsequently, real-time quantitative PCR was conducted using Premix Ex Taq™ kit (TaKaRa, RR390Q). The primer sequences used for the analysis are detailed in Supplementary Table 1.

Xenograft model

Six-week-old female BALB/c nude mice were obtained from the Beijing Center for Laboratory Animals and maintained in a specific pathogen-free (SPF) facility with controlled environmental conditions: temperature at 22 ± 2 ℃, humidity at 50 ± 5%, and a 12-hour light/dark cycle. Mice were provided with unrestricted access to water and sterilized food. For the in vivo experiments, Hep3B cells stably expressing MINPP1, MINPP1 + CTSB-WT, and MINPP1 + CTSB-2KR were harvested during the logarithmic growth phase. The cells were resuspended in DMEM at a density of 3 × 10⁶ cells/200 µL and injected subcutaneously into both flanks of the mice under sterile conditions. Tumor dimensions, including length (L) and width (W), were assessed every other day using calipers, and tumor volume was calculated using the formula V = 0.5 × L × W². After 4 weeks, the mice were euthanized, and tumors were excised under sterile conditions for weighing and further analysis. All animal procedures were conducted following the guidelines of the Animal Care Committee of Zhejiang University, ensuring compliance with animal welfare and ethical standards.

Measurement of metabolic indicators

To assess metabolic parameters, DMEM medium lacking phenol red was pre-warmed in an incubator set at 37 ℃ with 5% CO₂. Cells were plated into culture vessels at an optimal density and incubated for 15 h under stable conditions in the incubator. Following incubation, the culture vessels were carefully removed, and the culture medium was transferred to sterile centrifuge tubes using a pipette. The supernatant was then collected by centrifugation for further analysis. Glucose and lactate levels were measured using assay kits from BioVision (Mountain View, USA), following the manufacturer’s protocols, to evaluate cellular glucose uptake and lactate secretion.

Oxygen consumption rate and assays

The Seahorse XFe 96 Extracellular Flux Analyzer (Seahorse Bioscience, Billerica, MA) was utilized to precisely measure the cellular oxygen consumption rate (OCR), with the specific protocol described as follows. Before the experiment, the Seahorse XF Cell Mitochondrial Stress Test Kit (Agilent Technologies, Santa Clara, CA) was prepared, and the relevant reagents were formulated. Cells in the logarithmic growth phase were plated at a density of 1 × 10⁴ cells per well in a Seahorse XFe 96 cell culture microplate and incubated for 10 h at 37 ℃ with 5% CO₂ to ensure proper cell adhesion and optimal growth conditions. Before the completion of incubation, the Seahorse analyzer was preheated to 37 ℃, and the sensors were calibrated. The culture medium was then replaced with a pre-warmed CO₂-independent assay medium, and the plate was incubated for an additional hour in a CO₂-independent incubator to stabilize the cellular environment. Prepared reagents were loaded into the injection ports of the analyzer, and the plate was inserted to begin the measurement process. Following the measurement, data were analyzed using Seahorse XFe 96 Wave software to determine parameters such as basal respiration, ATP-linked respiration, maximal respiration, and spare respiratory capacity. The results were normalized to cell counts.

Glycolysis stress test

Following the manufacturer’s protocol (103020, Agilent Technologies), we utilized the Seahorse XF96 Analyzer to assess glycolytic capacity. Hep3B cells were seeded into the Seahorse XF96 Cell Culture Microplates and cultured in DMEM medium supplemented with 10% exosome-free fetal bovine serum at 37 °C in a cell incubator. After 12 h, the medium was replaced with a pre-warmed XF base medium containing 2 mM L-glutamine and incubated for 1 h in a non-CO₂ incubator before conducting the Seahorse assay. The baseline extracellular acidification rate (ECAR) was measured first, followed by sequential treatments with 10 mM glucose, 1 mM oligomycin, and 50 mM 2-deoxyglucose (2-DG). The data were analyzed using Wave Desktop 2.6 software (Agilent Technologies).

Immumohistochemical (IHC) staining

IHC staining was conducted in strict accordance with the standardized protocol, with the detailed procedures as follows: Initially, tissue sections were deparaffinized in xylene and then gradually rehydrated using graded ethanol solutions. Subsequently, antigen retrieval was performed by heating the sections in citrate buffer using a microwave. The sections were then incubated in 3% hydrogen peroxide solution at room temperature for 10 min to quench endogenous peroxidase activity. Following this, the sections were blocked with 10% normal goat serum at room temperature for 30 min, and the primary antibody was applied and incubated overnight at 4 °C. On the following day, after washing the sections with PBS buffer, the secondary antibody was added and incubated at room temperature for 2 h. DAB was then used for color development, and the nuclei were counterstained with hematoxylin. Finally, the sections were digitized using a digital slide scanner.

Bioinformatical analysis

In our bioinformatics analysis, we performed Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA) to elucidate the biological functions and pathways involved. We downloaded the expression dataset PDC000198 from the Proteomic Data Commons (PDC) database. To investigate the interaction pattern between ZRANB1 and CTSB, we uploaded their protein structure files to the ClusPro database, performed docking calculations with appropriate parameters, and downloaded the optimal docking model. To further elucidate the mechanism of the ZRANB1-CTSB complex, we uploaded the docking model in PDB format to the PDBePISA online database for analysis of the interaction interface and free energy.

Statistical analysis

To ensure the reliability of the findings, data were derived from a minimum of three independent experiments. Statistical analyses were performed using GraphPad Prism software (version 9.0.0). For comparisons between two groups, paired or unpaired two-tailed Student’s t-tests were utilized. When more than two groups were compared, a one-way analysis of variance (ANOVA) with subsequent multiple comparisons was applied. In experiments involving two factors, comparisons among four or more groups were assessed using two-way ANOVA followed by multiple comparisons. Statistical significance was determined at the levels of *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 to identify significant differences between experimental groups.

Results

MINPP1 induces ferroptosis in Hep3B cells

In previous studies, it has been established that MINPP1 functions as an oncogene involved in the glycolytic bypass pathway, which is negatively regulated by upstream miRNA-30b-5p [18]. This regulation promotes the activation of the glycolytic bypass specifically in HBV-positive HCC cells to enhance energy production. Recently, there have been increasing reports on the interplay between glycolysis and ferroptosis, mediated through metabolic reconnection to oxidative phosphorylation (OXPHOS) [19]. This reconnection leads to elevated cellular ROS levels, disrupting iron homeostasis and promoting lipid peroxidation. We established Hep3B cell lines with overexpression and knockdown of MINPP1 and detected the overexpression and knockdown effects by RT-qPCR (Supplementary Fig.  1A, B). We found that siMINPP1-1 had the best knockdown effect, so we chose siMINPP1-1 for further studies. We assessed relevant metabolic indicators such as glucose consumption, lactic acid production, and cell oxygen consumption rate. Our results demonstrated that overexpression of MINPP1 decreased glycolysis level in Hep3B cells (Supplementary Fig.  1C-F). Conversely, the knockdown of MINPP1 yielded opposite effects (Supplementary Fig.  1C-F), indicating that MINPP1 regulates metabolism by establishing a connection with OXPHOS.

We retrieved proteomic data (PCD00198) from the Proteomic Data Commons database, which represented hepatocellular carcinoma with HBV infection. Based on the median expression level, we categorized MINPP1 into high and low-expression groups to identify differentially expressed proteins and their impact on Gene Ontology (GO), signaling pathways (KEGG), and GSEA (Gene Set Enrichment Analysis). Our findings revealed that MINPP1 was significantly enriched in GO terms related to mitochondrial electron transport NADH to ubiquinone, NADH dehydrogenase complex, and oxidoreductase activity acting on NADPH (Supplementary Fig.  1G). Moreover, GSEA analysis demonstrated the enrichment of oxidative phosphorylation and fatty acid metabolism pathways (Supplementary Fig.  1H), both crucial for maintaining ferroptosis homeostasis. To explore this further, we utilized the FerrDb online database to obtain relevant proteins as well as differential proteins for comparison. Remarkably, more than 60% of these differential proteins overlapped with known ferroptosis-related substrates (Supplementary Fig.  1I). Collectively, our results strongly suggest that MINPP1 potentially participates in the regulation of ferroptosis.

Previous studies have demonstrated that MINPP1 exerts detrimental effects on the growth of HBV-positive HCC. To investigate the specific cell death mechanisms mediated by MINPP1 in HBV-positive HCC, we employed specific inhibitors targeting various modes of cell death. Notably, overexpression of MINPP1 significantly rescued the loss of Hep3B cells induced by ferroptosis activator Erastin when treated with two independent ferroptosis inhibitors (Fer-1 and DFO), while inhibition of necrotic death (Nec-1s) or apoptosis (Z-VAD) had only marginal effects (Fig. 1A and Supplementary Fig. 2A). These findings strongly suggest that ferroptosis is the primary mode of cell death regulated by MINPP1. Furthermore, we assessed mitochondrial superoxide levels in Hep3B cells with stable overexpression or knockdown of MINPP1. Intriguingly, elevated expression levels of MINPP1 were associated with increased mitochondrial superoxide production, whereas downregulation of MINPP1 yielded opposite results (Fig. 1B). Consistently, upregulation and downregulation of malondialdehyde (MDA) and glutathione (GSH) levels were observed upon overexpression of MINPP1 respectively in Hep3B and PLC/PRF/5 cells (Fig. 1C and Supplementary Fig. 2B, D). However, no significant changes were observed in Huh7 cells (Supplementary Fig. 2C, E). Simultaneously, an increase in the levels of intracellular reactive oxygen species (ROS) was also observed in Hep3B cells (Fig. 1D). Moreover, the well-known ferroptosis driver ACSL4 was upregulated while GPX4 was downregulated upon increased expression levels of MINPP1 in Hep3B and PLC/PRF/5 cells (Fig. 1E and Supplementary Fig. 2F), the changes in ACSL4 and GPX4 protein levels were not significant in Huh7 cells (Supplementary Fig. 2G). Additionally, heightened expression levels of MINPP1 led to an increase in cellular iron ion concentration within Hep3B cells (Fig. 1F-G), along with alterations in mitochondrial membrane potential (Fig. 1H). In summary, the results of this study demonstrate that MINPP1 is capable of modulating the ferroptosis process in HBV-positive cells, such as Hep3B and PLC/PRF/5 cells, whereas no significant effects were observed in HBV-negative cells, exemplified by Huh7 cells. These findings provide robust experimental evidence for the role of MINPP1 in promoting ferroptosis in HBV-positive cells and further elucidate the functional differences of MINPP1 across distinct cellular contexts and its underlying biological significance.

Fig. 1

MINPP1 induces ferroptosis in Hep3B cells. A: Effects of Erastin (10 µM) combined with DFO (10 µmol/L), Fer-1 (1 µmol/L), ZVAD-FMK (10 µmol/L), and Nec-1s (10 µmol/L) on Hep3B cell viability after 24 h as measured by the CCK8 assay. B: Representative images showing mitochondrial superoxide levels in stable MINPP1 knockdown or overexpressing Hep3B cells. Scale bars represent 20 μm. C: Measurement of MDA and GSH levels in stable MINPP1 knockdown or overexpressing Hep3B cells. D: Assessment of reactive oxygen species (ROS) levels in stable MINPP1 knockdown or overexpressing Hep3B cells, with scale bars representing 20 μm. E: Protein expression levels of ACSL4 and GPX4 in Hep3B cells transfected with siMINPP1 or OE-MINPP1 expression vectors. F-G: Detection of labile iron ion levels in stable MINPP1 knockdown or overexpressing Hep3B cells using FerroOrange dye. Scale bars represent 20 μm. H: Measurement of mitochondrial membrane potential in stable MINPP1 knockdown or overexpressing Hep3B cells

The regulation of ferroptosis through glycolysis is mediated by MINPP1

Recent literature has demonstrated that TUG1 plays a role in preventing ferroptosis in hepatic stellate cells by upregulating PDK4-mediated glycolysis [21]. Based on this finding, we hypothesized that MINPP1 regulates ferroptosis through the glycolytic pathway. To investigate the effect of MINPP1 on ferroptosis after inhibiting glycolysis, we employed a glycolysis inhibitor called 2-Deoxy-D-glucose (2-DG). The experimental results showed that in HBV-positive cells, the knockdown of MINPP1 significantly increased the protein levels of glycolysis markers, including hexokinase 2 (HK2), glucose transporter 1 (GLUT1), and lactate dehydrogenase A (LDH-A). However, treatment with 2-deoxy-D-glucose (2-DG) on this basis led to a significant decrease in these protein levels. In contrast, such changes were not evident in HBV-negative cells, indicating that 2-DG specifically inhibits glycolysis in HBV-positive cells (Fig. 2A and Supplementary Fig. 3A, B). We further employed the Seahorse assay to investigate the state of cellular glycolysis and observed that, compared with cells transfected with siMINPP1, the glycolytic rate and glycolytic capacity were inhibited in cells treated with 2-DG (Fig. 2B). Subsequently, we validated the markers associated with ferroptosis. Our results revealed that the addition of 2-DG mitigated the decrease in mitochondrial superoxide levels caused by MINPP1 inhibition in Hep3B cells, as depicted in Fig. 2C. Furthermore, it also ameliorated the decline in MDA levels and restored GSH levels in HBV-positive cells (Fig. 2D and Supplementary Fig. 3C, D). In addition, Fig. 2E and Supplementary Fig. 3E demonstrated that the introduction of 2-DG reinstated ROS levels in Hep3B and PLC/PRF/5 cells with the knockdown of MINPP1. The mitochondrial membrane potential was also improved by treatment with 2-DG (Fig. 2F and Supplementary Fig. 3F). Simultaneously, iron ion concentration within the cells increased upon exposure to 2-DG (Fig. 2G). These observations suggest that through its regulation of glycolysis, MINPP1 can alleviate the inhibition of ferroptosis caused by low expression levels observed in HBV-positive cells.

Fig. 2

The regulation of ferroptosis through glycolysis mediated by MINPP1. A: The protein expression levels of GLUT-1, LDH-A, and HK2 were detected using western blots. β-actin was used as an internal control. B: The extracellular acidification rate (ECAR) was measured using an XF96 Seahorse Analyzer. C: Representative images showing mitochondrial superoxide levels in Hep3B cells with MINPP1 knockdown treated with 2-Deoxy-D-glucose (2-DG, 4 µmol/L, 24 h), along with quantification data. Scale bars represent 20 μm. D: Measurement of Malondialdehyde (MDA) and Glutathione (GSH) levels in Hep3B cells with MINPP1 knockdown treated with 2-DG (4 µmol/L, 24 h). E: Assessment of Reactive Oxygen Species (ROS) levels in Hep3B cells with MINPP1 knockdown treated with 2-DG (4 µmol/L, 24 h). F: Quantification of mitochondrial membrane potential in Hep3B cells with MINPP1 knockdown treated with 2-DG (4 µmol/L, 24 h). G: FerroOrange staining for detection of labile iron ion levels in Hep3B cells with MINPP1 knockdown treated with 2-DG (4 µmol/L, 24 h). Scale bars represent 20 μm

The protein level reveals a significant correlation between MINPP1 and CTSB

To elucidate the mechanism through which MINPP1 facilitates iron-induced cell death in Hep3B cells, we retrieved protein interactions involving MINPP1 from the BioGRID database and identified proteins related to iron-induced cell death from the FerrDb database. By comparing their overlap factors, specifically CTSB and DPP4 (Fig. 3A), previous KEGG analysis revealed that MINPP1 is implicated in the process of 26 S proteasome-mediated protein degradation (Fig. 3B). Therefore, it is hypothesized that MINPP1 may regulate downstream proteins at the protein level. To explore this further, we utilized downloaded PCD00198 data to generate a correlation plot between CTSB and MINPP1, as well as DPP4 and MINPP1. The findings revealed a positive correlation between MINPP1 and CTSB expression at the protein level (Fig. 3C), while no correlation was observed with DPP4 (Fig. 3D). Moreover, MINPP1 expression was categorized into high and low groups according to median values, and differentially expressed proteins were subsequently identified through volcano plot analysis. Notably, CTSB was discovered among these differentially expressed proteins (Fig. 3E). We continued to utilize the results of RNA sequencing and observed no significant correlation between CTSB and MINPP1 in HBV-positive HCC samples (GSE140400) (Fig. 3F). Additionally, we downloaded GSE55092 and obtained consistent findings in HBV-positive HCC samples (Fig. 3G). These integrated findings provide compelling evidence that in HBV-positive HCC, MINPP1 and CTSB display a substantial positive correlation at the protein level, rather than at the transcriptional level.

Fig. 3

The protein level reveals a significant correlation between MINPP1 and CTSB. A: Venn diagram displaying overlapping genes between MINPP1 interacting substrates and genes associated with ferroptosis. B: KEGG Pathway Analysis of MINPP1. C: Correlation analysis of protein levels between MINPP1 and CTSB. D: Correlation analysis of protein levels between MINPP1 and DPP4. E: MINPP1 Differential Genes with Median as a Boundary. F and G: Correlation analysis of mRNA transcription levels between MINPP1 and CTSB

The CTSB attenuates the ferroptosis induced by MINPP1

To further elucidate whether the role of MINPP1 in promoting ferroptosis in Hep3B cells was mediated by CTSB, we initially examined the impact of MINPP1 knockdown on CTSB protein expression in Hep3B, PLC/PRF/5, and Huh7 cells. The findings revealed a substantial reduction in CTSB protein levels following MINPP1 knockdown in HBV-positive cells (Fig. 4A, B and Supplementary Fig. 4A, B). Simultaneously, we explored whether CTSB modulates MINPP1-induced cell death through an EdU assay, and discovered that the supplementation of CTSB mitigated the promoting effect of MINPP1 knockdown on cell proliferation in HBV-positive cells (Fig. 4C and Supplementary Fig. 4C, D). Furthermore, under conditions where CTSB was compensated for, we assessed common indicators of ferroptosis and found that the addition of CTSB mitigated the inhibitory effects of low MINPP1 expression on ferroptosis. This was specifically manifested as an increase in mitochondrial superoxide levels (Fig. 4D), elevated ROS levels (Fig. 4E), alterations in MDA and GSH levels (Fig. 4F, G and Supplementary Fig. 4E, F), changes in ACSL4 and GPX4 protein expressions (Supplementary Fig. 4G), improvements in mitochondrial membrane potential, and an elevation in cellular iron ion concentration in HBV-positive cells (Supplementary Fig. 4H, I). The glycolytic bypass, being an oxidative process under hypoxia, prompted us to investigate the potential regulatory role of CTSB in hypoxic MINPP1-induced tumor proliferation. As anticipated, CCK-8 assays indicated that the addition of CTSB notably counteracted the enhancement of Hep3B cell proliferation caused by hypoxic MINPP1 knockdown (Supplementary Fig. 4J). However, when CTSB expression was inhibited, overexpression of MINPP1 appeared to more effectively inhibit hypoxic Hep3B cell proliferation compared to normoxia (Supplementary Fig. 4K). These findings indicate that CTSB mitigates the effects of MINPP1 on ferroptosis.

Fig. 4

The CTSB attenuates the ferroptosis induced by MINPP1. A and B: Western blot analysis of the expression of relevant proteins (A) and their quantification (B). C: EdU analysis was used to detect the cell proliferation status of each group. D: Quantification of mitochondrial superoxide levels in Hep3B cells transfected or not with CTSB after silencing MINPP1. E: Quantification of reactive oxygen species (ROS) levels in Hep3B cells transfected or not with CTSB after silencing MINPP1. F and G: Measurement of MDA (F) and GSH (G) levels in Hep3B cells transfected or not with CTSB after silencing MINPP1

The stability of CTSB protein is enhanced by MINPP1

It has been reported that the O-GlcNAcylation at S210 of CTSB may influence the stability of mature CTSB, rather than affecting the conversion of pro-CTSB to its mature form [25]. The previous experiments have demonstrated that MINPP1 also exerts an influence on the expression of CTSB protein in HBV-positive cells (Fig. 4A and Supplementary Fig. 4A, B). However, our findings indicate that MINPP1 does not impact the expression of CTSB mRNA (Supplementary Fig. 5A). Additionally, it was observed that treatment with MG132 attenuated the reduction in CTSB expression induced by MINPP1 knockdown (Fig. 5A). Subsequently, we investigated the effect of MINPP1 on CTSB half-life using CHX drug treatment and found that MINPP1 knockdown promoted mature CTSB protein degradation (Fig. 5B and Supplementary Fig. 5B). Considering that CTSB primarily localizes to lysosomes, we discovered colocalization between MINPP1 and lysosomes in Hep3B cells, particularly those with larger diameters (Fig. 5C). Additionally, an immunoprecipitation experiment further confirmed the interaction between CTSB and MINPP1 (Fig. 5D).

Fig. 5

The stability of CTSB protein is enhanced by MINPP1. A: Western blot analysis and quantification of intracellular CTSB expression in Hep3B cells with or without knockdown of MINPP1 treated with the proteasome inhibitor MG132. B: Hep3B cells with or without knockdown of MINPP1 were treated with cycloheximide (CHX) at a concentration of 100 µg/ml for various periods. Protein expression was detected by Western blotting. C: Localization detection by immunofluorescence (IF) staining. The scale bar represents 20 μm. D: Co-immunoprecipitation (Co-IP) analysis to detect the interaction between MINPP1 and CTSB

ZRANB1 interacts with CTSB

Next, we investigated the mechanisms underlying MINPP1-mediated regulation of CTSB stability. We retrieved the interaction proteins of MINPP1 and CTSS from BioGRID and identified eight overlapping proteins (Supplementary Fig. 6A). Given our previous bioinformatics analysis indicating a role for MINPP1 in proteasome-mediated protein degradation (Fig. 3B), we focused on investigating the involvement of the E3 ligase TRIM67 and deubiquitinating enzyme ZRANB1. Intriguingly, previous data indicated that ZRANB1 was conspicuously downregulated in HBV-positive HCC, which impelled us to explore its role herein (18). Through the detection of the expression levels of ZRANB1 and MINPP1 under various circumstances, the regulatory relationship of MINPP1 over ZRANB1 was determined (Fig. 6A), and it was discovered that the expression of CTSB decreased with the inhibition of ZRANB1 (Fig. 6B), without affecting CTSB mRNA levels (Supplementary Fig. 6B). We further examined the effect of ZRANB1 on the stability of CTSB protein. Our results indicated that ZRANB1 markedly increased the stability of the mature CTSB protein (Fig. 6C and Supplementary Fig. 6C). Subsequently, we generated three-dimensional structural models for both CTSB and ZRANB1 (Supplementary Fig. 6D, E). Furthermore, we calculated the interaction area and complexation-free energy between CTSB and ZRANB1 using PDBePISA (Supplementary Fig. 6F). Notably, our results indicated a minimal binding free energy between ZRANB1 and CTSB, suggesting the formation of a stable complex. We utilized PyMOL to visualize the docking interface (Fig. 6D). Immunofluorescence and Co-immunoprecipitation (Co-IP) further corroborated the correlation between CTSB and ZRANB1 (Fig. 6E, F).

Fig. 6

ZRANB1 interacts with CTSB. A: Western blot analysis was performed to detect and quantify the expression levels of ZRANB1 in Hep3B cells following MINPP1 knockdown. Additionally, Western blot analysis was used to assess and quantify the expression levels of MINPP1 in Hep3B cells after ZRANB1 knockdown. B: Western blot analysis was performed to detect and quantify the expression levels of CTSB in Hep3B cells with ZRANB1 knockdown. C: Hep3B cells were transfected with ZRANB1 plasmid or corresponding control plasmid, followed by treatment with cycloheximide (CHX) at a concentration of 100 µg/ml for designated periods. Protein expression was detected using Western blot analysis. D: PyMOL was used to visualize the interaction interface between ZRANB1 and CTSB. E: Immunofluorescence (IF) was employed for the detection and localization of ZRANB1 and CTSB co-localization. The scale bar represents 20 μm. F: Co-immunoprecipitation (Co-IP) analysis was carried out to assess the interaction between ZRANB1 and CTSB

The ubiquitination of CTSB is regulated by ZRANB1

As widely recognized, ZRANB1 functions as a deubiquitinating enzyme within the ubiquitination modification system. To elucidate the regulatory role of ZRANB1 in CTSB stability, we initially generated a C443A mutant that lacked deubiquitinating enzyme activity [26]. Western blot experiments revealed that ZRANB1-WT significantly enhanced CTSB protein expression, whereas ZRANB1-C443A attenuated its regulation (Supplementary Fig. 7A). Our findings suggest that the influence of ZRANB1 on CTSB protein stability is predominantly mediated by its deubiquitinating enzyme activity. To investigate the interaction region between ZRANB1 and CTSB, truncated mutants encompassing different regions of both proteins were constructed (Supplementary Fig. 7B). Co-IP experiments revealed that the binding of ZRANB1 to CTSB was reliant on the presence of the OTU domain (Fig. 7A), while the interaction between CTSB and ZRANB1 mainly relied on the heavy chain of CTSB (Fig. 7B). Furthermore, ubiquitination modification experiments revealed that overexpression of ZRANB1 significantly reduced ubiquitination modification of CTSB protein in HBV-positive cells (Fig. 7C and Supplementary Fig. 7C, D). Our previous research has identified MINPP1’s effect on CTSB protein stability. To verify whether MINPP1 affects the ubiquitination of CTSB, we conducted in vitro ubiquitination experiments, which showed that the addition of MINPP1 weakened the ubiquitination of CTSB; moreover, the coexistence of both MINPP1 and ZRANB1 further diminished this process in HBV-positive cells (Fig. 7C and Supplementary Fig. 7C, D). To further clarify whether the regulation of CTSB ubiquitination by ZRANB1 depends on its enzymatic activity, we detected the ubiquitination modification of the CTSB protein in cells with wild-type (WT) and catalytically inactive mutant (CAmut) ZRANB1. As expected, when ZRANB1 lost its deubiquitinating enzyme activity, it was unable to regulate the ubiquitination of CTSB (Fig. 7D), indicating that the regulation of CTSB ubiquitination by ZRANB1 is dependent on its deubiquitinating enzyme activity. It has been reported that ZRANB1 mainly cleaves ubiquitin chains linked via K29 and K33 and exhibits higher activity towards K29 and K33-linked ubiquitins than K63-linked ubiquitins [27]. Experiments showed that ZRANB1 mainly regulates the deubiquitination of K33-linked ubiquitination (Supplementary Fig. 7E). To understand the ubiquitination sites of CTSB, we first predicted its ubiquitination sites using the GPS-Uber online database (Supplementary Fig. 7F) and further analyzed their conservation across three species using ESPript. We found that K220 and K237 were conserved among these species, and based on this, we constructed K220R and K237R mutants (Supplementary Fig. 7G). We subsequently investigated the impact of ZRANB1 on the ubiquitination status of CTSB-2KR. The findings revealed that when CTSB was subjected to a 2KR mutation, the deubiquitination activity of ZRANB1 was markedly diminished, reaching a level similar to that observed with the ZRANB1-C443A mutant (Fig. 7D), suggesting that ZRANB1 performs deubiquitination at the K220, K237 sites of CTSB.

Fig. 7

Ubiquitination of CTSB is regulated by ZRANB1 to affect ferroptosis. A: Western blot analysis of Hep3B cells transfected with Myc-tagged CTSB alone for 36 h or co-transfected with Flag-tagged WT ZRANB1 or its deletion mutant, followed by immunodetection with specific antibodies. B: Western blot analysis of Hep3B cells transfected with Flag-tagged ZRANB1 alone for 36 h or co-transfected with Myc-tagged CTSB or its deletion mutant, followed by immunodetection with specific antibodies. C: Stable Hep3B cells expressing control Flag-Vector or Flag-ZRANB1 were transfected with Myc-tagged CTSB and HA-tagged UB either alone or with MINPP1 and treated with MG132 (20 µM) for 8 h after 36 h of transfection. Cell lysates were subjected to HA pull-down beads, followed by immunodetection with specific antibodies. D: Hep3B cells were transfected with Myc-tagged CTSB wild-type and mutants, HA-tagged UB, Flag-tagged WT ZRANB1, and its mutants for 36 h and then treated with MG132 (20 µM) for 8 h. Cell lysates were immunoprecipitated with Myc antibody, followed by immunodetection with specific antibodies. E: Hep3B cells were transfected with CTSB-WT or mutant (2KR) and then treated with cycloheximide (CHX) (100 µg/ml) for designated periods. Protein expression was detected and quantified via Western blot analysis

Effects of ubiquitination of CTSB on ferroptosis

To explore how CTSB ubiquitination influenced its role in regulating MINPP1 in ferroptosis, we established a stable Hep3B cell line expressing CTSB-2KR (K220, 237R). In the CHX experiment, we observed that when ZRANB1 failed to deubiquitinate CTSB (CTSB-2KR), CTSB degradation was accelerated, as depicted in Fig. 7E and Supplementary Fig. 8A. As anticipated, the results from the EdU experiment demonstrated that compared to CTSB-2KR (Supplementary Fig. 8B), CTSB-WT more effectively mitigated the pro-proliferative effect of MINPP1 on Hep3B cells. However, it is noteworthy that even though CTSB-2KR significantly inhibited ferroptosis by exhibiting lower levels of mitochondrial superoxide (Supplementary Fig. 8C) and MDA (Supplementary Fig. 8D), along with higher GSH levels (Supplementary Fig. 8E) than CTSB-WT did, it also displayed reduced ROS levels (Supplementary Fig. 8F), elevated GPX4 protein expression, decreased ACSL4 protein expression (Supplementary Fig. 8G), increased mitochondrial membrane potential (Supplementary Fig. 8H), and decreased intracellular iron concentration (Supplementary Fig. 8I). To explore whether ubiquitination of CTSB affects the influence of MINPP1 on tumor proliferation in vivo, we subcutaneously injected stable Hep3B cells expressing either CTSB-WT or CTSB-2KR into mice and monitored tumor volume and weight changes over time. The findings indicated that compared to tumors formed by cells expressing CTAS-2KR (Fig. 8A, B), those formed by cells expressing WT exhibited better inhibition of tumor growth.

HBV infection is key to the regulation of ferroptosis by ZRANB1 promoting CTSB deubiquitination

Given that our initial experiments were performed in the HBV-positive Hep3B cell line, we proposed that HBV infection might significantly influence CTSB stability via a MINPP1-mediated ZRANB1 pathway, thereby affecting ferroptosis and promoting tumor progression. First, we detected the protein expression levels of MINPP1, ZRANB1, and CTSB in normal human hepatocyte cell lines, HBV-positive cell lines, and HBV-negative cell lines. According to the expression levels of MINPP1, ZRANB1, and CTSB in these cell lines, we found that the expressions of MINPP1, ZRANB1, and CTSB were reduced only in HBV-positive cells (Supplementary Fig. 9A). Additionally, we examined the expression of MINPP1 and CTSB in tissue sections from HBV-negative and HBV-positive HCC patients using immunohistochemical staining (Supplementary Fig. 9B). The results showed that these two proteins were lowly expressed in HBV-positive HCC patients, while they were highly expressed in HBV-negative HCC patients (Supplementary Fig. 9C). Subsequently, we transfected Huh7 cells (HBV-negative) with pHBV1.3 plasmid, a GFP-expressing plasmid, or an empty vector to further demonstrate the effect of HBV infection on CTSB. As shown in Supplementary Fig. 9D, the transfection of pHBV1.3 was successfully achieved. We investigated the biological functions of ZRANB1 knockdown. As shown in Supplementary Fig. 9E, ZRANB1 knockdown enhanced the viability of pHBV1.3-transfected Huh7 cells, whereas the reintroduction of CTSB markedly suppressed cell proliferation. Knockdown of ZRANB1 significantly inhibited ferroptosis in pHBV1.3-transfected Huh7 cells. Specifically, siZRANB1 reduced ACSL4 protein expression and elevated GPX4 protein expression (Fig. 8C). Meanwhile, siZRANB1 resulted in a decreased level of mitochondrial superoxide (Fig. 8D) and MDA (Supplementary Fig. 9F), an elevated level of GSH (Supplementary Fig. 9G), and an increased mitochondrial membrane potential (Supplementary Fig. 9H), while the intracellular ROS level was reduced (Fig. 8E). The addition of CTSB mitigated these phenotypes observed after siZRANB1 treatment. However, no significant changes were detected in the blank control group and GFP-transfected group. This suggests that HBV infection is the key to the regulation of ferroptosis by ZRANB1 promoting CTSB deubiquitination.

Fig. 8

The interplay among CTSB ubiquitination, tumor proliferation, HBV infection, and ZRANB1-mediated regulation of ferroptosis. A: Tumor volume and weight quantification following subcutaneous injection/implantation of stable MINPP1 and CTSB-WT or mutant (2KR) expressing Hep3B cells into nude mice. B: Representative immunohistochemical staining images of tumor tissue were performed using related antibodies. C: Protein expression levels of ACSL4 and GPX4 in Huh7 cells with MINPP1 knockdown or controls transfected with or without GFP, pHBV1.3, and CTSB were assessed. D: Mitochondrial superoxide levels in Huh7 cells with MINPP1 knockdown or controls transfected with or without pHBV1.3 and CTSB. Scale bars represent 20 μm. E: ROS levels in Huh7 cells with MINPP1 knockdown or controls transfected with or without pHBV1.3 and CTSB. Scale bars represent 20 μm. F: Schematic representation of the molecular mechanism

Discussion

Hepatitis B virus (HBV) infection is a major global risk factor for hepatocellular carcinoma (HCC). HBV can directly promote carcinogenesis through integration into the human genome [1]. Although vaccination programs have been widely implemented, resulting in a substantial reduction in new HBV infections among children, chronic HBV infection remains a major contributor to HCC globally, accounting for over half of all cases. In regions with high infection rates, this proportion can reach up to 85% [28, 29]. Patients undergoing antiviral therapy, even those with sustained viral suppression, continue to be at risk for hepatocellular carcinoma development [30]. Thus, elucidating the mechanisms by which HBV infection leads to HCC is essential for the effective treatment of HBV-positive HCC patients.

In this study, we have demonstrated that MINPP1 modulates ferroptosis in hepatocellular carcinoma cells via the glycolytic bypass. We highlight that HBV infection is a critical determinant for MINPP1-mediated regulation of ferroptosis. The underlying mechanism lies in MINPP1 regulating the expression of the deubiquitinase ZRANB1, which enhances the K33-linked deubiquitination of CTSB (Fig. 8F). In-depth research has identified the deubiquitination site of CTSB by ZRANB1, highlighting the key role of CTSB deubiquitination modification in MINPP1-mediated ferroptosis of HBV-related HCC cells. This study offers valuable therapeutic bases for HBV-related HCC.

Ferroptosis is a form of iron-dependent cell death, characterized by cellular iron overload and lipid peroxidation [31]. In chronic liver injury, ferroptosis exacerbates liver damage through various mechanisms. For instance, in non-alcoholic fatty liver disease (NAFLD), iron overload is closely related to disease progression [32]. Moreover, ferroptosis is also involved in drug-induced liver injury (DILI) and liver ischemia/reperfusion injury (IRI) [33,34,35]. In hepatocellular carcinoma (HCC), the role of ferroptosis is more complex. On the one hand, ferroptosis can increase the sensitivity of HCC cells to therapeutic drugs (such as sorafenib) [31]; on the other hand, some studies have found that ferroptosis may promote the development of HCC by exacerbating liver injury [36]. Although ferroptosis has potential application value in HCC treatment, its potential adverse effects on normal liver tissue should not be ignored. Firstly, ferroptosis inducers may cause iron overload and lipid peroxidation in normal hepatocytes, leading to cell damage and death [31]. Secondly, ferroptosis inducers may disrupt the liver’s antioxidant defense mechanisms, such as reducing glutathione (GSH) levels and the activity of glutathione peroxidase 4 (GPX4) [37]. These changes may lead to increased oxidative stress in normal liver tissue, thereby triggering inflammatory responses and fibrosis. In addition, ferroptosis inducers may affect the liver’s metabolic functions, such as lipid metabolism and amino acid metabolism [38]. These metabolic disorders may further affect the normal physiological functions of the liver and even promote the progression of chronic liver disease.

CTSB is an intracellular protease localized in the lysosomal membrane, exhibiting peptide-chain-cleaving activity and catalyzing protein hydrolysis within both lysosomal and extracellular environments. It has been demonstrated to participate in autophagy-induced inflammasome activation [39]. Moreover, it plays a crucial role in regulating autophagy, metabolism, cellular stress signal transduction, antigen presentation, and lysosome-dependent cell death [40]. CTSB is prominently implicated in various pathological processes, particularly cancer [41]. The association between CTSB and ferroptosis has been reported [42]. Post-translational modifications have been shown to regulate CTSB function; for example, the O-GlcNAcylation of CTSB at S210 may influence the stability of mature CTSB, rather than affecting the conversion of pro-CTSB to its mature form [25]. The present report also elucidated the interaction between MINPP1 and CTSB, thereby modulating the stability of CTSB protein to effectively regulate ferroptosis.

Our previous RNAseq results also indicated the downregulation of ZRANB1 in HBV-positive hepatocellular carcinoma. This finding further implies that ZRANB1 significantly contributes to the progression of HBV-positive hepatocellular carcinoma. Consequently, we initiated a study to explore the regulatory effects of ZRANB1 on CTSB protein. ZRANB1 is a deubiquitinating enzyme that selectively dismantles ubiquitin chains linked through K29 and K33 [27]. Previous studies have demonstrated that ZRANB1-mediated deubiquitination and stabilization of VPS34 act as positive regulators of autophagy [43]. We determined the interaction domain between ZRANB1 and CTSB through truncation analysis and Co-IP experiments, subsequently confirming that ZRANB1 enhances the stability of CTSB protein in a manner dependent on its deubiquitinating enzyme activity. Furthermore, we identified the specific type of deubiquitination mediated by ZRANB1 on CTSB, as well as the precise sites of deubiquitination modification. However, it is noteworthy that compared to the ZRANB1-C443A mutant, an elevated level of ubiquitination modification is observed on CTSB-2KR. This suggests the presence of potential additional modification sites that require further exploration.

In summary, our study revealed that MINPP1 enhances ferroptosis in HBV-positive HCC and suppresses the proliferation of Hep3B cells. The critical role of HBV infection in this process was also identified. Furthermore, we investigated the mechanism by which MINPP1 regulates ferroptosis. Specifically, MINPP1 modulates the expression of the ZRANB1 protein, thereby influencing the K33-linked deubiquitination of CTSB, predominantly at the K220 and K237 sites. This results in increased stability of the CTSB protein and promotes ferroptosis in HBV-positive HCC cells. These results identify a new molecular target for HCC treatment and shed light on the development of novel targeted therapies. However, it is essential to recognize certain limitations in our study. First, the study focused solely on the exploration of mechanisms at the cellular level and has not yet been validated in animal models or clinical samples. Although the cellular experiments revealed the potential role of ferroptosis regulation in HBV-related HCC, the lack of in vivo validation limits the extrapolation of the results. Secondly, in the selection of cell lines, the differences in p53 background between Hep3B (p53-deficient) and Huh7 (mutant p53) cell lines were not fully considered, which may interfere with the experimental results. Future studies will improve the mechanistic analysis by introducing p53 wild-type cell models and conducting p53 functional validation experiments. Additionally, the efficacy and safety will be validated in animal models to address the current limitations of the study. Finally, we did not explore the influence of MINPP1 on immune checkpoint (ICB) therapy in liver cancer patients, nor did we examine its effects on drug sensitivity. Additionally, our study lacked clinical evidence to corroborate. Therefore, additional research is required to comprehensively evaluate MINPP1 as a potential therapeutic strategy for liver cancer patients.