Abstract Hepatocellular carcinoma (HCC) is one of the most prevalent malignant tumors and the fourth leading cause of cancer-related death globally, which is characterized by complicated pathophysiology, high recurrence rate, and poor prognosis. Our previous study has demonstrated that disulfiram (DSF)/Cu could be repurposed for the treatment of HCC by inducing ferroptosis. However, the effectiveness of DSF/Cu may be compromised by compensatory mechanisms that weaken its sensitivity. The mechanisms underlying these compensatory responses are currently unknown. Herein, we found DSF/Cu induces endoplasmic reticulum stress with disrupted ER structures, increased Ca^2+ level and activated expression of ATF4. Further studies verified that DSF/Cu induces both ferroptosis and cuproptosis, accompanied by the depletion of GSH, elevation of lipid peroxides, and compensatory increase of xCT. Comparing ferroptosis and cuproptosis, it is interesting to note that GSH acts at the crossing point of the regulation network and therefore, we hypothesized that compensatory elevation of xCT may be a key aspect of the therapeutic target. Mechanically, knockdown of ATF4 facilitated the DSF/Cu-induced cell death and exacerbated the generation of lipid peroxides under the challenge of DSF/Cu. However, ATF4 knockdown was unable to block the compensatory elevation of xCT and the GSH reduction. Notably, we found that DSF/Cu induced the accumulation of ubiquitinated proteins, promoted the half-life of xCT protein, and dramatically dampened the ubiquitination–proteasome mediated degradation of xCT. Moreover, both pharmacologically and genetically suppressing xCT exacerbated DSF/Cu-induced cell death. In conclusion, the current work provides an in-depth study of the mechanism of DSF/Cu-induced cell death and describes a framework for the further understanding of the crosstalk between ferroptosis and cuproptosis. Inhibiting the compensatory increase of xCT renders HCC cells more susceptible to DSF/Cu, which may provide a promising synergistic strategy to sensitize tumor therapy and overcome drug resistance, as it activates different programmed cell death. Keywords: Hepatocellular carcinoma, DSF/Cu, Ferroptosis, Cuproptosis, xCT Graphical abstract Image 1 [55]Open in a new tab 1. Introduction Hepatocellular carcinoma (HCC) is an invasive type of cancer with increasing incidences and mortality rates, particularly in East Asia and Africa [[56]1]. The 5-year survival rate of HCC patients is still below 20 % leading to a poor outcome, which is still a global health challenge [[57]2]. The stage of the tumor greatly influences the prognosis. Early HCC patients' treatment includes liver transplantation, surgical resection, and locoregional therapies, but unfortunately, most patients are diagnosed at an advanced stage [[58]3]. Conventional cancer chemotherapeutics such as paclitaxel and cisplatin have limited effectiveness in HCC [[59]4]. Currently, the FDA-approved chemotherapeutics for the treatment of HCC involve tyrosine kinase inhibitors sorafenib and lenvatinib [[60]5,[61]6]. However, the acquired drug resistance of tyrosine kinase inhibitors exhibits certain limitations for the treatment of HCC [[62]7]. Therefore, it is urgent to discover novel and effective drugs for HCC treatment. The time and capital cost of new drug research confer it with great risks, thus drug repurposing is a fast and low-cost approach [[63]8]. Disulfiram (DSF) is an acetaldehyde dehydrogenase inhibitor approved by FDA in 1951 to treat chronic alcohol addiction. Previous research has reported that DSF could be metabolized into diethyldithiocarbamate (DTC) by the disulfide fragmentation at the center of symmetric structure, and then DTC combines with Cu to form the DTC-Cu complex (CuET) [[64]9,[65]10]. CuET is a Cu (II) diethyldithiocarbamate complex, serves as the primary component responsible for the anti-tumor effect of DSF/Cu [[66]11]. Further studies proved that DSF/Cu can induce the death of various tumor cells such as breast cancer cells [[67]12], non-small cells [[68]13], lung cancer cells [[69]14], esophageal cancer cells [[70]15], meningioma cells [[71]16], and mesothelioma cells [[72]17]. For HCC, DSF/Cu was found to inhibit metastasis by NF-κB and TGF-β pathways [[73]18]. Moreover, our previous study indicated that DSF/Cu could disrupt mitochondrial homeostasis, increase free iron, and exert an important role of ferroptosis in DSF/Cu-caused cell death [[74]19]. Ferroptosis is a new programmed cell death first discovered in 2012, indivisibly linked with lipid peroxide accumulation [[75]20]. The major molecular pathways of ferroptosis are iron metabolism, GPX4/GSH, NRF2, System Xc^-, mitochondrial metabolism, and FSP1-CoQ10 [[76][21], [77][22], [78][23]]. Several drugs, such as sorafenib, sunitinib, zalcitabine, dihydroartemisinin, and cetuximab, have been identified as implicated in the regulation of ferroptotic cell death and increased sensitivity of cancer cells to conventional therapy [[79][24], [80][25], [81][26], [82][27]]. In HCC, targeting ferroptosis is verified significantly improve the therapeutic effect of HCC by reducing the drug resistance of HCC cells to sorafenib [[83]28,[84]29]. Therefore, inducing ferroptosis is a promising therapeutic strategy. Endoplasmic reticulum (ER) is a key organelle that maintains intracellular homeostasis, participates in protein folding, assembly, transport, and maintains the homeostasis of intracellular calcium levels [[85]30]. Whereas, various exogenous factors, such as hypoxia and oxidative stress, easily cause unfolded protein response (UPR) in ER, and lead to ER stress [[86]31]. The UPR serves as an adaptive response and endogenous protective mechanism in response to cellular stimuli. It transmits stress signals from the ER to the nucleus, ultimately restoring ER homeostasis through mechanisms such as reducing protein synthesis, facilitating proper protein folding, and enhancing the degradation of misfolded or unfolded proteins [[87]32,[88]33]. Emerging evidence indicated ER stress response is activated synchronously during the initiation of ferroptosis [[89]34]. PERK-eIF2α-ATF4 signal pathway was capable of ameliorating ferroptosis by up-regulating HSPA5 and System Xc^- [[90]35,[91]36]. On the other hand, erastin-induced ER stress activates the PERK-eIF2α-ATF4-CHOP pathway, leading to an increase in PUMA expression, which participates in the regulation of ferroptosis [[92]37]. In our study, we aimed to explore the specific mechanism of DSF/Cu in HCC. It is found that DSF/Cu induces ER stress, as evidenced by the expanded ER, increased Ca^2+ level and activated expression of ATF4. Further studies verified that DSF/Cu induces both ferroptosis and cuproptosis, which is accompanied by the depletion of GSH, elevation of lipid peroxides, and compensatory elevation of xCT. Remarkably, our finding indicated that xCT expression was elevated through the inhibition of ubiquitination–proteasome mediated degradation rather than ATF4-dependent transactivation [[93]38]. Suppressing the compensatory increase of xCT facilitated the sensitivity of HCC to DSF/Cu-induced cell death. In conclusion, these results provide evidence that DSF/Cu may induce multiple forms of cell death simultaneously, and suppressing the elevated xCT strengthens HCC cells more vulnerable to DSF/Cu. 2. Materials and methods 2.1. Reagents and antibodies DSF and Cu were freshly mixed at a l:l ratio and used in the in vitro experiments. Enhanced Cell Counting Kit-8 (CCK-8 kit), Calcein/PI kit, EdU assay kit, Hoechst 33342, ER-Tracker Red and Fluo-4 AM were acquired from Beyotime (Shanghai, China). DCF-DA and C11-BODIPY (581/591) were purchased from Thermo Fisher Scientific (Waltham, MA). Disulfiram (DSF), CuCl[2], Monochlorobimane (MCB), Tauroursodeoxycholic acid (TUDCA), Deferoxamine (DFO), Tetrathiomolybdate (TTM), Bathocuproine disulfonic acid (BCS), Glutathione (GSH), Cycloheximide (CHX) and MG132 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Salicylazosulfapyridine (SASP), Erastin, Z-VAD-FMK, Necrosuifonamide (Necro), and Bafilomycin A1 (Baf-A1) were obtained from Selleck Chemicals (Houston, TX). The antibodies to GAPDH (ab8245), β-Actin (ab8226), p-eIF2α (ab32157), HSP70 (ab194360), NRF2 (ab62352), xCT (ab175186), FTH (ab75973), CSE (ab189916), GPX4 (ab125066), Glutaminase (ab156876), Glucose transporter 1 (GluT1) (ab115730), Recombinant Glutathione Reductase 1 (GRD1) (ab124995), Ki-67 (ab15580), ATP7B (ab124973), NFS1 (ab197253) and NDUFV2 (ab183717) were purchased from Abcam. eIF2α (11170-1-AP), ATF4 (10835-1-AP), CHOP (15204-1-AP), IRE1 (27528-1-AP) FDX1 (12592-1-AP) and Ubiquitination (23516-1-AP) were purchased from Proteintech. CBS (A11612) were purchased from ABClonal. GAPDH and β-Actin antibodies were used with 4000-fold dilution and other antibodies were used with 1000-fold dilution. 2.2. Cell culture SMMC-7721, Huh7 and 293T cells were preserved and passaged in our laboratory and cultured in a DMEM medium (Gibco, USA) containing 10 % fetal bovine serum (Gibco, USA) and 1 % penicillin-streptomycin in a cell culture incubator which was kept at the temperature of 37 °C and 5 % CO[2] with humidification. All experiments were performed using cells with more than 90 % viability in the logarithmic phase of growth. 2.3. CCK-8 assay CCK-8 was utilized to evaluate cell viability which consists of tetrazolium salts that react with dehydrogenase in cells to produce yellow formazan dye [[94]39]. Firstly, cells were plated on 96-well plates with a density of 8,000 cells per well, cultured in DMEM medium involving 10 % fetal bovine serum for the night. Then, the medium was replaced with a DMEM medium containing various drug concentrations to culture cells for 12 h. Before measuring at 450 nm using a microplate reader (Thermo Fisher), 10 μL CCK-8 was added to each well and incubated for 2 h at 37 °C. 2.4. Lentiviral packaging and transduction The expression of ATF4 and xCT were silenced using pLVX-shRNA Lentivector (Takara, Japan) according to our previous publication [[95]40]. The shRNA knockdown sequences are shown in [96]Table S1 xCT cDNA was obtained from Sino Biological (Beijing, China) and subcloned into a pLVX-IRES-Neo lentivirus vector (Takara, Japan). The recombinant lentiviral plasmids were verified by sequencing and co-transfected with PMD2.G and PSPAX2 in 293T cells for lentiviral packaging. Then the supernatants were collected and filtered through a 0.45 μm filter. SMMC-7721 and Huh7 cells were plated in a 24-well plate overnight and transfected with the corresponding lentivirus for 2 days. Next, 1 mg/mL G418 or 2 μg/mL puromycin was added for the selection of stably transfected cell lines. 2.5. EdU assay The 5-ethynyl-2-deoxyuridine (EdU) is a pyrimidine analog that can be integrated into DNA double-strand during DNA synthesis [[97]41]. EdU assay was performed to detect the cell proliferation capacity of SMMC-7721 and Huh7 cells treated with or without DSF/Cu. Images were captured by confocal microscopy. 2.6. Live/dead cell staining assay SMMC-7721 and Huh7 cells (8 × 10^4) were seeded in a confocal dish (NEST Biotechnology). DSF/Cu (0–1 μM) was added into the indicated wells for 12 h. Then, the cells were incubated with Calcein-AM and PI for the staining of live and dead cells. Nuclei were also counterstained with Hoechst 33342 (1 μg/mL) for 10 min. The fluorescence was observed under a confocal microscope (Leica TCS SP5, German). 2.7. Confocal microscopy assay The cells were plated in a confocal dish for 12 h. After the treatment of DSF/Cu, the cells were stained with different probes, including C11-BODIPY (4 μM), ER-Tracker Red (0.5 μM), Fluo-4 AM (1 μM), MCB (25 μM), for 30 min. Meanwhile, the cells were co-stained with Hoechst 33342 (1 μg/mL). Then the stained cells were washed with PBS three times and visualized under laser confocal microscopy. 2.8. Real-time PCR (RT-PCR) The total RNA from the treated HCC cells was extracted by the AG RNAex Pro Reagent (Accurate Biology). Afterward, RNA was reversed into cDNA utilizing Evo M-MLVRT Kit (Accurate Biology) according to the instructions. Gene expression was quantified by SYBR Green base RT-PCR and normalized to the level of the housekeeping gene. The primers of quantitative real-time PCR are shown in [98]Table S2. 2.9. Western blot assay After being treated with drugs, cells were lysed by scraping in RIPA buffer. Supernatants including total cellular proteins were obtained following centrifugation. Proteins were quantified by the BCA Protein Assay Kit (Boxbio Science & Technology, Beijing), and were separated by 10 % SDS-PAGE. Subsequently, proteins were transferred onto the PVDF membrane (Bio-Rad) and blocked with 5 % non-fat milk. The membranes were incubated with the corresponding primary antibody at 4 °C for at least 8 h. After being washed by TBST three times, the membranes were incubated with HRP-conjugated anti-rabbit or anti-mouse antibodies for 1 h. Finally, proteins were visualized by a Fdbio-Plco ECL kit (Fdbio science) in an enhanced chemiluminescence system (Bio-Rad, United States). ImageJ was used to quantify the band intensities. 2.10. CoIP assay CoIP assay was performed using the Capturem™ IP&COIP kit (TaKaRa, Japan) according to the manufacturer's instructions. Huh7 cells with xCT over-expression were lysed on ice by cell lysis buffer for 15 min and centrifuged at 4 °C for 10 min. The lysed proteins were incubated with the corresponding antibody overnight at 4 °C under constant mixing. After washings in the column, antibody-protein complexes were eluted with 40 μl of Elution Buffer and subjected to WB analysis. 2.11. Animal experiments The animal experiments were approved by the ethics committee of Zhejiang Provincial People's Hospital. Six-week-old BALB/c nude mice were purchased from the Shanghai Laboratory Animal Center and fed under SPF conditions. Huh7 cells were collected and resuspended in DMEM medium at a concentration of 2 × 10^7/mL. The right flank of each BALB/c nude mouse was injected subcutaneously with 0.2 mL of the Huh7 cell suspension. The animals were randomly divided into the Control group and DSF/Cu-treated group when the tumor reached 100–200 mm^3. In the DSF/Cu-treated group, CuCl[2] (0.06 mg/kg) was injected intramuscularly and 50 mg/kg DSF was taken intraperitoneally 2 h later. The mice's tumor volume was measured every 3 days. At the end of the experiments, the mice were euthanized by CO[2] inhalation. The blood samples were collected for safety assessment, and the tumors and organs were removed and collected. 2.12. Hematoxylin-eosin (H&E) and immunohistochemistry staining Tumor samples obtained from mice were fixed with 4 % paraformaldehyde (Leagene Biotechnology, China), embedded into the paraffin and cut into 4 μm thick sections. The tissue sections were performed with routine hematoxylin and eosin (H&E) staining. The immunohistochemistry assay was carried out after dewaxed, hydrated and antigen retrieval. The endogenous peroxidases were inactivated with 3 % peroxide and the sections were blocked with 5 % BSA. Ki-67 antibody was used with 100-fold dilution. Subsequently, the sections were incubated with normal rabbit IgG (Beyotime, China) for 30 min and added with horseradish for 15 min. The stained sections both were viewed and photographed under a microscope. 2.13. Statistical analysis All statistical calculations were performed using GraphPad Prism (version 9.0). Results are represented as mean ± SD. The differences between the two groups were performed by the student's t-test. Comparisons among multiple groups were analyzed by one-way ANOVA. Statistical significance was defined as * P < 0.05, **P < 0.01, ***P < 0.001. 3. Results 3.1. DSF/Cu inhibits cell viability and proliferation of HCC cells Our and other previous studies have shown that DSF/Cu exerts preferential toxicity toward HCC cells while sparing lower toxicity toward the non-malignant cells [[99]10,[100]42]. By comparing the toxic effects of DSF and CuCl[2] alone and in combination, the results revealed that the anticancer effect of DSF in HCC cells was copper-dependent ([101]Figs. S1A–B). To further verify the antitumor activity of DSF/Cu in HCC cells, the Calcein-AM/PI staining was utilized to calculate the percentage of live (Calcein-AM^+/PI^-) and dead (Calcein-AM^-/PI^+) cells. The results showed that the amounts of dead cells increased with the increase of DSF/Cu concentration, while the live cells characterized as calcein positive were reduced significantly, indicating that DSF/Cu accelerated cell death of HCC cells ([102]Fig. 1A and B). We then tested whether the administration of DSF/Cu could suppress the cell proliferation of HCC cells. As expected, the results of the EdU incorporation assay revealed that DSF/Cu significantly disrupted the growth of HCC cells, evidenced by the significant decrease of EdU-positive cells ([103]Fig. 1C–F). Taken together, these results showed that DSF/Cu could inhibit the viability and suppress the proliferation of HCC cells. Fig. 1. [104]Fig. 1 [105]Open in a new tab The cytotoxic effects of DSF/Cu on HCC cells (A, B) The cytotoxicity of SMMC-7721 and Huh7 were detected by Calcein/PI assay under treatment of DSF/Cu (0–1 μM) for 12 h. Green fluorescence is marked with living cells stained by Calcein, and red fluorescence is marked with dead cells stained by PI. Scale bar: 100 μm. (C–F) Representative images of SMMC-7721 and Huh7 stained for 5-ethynyl-2′-deoxyuridine (EdU) after treatment of DSF/Cu (0–1 μM) for 12 h. Scale bar: 100 μm. The quantitative histograms were displayed below the images. (Compared with control, *** indicated P < 0.001). (For interpretation of the references to colour in this figure legend,