USP11 is involved in the sensitivity of liver cancer cells to ferroptosis and taxanes through the regulation of NRF2 ubiquitin-mediated degradation.

Introduction
Hepatocellular carcinoma (HCC) is the sixth most common type of cancer worldwide and the fourth leading cause of cancer-related death [1], primarily due to hepatitis, persistent viral infections, alcohol consumption, contact with carcinogens, and genetic disorders [2,3]. While surgical removal is the optimal treatment for HCC, only 20% to 30% of HCC cases are amenable to surgical excision [4]. Chemotherapy, alongside surgery, remains the primary therapy for HCC, yet prolonged chemotherapy may lead to drug resistance [5], curtailing the effectiveness of chemotherapy and posing a significant threat to HCC patient survival. Therefore, identifying the relevant molecules or targets that influence the progression of HCC and the sensitivity to chemotherapy is highly important.
Taxanes, such as paclitaxel, docetaxel, and cabazitaxel, serve as microtubule inhibitors in chemotherapy, contributing to cancer therapy by preventing the breakdown of microtubules, disrupting their equilibrium, and impeding the progression of the G2‒M phase of the cell cycle [6]. Recently, taxanes have been applied to treat a range of cancers, such as breast cancer, prostate cancer, lung adenocarcinoma, ovarian cancer, and HCC [[7], [8], [9], [10]]. Research indicates that taxanes have the capacity to influence various cancer-causing signaling routes and exhibit potent cytotoxic effects on HCC [11]. However, drug resistance remains a key factor contributing to the poor prognosis of patients with HCC [12]. Therefore, enhancing the sensitivity of HCC to taxanes is a pressing issue that currently needs to be addressed.
Ferroptosis is a regulated form of cell death triggered by iron-dependent lipid peroxidation [13], and it is associated with various physiological and pathological processes. Increasing evidence suggests that the suppression of the cystine/glutamate transporter solute carrier family 7 member 11 (SLC7A11/xCT) and glutathione peroxidase 4 (GPX4) and the accumulation of iron and lipid reactive oxygen species (ROS) can induce ferroptosis [[14], [15], [16]]. Earlier research has indicated that blocking ubiquitin-specific protease 8 (USP8) hinders the expansion of HCC tumors through triggering cellular ferroptosis [17]. Another study indicated that transmembrane protein 147 (TMEM147) exacerbates HCC development by hindering ferroptosis [18]. Consequently, triggering ferroptosis may represent a novel approach to prevent HCC.
Ubiquitin dysregulation is closely related to the occurrence and progression of human cancers [17]. As a member of the deubiquitinase family, ubiquitin-specific protease 11 (USP11) can remove ubiquitin from proteins, prevents substrate degradation, and has been verified to play a role in the development of HCC [19]. High levels of USP11 promote the growth of HCC tumors and vascular infiltration in tissues, and they are also associated with a poor prognosis in patients with HCC [20]. Furthermore, USP11 can participate in cancer progression by regulating ferroptosis. For example, USP11 promotes the progression of breast cancer by inhibiting ferroptosis through phosphoglycerate mutase 5 (PGAM5) [21]. USP11 inhibits ferroptosis in colorectal cancer by mediating the deubiquitination of lymphoid-specific helicase (LSH) [22]. Studies have shown that overexpression of USP11 promotes the growth and migration of gastric cancer cells and increases their resistance to paclitaxel [23]. However, the role of USP11 in ferroptosis and taxane drug sensitivity in HCC remains unknown.
Furthermore, nuclear factor erythroid-2-related factor 2 (NRF2), a transcription factor linked to antioxidant proteins, plays a role in regulating key biological functions related to reducing ROS and defending against oxidative stress [24]. Reports suggest that NRF2 is involved in the metastasis of HCC [25] and is upregulated in HCC, and its expression is associated with a poor prognosis and a malignant phenotype [26]. Furthermore, NRF2 plays a crucial role in regulating ferroptosis [27]. Studies have shown that the downregulation of NRF2 inhibits the expression of SLC7A11 and GPX4, leading to ferroptosis in HCC cells [28]. Additionally, NRF2 is a key factor in the development of chemotherapy resistance in HCC [29]. Inhibiting the nuclear translocation and activation of NRF2 can induce ferroptosis to overcome the resistance of HCC to lenvatinib and sorafenib [30,31]. However, it remains unclear whether NRF2-mediated ferroptosis affects the sensitivity of HCC to taxanes. Furthermore, the regulation of ubiquitination and deubiquitination can modulate the stability of NRF2. For example, the E3 ubiquitin ligase Mind bomb 1 (MIB1) promotes the proteasomal degradation of NRF2 and increases the sensitivity of lung cancer cells to ferroptosis [32]. Studies have shown that members of the USP family, including USP9X and USP32, can mediate the deubiquitination of NRF2 to stabilize its expression, thereby regulating ferroptosis and influencing disease progression [33,34]. Crucially, USP11 can deubiquitinate and stabilize the expression of NRF2 and is involved in cell proliferation and ferroptosis [35]. However, there have been no reports on whether USP11 affects HCC ferroptosis and taxane sensitivity by regulating the deubiquitination of NRF2.
On the basis of the above analysis, in this study, we investigated the effects of the USP11/NRF2 molecular axis on ferroptosis and sensitivity to taxane drugs in HCC. We revealed that the USP11/NRF2 axis is an important pathway for regulating ferroptosis and chemotherapy resistance in HCC.

Materials and methods

Clinical sample collection
Surgical samples of tumor and paracancerous tissues were collected from 24 patients with HCC at our hospital. The participants in this research were not subjected to radiotherapy, chemotherapy, or any other related neoadjuvant treatment prior to their surgery. Prior to the collection of samples, all participants provided written informed consent. This work was approved by the ethics committee of our hospital.

Cell culture
The immortalized hepatocyte cell line THLE2 and the human HCC cell lines HepG2, SNU449, and Hep3B were obtained from the American Type Culture Collection (ATCC, USA), while Huh7 and HEK293T cells were obtained from the Shanghai Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were cultured in Dulbecco's modified Eagle's medium (DMEM, HyClone, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA) and 100 U/mL penicillin/streptomycin (Sigma, USA), followed by incubation at 37°C in a 5% CO2 incubator. The culture medium was replaced every 2 days, and cells from passages 3 to 6 were used for subsequent experiments.

Construction and treatment of stable cell lines
USP11- and NRF2-knockdown cell lines (sh-negative control (NC), sh-USP11, and sh-NRF2) and stable USP11- and NRF2-overexpressing cell lines (overexpressing (OE)-NC, OE-USP11, and OE-NRF2) were generated using psPAX2 and pMD2.G. In brief, lentiviral plasmids were introduced into HEK293T cells, after which the viral supernatant was collected after 72 hours. The virus suspension was mixed with 5 μg/mL polybrene (40804ES76, Yeasen, China) and then used to infect the target cells. After 48 hours, puromycin was added to screen for positive clones, and the transfection efficacy was assessed using western blotting. The sequences of the shRNAs used were as follows:sh-NC,

sense: 5′-TTCTCCGAACGTGTCACGT-3′, antisense: 5′-ACGTGACACGTTCGGAGAA-3′;

sh-USP11,

sense: 5′-GCTGGTTCCTTGTGGAGAA-3′, antisense: 5′-TTCTCCACAAGGAACCAGC-3′;

sh-NRF2,

sense: 5′-CTGTTGATCTGTTGCGCAA-3′, antisense: 5′-TTGCGCAACAGATCAACAG-3′.

To trigger ferroptosis, cells with either USP11 overexpression or USP11 knockdown were exposed to erastin (20 μM; HY-15763; MedChemExpress, USA), a ferroptosis inducer, for 24 hours. Dimethyl sulfoxide (DMSO)-treated cells served as the blank control group.
To investigate the therapeutic effects of taxanes on HCC cells, varying doses of paclitaxel (PTX; 0, 2.5, 5, 10, 20, 40 nM; 580555, Sigma‒Aldrich, USA), docetaxel (DTX; 0, 5, 10, 20, 40, 80 nM; HY-B0011, Sigma‒Aldrich, USA), and cabazitaxel (CTX; 0, 5, 10, 20, 40, 80 nM; SML2487, Sigma‒Aldrich, USA) were used to treat cells overexpressing USP11 or silencing USP11 for 48 hours.

Xenograft animal model construction
Male BALB/c nude mice (18–22 g; 5 weeks) were purchased from the Animal Experiment Center of Kunming Medical University. They were raised in a specific pathogen-free clean environment and acclimatized for one week before being randomly divided into 9 groups, with 10 mice in each group: the control group, sh-NC group, sh-USP11 group, PTX group, DTX group, CTX group, sh-USP11+PTX group, sh-USP11+DTX group, and sh-USP11+CTX group. Every group of mice received a subcutaneous injection containing 0.2 mL of approximately 3 × 106 HepG2 cells transfected with sh-NC or sh-USP11. The control group of mice received injections of only HepG2 cells. After one week, the nude mice in each group received intraperitoneal injections of 10 mg/kg PTX, DTX, or CTX every three days, resulting in two total injections. The tumor size was measured every 7 days after the injection of cells, and its volume was determined using the following equation: volume = length × width2 × 0.5. On the 28th day, the nude mice were euthanized, and tumor samples were collected for follow-up experiments.

Real-time quantitative polymerase chain reaction (RT‒qPCR)
Total RNA was isolated from clinical tissues using the TRIzol reagent (15596026, Invitrogen, USA). The first-strand cDNA synthesis kit (P4202, Genenode, China) was used to reverse transcribe RNA for cDNA synthesis. RT‒qPCR was performed using a SYBR Green real-time fluorescence quantitative PCR kit (SR1110, Solarbio, China) with β-actin as the internal reference. The results were calculated using the 2−ΔΔCt method. The primer sequences are shown in Table 1.

western blot analysis
The proteins were isolated from tissues and cells using radioimmune precipitation assay (RIPA) buffer (R0278; Sigma‒Aldrich, USA), which included 1% protease and phosphatase inhibitors. Protein levels were measured using a bicinchoninic acid (BCA) detection kit (P0012, Beyotime, China). The total proteins were separated through sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE). The separated proteins were transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, USA) and subsequently blocked with 5% skim milk at room temperature for 1.5 hours. The membranes were then incubated with the diluted primary antibody at 4°C overnight. The membrane was incubated with a secondary antibody (1:4000, ab97051, Abcam, UK) at room temperature for one hour, followed by development using an enhanced chemiluminescence (ECL) kit (34580, Thermo Fisher Scientific, USA). Ultimately, the bands were semiquantitatively examined using ImageJ software. The primary antibodies used included USP11 (1:1000, 10244-1-AP; Proteintech, USA), NRF2 (1:1000, PA5-27882; Thermo Fisher Scientific, USA), SLC7A11 (1:1000, ab307601; Abcam, UK), GPX4 (1:1000, ab125066; Abcam, UK), and β-actin (1:1000, ab8226; Abcam, UK).

Immunohistochemistry
Human HCC tissues or tumor samples from nude mice encased in paraffin were deparaffinized and rehydrated, followed by antigen recovery using 0.01 M citrate buffer (pH 6.0). The samples were subsequently incubated with USP11 (1:500, 10244-1-AP, Proteintech, USA), NRF2 (1:500, PA5-27882, Thermo Fisher Scientific, USA), Ki67 (1:500, ab15580, Abcam, UK), SLC7A11 (1:500, ab307601, Abcam, UK), and GPX4 (1:100, ab125066, Abcam, UK) antibodies at 4°C overnight and horseradish peroxidase (HRP)-linked goat anti-rabbit IgG (H+L) secondary antibodies (1:4000, ab205718, Abcam, UK) for one hour at room temperature. These samples were then treated with 3,3′-diaminobenzidine (DAB) chromogenic reagent for 10 minutes at room temperature and then examined and imaged using a microscope (NiKon, Japan).

Cell Counting Kit-8 (CCK-8) assay
The proliferation of the cells was assessed using a CCK-8 kit (C0037, Beyotime, China). In brief, HepG2 and SNU449 cells (2 × 103 cells per well) were cultured in 96-well plates, after which the cells in each category were processed. After treatment, 10 μL of CCK-8 reagent was added to each well, and the absorbance values at 450 nm were measured in each well using a microplate reader at 0, 24, and 48 hours.

Colony formation experiment
HepG2 and SNU449 cells were seeded in 6-well plates (1 × 103 cells/well), followed by a continuous cultivation period of 2 weeks. After the cells were washed with PBS, 2 mL of 4% paraformaldehyde was added to each well to fix the cells for 30 minutes. Next, the cells were subjected to a 15-minute staining process with a crystal violet solution (G1063, Solarbio, China). After rinsing with tap water, the stained cells were photographed for observation.

Scratch test
The HepG2 and SNU449 cells in the logarithmic growth phase were collected. After normal digestion and passage, the cells were inoculated into 6-well plates. Upon reaching 90% confluence, the cells were scratched with a pipette tip. Following a 24-hour cultivation period, the migration of the cells was observed under a microscope (Nikon, Japan) and photographed, and the migration rate was calculated.

Transwell assay
Cell invasion experiments were conducted using a Transwell chamber (Corning, USA). HepG2 and SNU449 cells were adjusted to a concentration of 1 × 105 cells/mL in serum-free DMEM. Two hundred microliters of cell suspension were added to the upper chamber of a Transwell plate coated with Matrigel. In the bottom section of a 24-well plate, 600 μL of DMEM supplemented with 10% FBS was added. Following a 24-hour cultivation period, the invading cells were fixed with 4% paraformaldehyde, stained with crystal violet (G1063, Solarbio, China), and examined using an inverted microscope (Olympus, Japan).

ROS fluorescence staining
The ROS levels were detected using a ROS detection kit (S0033S, Beyotime, China). The cells from each group were subjected to trypsin digestion, and their density was modified using cell culture medium. The cells were seeded in 6-well plates at a density of 2 × 105 cells per well and incubated in a 5% CO2 cell incubator for 24 hours. The treatment of the cells varied on the basis of their grouping. The cells were subsequently incubated with Hanks solution. Concurrently, 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) was mixed with Hanks solution to achieve a concentration of 10 μmol/L. Each sample was treated with a diluted DCFH-DA solution and incubated in a cell incubator in the dark for 20 minutes. To ensure that the cells could fully combine with the probes, the mixture was shaken evenly every 5 minutes. The cells were washed with Hanks solution three times and incubated once more, and the levels of ROS were measured using fluorescence microscopy.

Detection of glutathione (GSH), malondialdehyde (MDA), and Fe2+ levels
The levels of GSH and MDA in the cell lysates were detected using a GSH kit (A006-2-1; Nanjing Jiancheng Bioengineering Institute, China) and an MDA assay kit (A003-1-2; Nanjing Jiancheng Bioengineering Institute, China). An Fe2+ kit (MAK025, Sigma‒Aldrich, USA) was used to measure the levels of Fe2+ in animal tissues and cell lysates. According to the instructions supplied with the kit, the tumor tissues or cells were homogenized and centrifuged, and the supernatant was collected. Each working solution was subsequently added, and the levels of GSH, MDA and Fe2+ in the tissue or cells were detected.

Coimmunoprecipitation
HepG2 and SNU449 cells were collected and lysed using IP lysis buffer, and the supernatants were isolated through centrifugation. A mixture of protein A/G Sepharose (Santa Cruz Biotechnology, USA), anti-USP11 (1:300, 10244-1-AP, Proteintech, USA), and anti-NRF2 (1:500, PA5-27882, Thermo Fisher Scientific, USA) antibodies was prepared. Each IP mixture was preincubated at 4°C for one hour with mild agitation; these IPs were incubated at 4°C overnight with gradual shaking; subsequently, the beads were collected through centrifugation and rinsed three times with lysis buffer. Western blot analysis was conducted on the immunoprecipitates.

Protein stability and degradation analysis
Cycloheximide (CHX) inhibits the synthesis of proteins in eukaryotes. MG-132 is a proteasome inhibitor that inhibits protein degradation in a protease-dependent manner [36]. HepG2 and SNU449 cells that had been transfected with OE-NC, OE-USP11, sh-NC, or sh-USP11 were incubated with 50 µg/mL CHX (HY-12320, MedChemExpress, USA) for 0, 2, 4, or 6 hours, and total protein was extracted at each time point for western blot analysis. HepG2 and SNU449 cells were treated with 20 μM MG132 (HY-13259; MedChemExpress, USA) for 6 hours, and total protein was extracted for western blot analysis.

Determination of ubiquitination levels
MG132 was added to the cell culture medium for 6 hours, followed by the addition of immunoprecipitation lysis buffer supplemented with protease and phosphatase inhibitors for 30 minutes. Immunoprecipitation of the lysate was performed using anti-NRF2 (1:1000, PA5-27882, Thermo Fisher Scientific, USA), followed by an overnight rotation at 4°C. Ubiquitination of NRF2 was detected using an anti-ubiquitin antibody. The immunoprecipitated proteins were analyzed by western blotting.

Statistical analysis
All the experimental data in this study are expressed as the means ± standard deviations (means ± SDs). The data were analyzed and visualized using GraphPad Prism 8. A t test was used to compare two groups, one-way analysis of variance (ANOVA) was used for multiple group comparisons, and two-way ANOVA was used to compare pairs between groups. P<0.05 indicated that the difference was statistically significant.

Results

Upregulation of USP11 expression in HCC
To evaluate USP11 expression in HCC and normal tissues, we analyzed the expression of USP11 in 369 HCC samples and 50 adjacent tissues from The Cancer Genome Atlas (TCGA) database. The findings revealed an increase in the expression of USP11 in tumor tissues (Fig. 1A). The levels of USP11 mRNA and protein expression were subsequently detected in clinical tissue samples, revealing a notable increase in HCC tumor tissues (Fig. 1B-C). Further immunohistochemical experiments were conducted to detect the expression level of USP11, and the results revealed the same trend (Fig. 1D). These findings demonstrated that USP11 expression was upregulated in HCC.

USP11 is associated with HCC cell proliferation and migration
To explore the potential role of USP11 in HCC progression, we measured its expression in THLE2 cells and HCC cell lines (HepG2, SNU449, Huh7, and Hep3B). Compared with THLE2 cells, HCC cells showed higher expression of USP11 (Fig. 2A). Furthermore, HepG2 and SNU449 cells were selected for subsequent experimental procedures. Next, USP11 was overexpressed and knocked down in HCC cells, and the results revealed significantly increased expression of USP11 in the OE-USP11 group but significantly decreased expression in the sh-USP11 group (Fig. 2B-C). The proliferation, clonogenicity, migration and invasion abilities of the cells were subsequently examined. Compared with the control, USP11 overexpression significantly promoted the proliferation, colony formation, migration and invasion abilities of HepG2 and SNU449 cells, whereas USP11 knockdown had the opposite effect on HepG2 and SNU449 cells (Fig. 2D-K). These results indicated that overexpression of USP11 promoted the proliferation, migration and invasion of HCC cells in vitro.

USP11 inhibits erastin-induced ferroptosis in HCC cells
Studies have shown that inhibiting ferroptosis is a promising treatment method for suppressing the growth of liver tumors [37]. To investigate the role of USP11 in HCC cell ferroptosis, we triggered ferroptosis by introducing erastin in vitro. Findings from the CCK-8 test revealed that overexpression of USP11 increased the viability of HepG2 and SNU449 cells, whereas treatment with erastin inhibited cell viability. Additionally, overexpression of USP11 weakened the inhibitory effect of erastin, while knockdown of USP11 had the opposite effect (Fig. 3A-B). Indicators related to ferroptosis revealed that without erastin induction, there was a notable reduction in Fe2+, MDA, and ROS levels in HepG2 and SNU449 cells due to USP11 overexpression, whereas GSH levels increased. After the addition of erastin, there was a notable increase in the levels of Fe2+, MDA, and ROS and a reduction in that of GSH. While overexpression of USP11 weakened the effect of erastin, USP11 knockdown showed the opposite regulatory effect (Fig. 3C-H; Fig. S1A-B). Ultimately, expression of the ferroptosis-associated proteins SLC7A11 and GPX4 was detected. These findings indicated that overexpression of USP11 enhanced the protein levels of SLC7A11 and GPX4 in the absence of erastin induction. Under erastin induction conditions, overexpression of USP11 partially restored the expression of the SLC7A11 and GPX4 proteins that were inhibited by erastin, whereas knockdown of USP11 had the opposite effect (Fig. S1C-F). These findings suggested that overexpression of USP11 was capable of preventing erastin-induced ferroptosis in HCC cells.

USP11 is associated with HCC cell sensitivity to taxanes
Studies have demonstrated the frequent use of taxanes such as PTX in HCC chemotherapy, yet their clinical use is hindered by significant drug resistance, restricting their effectiveness [38,39]. To investigate the effect of USP11 on taxane sensitivity, we first transfected OE-USP11 and sh-USP11 into HepG2 and SNU449 cells, respectively, and then treated these cells with different concentrations of PTX, DTX and CTX. A CCK-8 assay was used to detect cell viability. Compared with the OE-NC group, USP11 overexpression reduced the sensitivity of HepG2 and SNU449 cells to PTX, with IC50 values of 24.27 nM and 15.47 nM, respectively. However, USP11-knockdown cells were more sensitive to PTX, with IC50 values of 3.493 nM and 4.844 nM, respectively. Similarly, overexpression of USP11 reduced the sensitivity of HepG2 and SNU449 cells to DTX and CTX, whereas USP11 knockdown had the opposite effect (Fig. 4A-F). The cells were subsequently treated with 5 nM PTX, 10 nM DTX, and 10 nM CTX, followed by a colony formation assay. The findings indicated that administering PTX, DTX, or CTX to cells inhibited cell proliferation. Overexpression of USP11 promoted the resistance of cells to PTX, DTX, and CTX. In contrast, reduction of USP11 led to the opposite trend (Fig. 4G-L). These results indicated that USP11 reduced the sensitivity of HCC cells to taxanes and promoted drug resistance in these cells.

USP11 stabilizes NRF2 through deubiquitination
NRF2, a crucial transcription factor, can activate regulatory genes involved in the ferroptosis process [40] and promote drug resistance in cancer treatment [41]. Consequently, we investigated the regulatory relationship between USP11 and NRF2. The results from RT‒qPCR and western blot analyses revealed a notable increase in NRF2 expression in HCC tumor tissues (Fig. 5A‒B), which were further verified through immunohistochemical studies (Fig. 5C). Additionally, we found a positive correlation between USP11 and NRF2 (Fig. 5D), suggesting a potential synergistic effect between the two proteins in the tumor environment. Moreover, NRF2 expression in cells further demonstrated that NRF2 was highly expressed in HepG2 and SNU449 cells (Fig. 5E). USP11 is a deubiquitinating enzyme, and ubiquitination is the main mechanism responsible for regulating the stability and expression of NRF2 [42]. USP11 protects NRF2 from proteasome degradation by removing the K48 (lysine residue)-linked polyubiquitin chain of NRF2 [43]. We speculated that USP11 exerted its effect by regulating the ubiquitination and degradation of NRF2. Coimmunoprecipitation results demonstrated an interaction between USP11 and NRF2 (Fig. 5F). Western blot analysis revealed that in HepG2 and SNU449 cells, overexpression of USP11 significantly promoted NRF2 expression, whereas reducing USP11 levels had the opposite effect (Fig. 5G-H). After the cells were treated with the proteasome inhibitor MG132, NRF2 accumulated moderately in cells in the OE-NC + MG132 group. Overexpression of USP11 further increased the protein level of NRF2, while reducing the USP11 level weakened the inhibitory effect of MG132 on the degradation of NRF2 protein (Fig. 5I-J). CHX is a protein synthesis inhibitor. The cells were treated with CHX to assess the effect of USP11 on the stability of NRF2. Compared with the OE-NC group, overexpression of USP11 following CHX addition increased the protein stability of NRF2, whereas reduced USP11 content led to a decrease in the protein level of NRF2 (Fig. 5K-L). The results of the ubiquitination assay revealed a reduction in NRF2 ubiquitination following the overexpression of USP11, while knocking down USP11 promoted the ubiquitination of NRF2 (Fig. 5M-N). In summary, overexpression of USP11 led to the stabilization of NRF2 via deubiquitination.

USP11 inhibits erastin-induced ferroptosis in HCC cells through NRF2
We further investigated whether USP11 could regulate HCC ferroptosis through NRF2. Consequently, overexpression or knockdown of NRF2 was performed. Overexpression significantly increased the level of NRF2, while knockdown effectively reduced NRF2 expression (Fig. 6A). Erastin was subsequently used to trigger ferroptosis in HepG2 and SNU449 cells. The findings from the CCK-8 assay revealed that without the addition of erastin, knockdown of NRF2 significantly weakened the promoting effect of USP11 overexpression on cell proliferation. Moreover, erastin-induced treatment led to a more pronounced inhibitory effect. Conversely, overexpression of NRF2 somewhat reversed the suppressive effect of USP11 knockdown during erastin induction, yet the cell survival rates remained lower than those without erastin induction (Fig. 6B-C). The indicators related to ferroptosis were then detected, showing that post-erastin induction, there was an elevation in Fe2+, MDA, and ROS levels and a reduction in GSH levels. Reducing NRF2 weakened the inhibitory effect of USP11 overexpression on ferroptosis. Overexpression of NRF2 had the opposite effect (Fig. 6D-I; Fig. S2A-B). The levels of SLC7A11 and GPX4 expression were ultimately detected using western blotting. Without erastin induction, reducing NRF2 levels decreased the promoting effect of USP11 overexpression on the expression of SLC7A11 and GPX4. After erastin induction, NRF2 knockdown further reduced the expression of SLC7A11 and GPX4. In contrast, overexpression of NRF2 had the opposite effect (Fig. S2C-D). Research has indicated that USP11 inhibits erastin-induced ferroptosis in HCC cells through NRF2.

USP11 inhibits the sensitivity of HCC cells to taxanes through NRF2
We further explored the influence of the USP11/NRF2 signaling pathway on the sensitivity of HCC cells to taxanes. HepG2 and SNU449 cells were treated with 5 nM PTX, 10 nM DTX or CTX. Proliferation of the cells was subsequently detected using the CCK-8 and colony formation assays. Treatment with PTX, DTX or CTX reduced cell viability and clonogenic ability. Moreover, knockdown of NRF2 further weakened the promoting effect of USP11 overexpression on cell proliferation during treatment with PTX, DTX or CTX. Conversely, overexpression of NRF2 effectively alleviated the inhibitory effect of USP11 knockdown on cell proliferation (Fig. 7A-F; Fig. S3A-F). These findings indicated that USP11 inhibited the sensitivity of HCC cells to taxanes via NRF2.

Knockdown of USP11 in vivo inhibits tumor growth and promotes ferroptosis and sensitivity to taxanes
Finally, we injected HepG2 cells subcutaneously into nude mice to establish a xenograft tumor model to verify the effects of USP11 on ferroptosis and sensitivity to taxanes. Tumor size assessments in nude mice revealed that, compared with the sh-NC group, USP11 knockdown inhibited the growth of tumors in nude mice (Fig. 8A-C). Immunohistochemical analysis showed that USP11 knockdown led to a decrease in the protein levels of Ki67, USP11, NRF2, SLC7A11, and GPX4 proteins in nude mice (Fig. 8D-I). Detection of Fe2+ levels revealed that knocking down USP11 could increase Fe2+ levels in nude mice (Fig. 8J). Taxanes were subsequently used as a treatment for the nude mice. Measurement of tumor size indicated that USP11 knockdown or treatment with taxanes inhibited tumor growth, with a more pronounced effect observed with the combination of USP11 knockdown and taxane treatment (Fig. S4A-C). Immunohistochemical analysis was used to detect the expression of Ki67 in the tumors. The results revealed that knockdown of USP11 or treatment with taxanes decreased Ki67 expression. When sh-USP11 was combined with taxanes, the decrease in Ki67 expression was more significant (Fig. S4D). These findings suggested that reducing USP11 levels inhibited tumor growth, promoted ferroptosis and enhanced sensitivity to taxanes.

Discussion
HCC is a common and highly aggressive malignant tumor [17]. With the increasing burden of HCC, there is an urgent need for new biomarkers to facilitate early diagnosis, optimize treatment, and improve prognosis [4]. Our research confirmed that USP11 was highly expressed in HCC and led to more aggressive behavior of HCC cells. The primary aim of our research was to explore the mechanism by which USP11 could regulate ferroptosis and sensitivity to taxanes in HCC. The findings indicated that USP11 could inhibit the ubiquitination and degradation of NRF2, increase the level of NRF2, and thereby inhibit erastin-induced ferroptosis in HCC cells and the sensitivity of these cells to taxanes.
USP11 is a cysteine protease that is located in the cell nucleus. USP11 is involved in the development of cancer [44,45]. USP11 is pivotal for multiple biological functions, including the control of cell growth, cancer progression, metastasis, resistance to cancer treatments, and brain hemorrhage [35,[46], [47], [48]]. Furthermore, elevated levels of USP11 in HCC [49] and melanoma [50] are intimately linked to unfavorable survival rates. Our research confirmed the notably high expression of USP11 in HCC. Furthermore, research indicates that blocking USP11 may impede the proliferation, movement, and longevity of gastric cancer cells [23]. Knockdown of USP11 promotes the growth, migration and invasion of renal cancer cells [51]. Our research revealed that overexpression of USP11 promoted the proliferation, migration and invasion abilities of HCC cells, whereas knockdown of USP11 suppressed these abilities. These findings highlighted the significant function of USP11 in the development of HCC. The function of USP11 may vary depending on the type of tumor, and this variation may be determined by tumor type-specific factors.
Ferroptosis is a type of iron-dependent programmed cell death mechanism characterized by iron accumulation and lipid peroxidation [52]. Research has shown that ferroptosis is closely related to the progression of HCC [53]. For example, Yang et al. reported that polyphyllin I inhibited HCC cell proliferation, invasion and metastasis by inducing ferroptosis through the NRF2/heme oxygenase-1 (HO-1)/GPX4 axis [54]. Another study indicated that activating transcription factor 4 (ATF4) reduces the progression of HCC by hindering ferroptosis [55]. Importantly, USP11 can also regulate ferroptosis and plays an important role in colorectal cancer [22] and lung adenocarcinoma [42]; however, whether USP11 influences HCC development by regulating ferroptosis is still unclear. Our research revealed that overexpression of USP11 weakened erastin-induced ferroptosis, resulting in decreased levels of Fe2+, MDA, and ROS in HepG2 and SNU449 cells and increased levels of GSH, SLC7A11, and GPX4. The opposite trend was observed in the knockdown group. Consequently, our research indicated that USP11 effectively inhibited erastin-induced ferroptosis in HCC cells.
Taxanes, including PTX, DTX, and CTX, function as inhibitors of microtubules, and these chemotherapeutic agents exhibit effective and wide-ranging antitumor properties [10]. Reports indicate that the use of cell cycle inhibitors such as PTX may prolong mouse survival by hindering the proliferation of HCC [11]. Earlier research has indicated that DTX hinders cell growth by causing HCC cell death and that reduced SET8 levels heighten the vulnerability of SMMC-7721 cells to DTX therapy [56]. Furthermore, CTX exhibits high toxicity toward HCC cell lines, varying with time and dosage, by triggering G2/M phase arrest and apoptosis in vitro, and it also markedly hinders the growth of HCC tumors in vivo [57]. Research indicates a connection between USP-related proteins and the responsiveness of certain medications, such as PTX, potentially assisting in decisions regarding appropriate drug therapies [58]. Our research confirmed that the taxanes PTX, DTX, and CTX were capable of reducing HCC cell viability and that USP11 could reduce the sensitivity of HCC cells to taxanes. Therefore, inhibiting USP11 could increase the sensitivity of HCC to chemotherapy.
NRF2 plays a crucial role in controlling antioxidant reactions [59]. Irregular expression of NRF2 frequently occurs in multiple cancers, including HCC, and is linked to the development and advancement of tumors [60]. Reports have indicated an increase in NRF2 levels in HCC tumor tissues, correlating its expression with malignancy and an unfavorable prognosis [26]. Furthermore, NRF2 is crucial for controlling ferroptosis [27]. Notably, USP11 has the ability to regulate cell proliferation and ferroptosis by deubiquitinating and stabilizing NRF2 [35]. Our research further revealed that USP11 could enhance NRF2 expression via deubiquitination. Subsequent research indicated that with erastin treatment, reducing NRF2 levels weakened the suppressive effect of USP11 overexpression on ferroptosis, whereas increasing NRF2 expression had the opposite effect. These findings revealed that USP11 inhibited erastin-induced ferroptosis by promoting NRF2 expression. Furthermore, excessive NRF2 expression suppresses cell death and results in resistance to chemotherapy in a range of cancers [61]. Research has further confirmed that USP11 inhibits the sensitivity of HCC cells to taxanes by promoting the expression of NRF2.
In conclusion, our results indicated that USP11 stabilized NRF2 through deubiquitination, thereby inhibiting HCC ferroptosis and sensitivity to taxanes, providing new insights into the role of USP11 in regulating HCC ferroptosis and tumor drug resistance and suggesting that targeted inhibition of USP11 might be a promising therapeutic strategy for HCC. Therefore, the use of USP11-specific inhibitors, such as mitoxantrone [62], may increase the sensitivity of HCC to taxanes. However, the response rate of mitoxantrone is relatively low. Mitoxantrone is commonly used in combination with other antitumor drugs for the treatment of tumors, and its therapeutic effect is superior to that of mitoxantrone alone [63]. We speculate that mitoxantrone combined with taxanes may synergistically enhance the anti-HCC effect. Further experiments are needed to verify this hypothesis.

Funding
This study was supported by the 10.13039/501100008871Yunnan Provincial Department of Science and Technology- Basic Research Joint Special Project of Kunming Medical University - General Project (202301AC070250); 10.13039/501100008871Yunnan Provincial Department of Science and Technology - Kunming Medical University Basic Research Joint Project (202401AY070001-146).

Ethics approval and consent to participate
All the authors confirm that informed consent was obtained from all the subjects. All the methods/studies were conducted in accordance with the Declaration of Helsinki. All animal experimental protocols were approved by the Animal Ethics Review Committee of Kunming Medical University (kmmu20240079), and the animal procedures adhered to the ARRIVE guidelines 2.0.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

CRediT authorship contribution statement
Shujia Kong: Writing – original draft, Methodology, Investigation, Data curation. Chen Zhao: Validation, Data curation, Conceptualization. Jiaxun Li: Visualization, Supervision, Software. Xin Pan: Visualization, Software, Formal analysis. Yanwen Li: Writing – review & editing, Resources, Project administration, Funding acquisition.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.