PSMD8 cooperates with USP14 to promote bladder cancer progression by inhibiting ferroptosis

Summary

The ubiquitin-proteasome system (UPS) is crucial for regulating protein stability and essential cellular functions, including cell survival and proliferation. Dysregulation of UPS components contributes directly to malignant tumor development. This study investigates the significance of the proteasome subunit PSMD8 in bladder cancer (BLCA) pathogenesis. We found that elevated PSMD8 expression in BLCA specimens is strongly associated with a worse patient prognosis. Mechanistic studies demonstrated that PSMD8 promotes BLCA cell proliferation, migration, and carcinogenesis. PSMD8 negatively regulates ferroptosis, but not apoptosis, by directly interacting with and stabilizing SLC7A11, a known ferroptosis suppressor. Furthermore, we illustrate that ubiquitin-specific peptidase 14 (USP14) collaborates with PSMD8 to enhance SLC7A11 protein abundance. Reduction of PSMD8, SLC7A11, or USP14 sensitizes BLCA cells to cisplatin. Our findings demonstrate that the PSMD8/USP14 axis stabilizes SLC7A11 to suppress ferroptosis, promoting malignant growth, which suggests that targeting SLC7A11 or USP14 may significantly benefit BLCA patients with PSMD8 overexpression.

Graphical abstract

Highlights

• •PSMD8 overexpression correlates with poor bladder cancer prognosis
• •PSMD8 promotes bladder cancer proliferation, metastasis, and invasion
• •PSMD8 collaborates with USP14 to stabilize SLC7A11, thereby suppressing ferroptosis
• •Inhibition of the PSMD8/SLC7A11/USP14 axis enhances cisplatin efficacy

Molecular biology; Cell biology; Cancer

Introduction

Approximately 70%–75% of BLCA patients are diagnosed as non–muscle-invasive bladder cancers (NMIBC), while the remaining cases are muscle-invasive bladder cancers (MIBC).¹ The majority of BLCA patients die within 2 years because the suboptimal therapeutic responses and tolerance.² Despite the rapid progress in therapy research, few drug targets have succeeded in diagnosis and treatment. The ubiquitin-proteasome system (UPS) is a complicated system consisting of ubiquitin, various ubiquitin enzymes, and the 26S proteasome complex, plays a pivotal role in modulating the majority of protein ubiquitination events in eukaryotic cells.^3,4,5 Homeostasis of UPS is critical for the stability of intracellular proteins and cellular activities.^6,7,8 For the past years, a wide range of studies have demonstrated that dysregulation of UPS members is involved in the development of cancers.^9,10,11 Foundational studies culminated in bortezomib, the first selective 20S proteasome inhibitor, for treating cancers like multiple myeloma.^12,13 Bortezomib’s clinical efficacy is suboptimal due to severe side effects and limited application in solid tumors. Identifying UPS-specific driver genes as cancer therapeutic targets is crucial.

As components of 19S regulatory complexes in 26S proteasome, proteasome 26S subunit, non-ATPases (PSMDs) play a critical role in regulating the deubiquitination and stability of proteins. Accumulating evidence indicates that aberrant expression of PSMDs contributes to the malignant growth, migration, and invasion of cancer cells. PSMD proteins exhibit diverse roles in cancer progression, with PSMD1 and PSMD3 promoting chronic myeloid leukemia cell survival and PSMD2 contributing to breast cancer development via p21 and p27 degradation.^14,15While PSMDs’ roles in BLCA are evident, with PSMD14 promoting tumorigenesis via GPX4 regulation.¹⁶ Many studies have explored bladder cancer prognosis-related targets and drugs based on bioinformatics, clinicopathological analyses, and cellular and molecular biology methodologies.^{17,18,19,20,21} Meanwhile, PSMD8’s function in BLCA development remains elusive, prompting an investigation into its role in bladder cancer.

Herein, this study illustrated the clinical relevance and function of PSMD8 during the progression and BLCA based on human BLCA samples, in vitro and in vivo experiments. The underlying mechanisms of PSMD8 were investigated by mass spectrum and co-immunoprecipitation assays. We demonstrated that PSMD8 overexpression contributed to the progression of BLCA through repression of SLC7A11-mediated ferroptosis.

Results

Overexpression of PSMD8 confers poor prognosis in patients with BLCA

BLCA tissues showed significantly higher PSMD8 expression compared to adjacent normal tissues in both paired and unpaired analyses of TCGA data (Figure 1A). To validate these findings, we immunohistochemistry staining PSMD8 in fourteen paired BLCA samples, confirming its overexpression in cancer tissues, the IHC scores of tumor tissues were significantly higher than those of paracancerous normal tissues (Figure 1B). Importantly, PSMD8 expression did not correlate with age, smoking history, or gender (Figure 1C; Table 1). To evaluate the link between PSMD8 and BLCA progression, we divided patients by stage and grade. PSMD8 levels were higher in higher grades (Figure 1D). Similarly, stage IV patients had the highest PSMD8 expression compared to stages III and II&I (Figure 1D). To assess the BLCA prognosis linked to PSMD8 expression, we analyzed TCGA and GEO databases. We compared survival rates between high and low PSMD8 groups. Analysis across datasets (TCGA-BLCA, GSE13507, GSE31684, and GSE32548) revealed a significant link between high PSMD8 and worse overall survival (Figure 1E). A survival curve (Figure 1F) confirmed this. Similarly, disease-specific survival analysis in TCGA indicated a poorer outlook for high PSMD8 patients (Figure 1G). Collectively, PSMD8 high expression is a predictive marker for the progressiveness and poor prognosis of BLCA patients.

Figure 1.

BLCA patients with overexpression of PSMD8 have progressive disease and a worse prognosis

(A) Relative expression of PSMD8 was evaluated in BLCA tissues and non-paired or paired normal tissues.

(B) Immunohistochemical staining of PSMD8 was performed in the cancer and normal tissues of BLCA

patients. The IHC score of PSMD8 was quantified. Scale bar, 100 μm.

(C) Relative expression of PSMD8 w in BLCA patients based on their characteristics, including age, smoking, and gender, was analyzed based on the TCGA database. The difference was not significant.

(D) Relative expression of PSMD8 was analyzed in BLCA patients with different grades (low grade vs. high grade) or stages (stages I&II vs. stage III vs. stage IV).

(E) Meta-analysis of HR value comparison in PSMD8 high and low expression groups based on four public BLCA datasets.

(F) Overall survival of BLCA patients was compared in PSMD8 low and high expression groups from TCGA (p = 0.008), GSE13507 (p = 0.035), GSE31684 (p = 0.061), and GSE32548 (p = 0.054) cohort.

(G) Disease-specific survival (DSS) of BLCA patients was compared in the PSMD8 low and high expression group from the TCGA cohort, p = 0.005. ∗p < 0.05.∗∗∗p < 0.001. Error bars represent the mean values ± SEM. Statistical analyses were performed using Student’s t tests for two groups and one-way ANOVA for multiple comparisons.

Table 1.

Clinical features of bladder cancer patients from the TCGA database

characteristics	Low expression of PSMD8	High expression of PSMD8	P valuea
n	206	206	–
Age, n (%)	–	–	0.842533036
≤ 70	115 (27.9%)	117 (28.4%)
>70	91 (22.1%)	89 (21.6%)	–
Smoker, n (%)	–	–	0.425782207
No	59 (14.8%)	50 (12.5%)	–
Yes	144 (36.1%)	146 (36.6%)
Gender, n (%)	–	–	0.654092208
Female	52 (12.6%)	56 (13.6%)	–
Male	154 (37.4%)	150 (36.4%)
Histologic grade, n (%)	–	–	0.000810487
Low grade	18 (4.4%)	3 (0.7%)	–
High grade	187 (45.7%)	201 (49.1%)
Pathologic stage, n (%)	–	–	0.915984127
Stage I&Stage II	66 (16.1%)	67 (16.3%)	–
Stage III&Stage IV	139 (33.9%)	138 (33.7%)

^aChi-square tests were employed to compare proportions between two groups.

PSMD8 overexpression contributes to the proliferation and migration of BLCA cells

To explore the role of PSMD8 on the proliferation and migration capacity of BLCA cells, we first determined the expression of PSMD8 in various BLCA cells. T24 cells had a relatively lower expression of PSMD8, while 5637 expressed a much higher level of PSMD8 (Figure 2A). Thus, we performed gain-of-function assays in T24 cells and loss-of-function experiments in 5637 cells to study the biological function of PSMD8. Knockdown and overexpression efficacy were demonstrated by RT-qPCR and western blot results (Figures 2B, 2C, and S1). CCK-8 and colony formation results showed that cell proliferation is diminished after PSMD8 knockdown, while the overexpression of PSMD8 promoted the proliferation and growth of BLCA cells (Figures 2D, 2E, and S2). EdU incorporation was suppressed by PSMD8 knockdown and was promoted by PSMD8 overexpression, suggesting that PSMD8 potentiates the DNA synthesis in BLCA cells (Figures 2F and and 2G). In addition, PSMD8 overexpression promoted the migration ability of BLCA cells (Figures 2H and and 2I). Taken together, PSMD8 could act as an oncogenic protein triggering the proliferation and migration of BLCA cells.

Figure 2.

PSMD8 functions as an oncogenic protein in BLCA

(A) The relative mRNA level of PSMD8 was checked by RT-qPCR in T24, 5637, and RT-4 cells.

(B and C) The mRNA and protein expression of PSMD8 was detected by RT-qPCR and immunoblotting assays in siCtrl, siPSMD8#1, and siPSMD8#2 5637 cells (B) and in Ctrl and PSMD8 overexpressed T24 cells (C). ∗∗p < 0.01.

(D and E) CCK-8 and colony formation were performed in BLCA cells with knockdown or overexpression of PSMD8. ∗p < 0.05. ∗∗p < 0.01.

(F–I) EdU staining analysis of DNA synthesis and Transwell detection of cell migration ability were conducted in BLCA cells with knockdown or overexpression of PSMD8.

(F and G) EdU staining analysis of DNA synthesis was conducted in BLCA cells with knockdown or overexpression of PSMD8, the scale bar in Figure 2F is 50 μm.

(H and I) Transwell detection of cell migration ability was conducted in BLCA cells with knockdown or overexpression of PSMD8, the scale bar in Figure 2H is 100 μm. ∗p < 0.05. ∗∗p < 0.01.

PSMD8 suppresses lipid peroxidation and ferroptosis in BLCA cells

Resisting cell death is a hallmark of cancer.²² To determine whether PSMD8 regulates cell death, we initially performed PI/Annexin V staining after PSMD8 knockdown and overexpression in BLCA cells. We found that PSMD8 downregulation dramatically enhanced the number of PI⁺/Annexin V⁻ cells, but had minimal effect on apoptosis (PI⁺/Annexin V⁺ cells). Consistent results were observed in T24 cells with PSMD8 overexpression (Figure 3A). These results indicate that PSMD8 does not regulate apoptosis, but has an obvious impact on other cell death types. Except for apoptosis, the well-established types of cell death include ferroptosis, necrosis, necroptosis, and pyroptosis.²³ A previous study used PI staining to detect ferroptosis,²⁴ suggesting that ferroptosis is PI-positive. Therefore, we suspected that PSMD8 might regulate ferroptosis in BLCA cells. To address this question, we treated the siCtrl, siPSMD8#1, and siPSMD8#2 transfected 5637 cells with inhibitors of different cell death types, including ferroptosis (ferrostatin-1, ferr-1) and apoptosis (Z-VAD-FMK, Z-VAD). Results showed that only ferr-1, but not other inhibitors, could reverse the cell death triggered by PSMD8 downregulation (Figure 3B). Additionally, ferroptosis-specific-inducer erastin more obviously potentiated cell death in 5637 cells with PSMD8 knockdown as compared to siCtrl cells (Figure 3C). The enhanced cell death could be reversed by ferr-1 but not by other cell death inhibitors (Figure 3C). Consistently, PSMD8 overexpression overcame the cell death rate induced by erastin, but not other cell death inducers (Figure 3D). Erastin-induced cell death could be reversed by ferr-1, but not other cell death inhibitors (Figure 3E). As ferroptosis depends on cellular peroxidation of fatty acid, we lastly measured lipid ROS and MDA levels. Results showed that PSMD8 knockdown promoted, while its overexpression suppressed lipid peroxidation in BLCA cells (Figures 3F and 3G). Thus, PSMD8 suppresses ferroptosis and lipid peroxidation in BLCA cells.

Figure 3.

PSMD8 inhibits lipid peroxidation and ferroptosis in BLCA cells

(A) BLCA cells with PSMD8 knockdown and overexpression were stained with PI and Annexin V and the percentage of Annexin V⁻/PI⁻, Annexin V⁺/PI⁻, Annexin V⁺/PI⁺, Annexin V⁻/PI⁺ was analyzed on the flow cytometry system.

(B) siCtrl, siPSMD8#1, and siPSMD8#2 5637 cells were treated with different cell death inhibitors, including ferrostatin-1 (ferr-1) and Z-VAD-FMK (Z-VAD). Viable cells were examined by AO/PI staining.

(C) siCtrl, siPSMD8#1, and siPSMD8#2 5637 cells were treated with DMSO or ferroptosis inducer erastin, in combination with cell death inhibitors, such as ferr-1, Z-VAD, and necrostatin-1 (necro-1). Viable cells were examined by AO/PI staining.

(D) Ctrl and PSMD8 T24 cells were treated with different cell death inducers, including erastin, apoptosis activator-2, and TNF-α. Viable cells were examined by AO/PI staining.

(E) Ctrl and PSMD8 T24 cells were treated with DMSO or erastin, in combination with cell death inhibitors, such as ferr-1, Z-VAD, and necro-1. Viable cells were examined by AO/PI staining.

(F and G) Lipid peroxidation was assessed by measuring cellular BODIPY and MDA after PSMD8 knockdown and overexpression. ∗p < 0.05. ∗∗p < 0.01. Error bars represent the mean values ± SEM. Statistical analyses were performed using Student’s t tests for two groups.

PSMD8 promotes the stability of SLC7A11 protein in BLCA cells and patients

To identify the substrate of PSMD8, we performed a co-immunoprecipitation (CO-IP) assay. The protein lysates of IgG and Flag groups were subjected to mass spectrum analysis. Hundreds of candidates were identified to interact with PSMD8. Top ten proteins included PSMD8, RPL3, RPL7A, SLC7A11, HSPA1B, H3C1, H3C15, H3-3A, PGK1, and HSP90AA1 (Figure 4A). Since SLC7A11 is an essential protein regulating ferroptosis, we then focused on whether PSMD8 regulated the stability of SLC7A11. PSMD-Flag was overexpressed in T24 cells and the cells were subjected to IP with IgG and Flag antibodies. The immunoprecipitates included both PSMD8 and SLC7A11. In turn, when SLC7A11-Flag was ectopically expressed in T24 cells, the immunoprecipitates contained both SLC7A11 and PSMD8 (Figure 4B). GST Pull-down experiments also confirmed direct binding between SLC7A11 and PSMD8 (Figure 4C). These results suggested that there was an interaction between PSMD8 and SLC7A11. To assess whether PSMD8 influences the protein stability of SLC7A11, we treated siCtrl, and siPSMD8#1 transfected 5637 cells with MG132, a proteasome inhibitor. The results showed that the SLC7A11 protein was dramatically reduced in siPSMD8 cells as compared with the siCtrl group (Figure 4D). By contrast, PSMD8 overexpression overcame the protein degradation when treated with cycloheximide (Figure 4E). In addition, Ctrl and PSMD8-overexpressed T24 cells were overexpressed with ubiquitin and treated with MG132. The cells were subjected to immunoprecipitation with SLC7A11 antibody and immunoblotting with ubiquitin. We demonstrated that PSMD8 suppressed the ubiquitination of SLC7A11 (Figure 4F). Lastly, we checked the protein levels of PSMD8 and SLC7A11 in BLCA patients and found that there was a positive correlation between them in normal and cancer tissues from BLCA patients (Figures 4G and 4H). Taken together, PSMD8 promotes the stability of SLC7A11 by decreasing the ubiquitination levels. See Figure S3 for the original image of the western blotting in Figure 4.

Figure 4.

PSMD8 promotes the stability of the SLC7A11 protein in BLCA cells and patients

(A) PSMD8/Flag was overexpressed in HEK293 cells. The cell lysates were coimmunoprecipitated with IgG or Flag antibody. The proteins were subjected to mass spectrum assay. The top ten proteins that might have interaction with PSMD8 were listed.

(B) PSMD8/Flag or SLC7A11/Flag was overexpressed in T24 cells. The cell lysates were subjected to immunoprecipitation with IgG or Flag antibody. Immunoblotting was performed with indicated antibodies.

(C) Purified GST-PSMD8 was incubated with HIS-SLC7A11, and immunoblotting was performed with indicated antibodies.

(D) siCtrl and siPSMD8#1 5637 cells were treated with MG132 for different hours. The cells were subjected to immunoblotting analysis of SLC7A11.

(E) Ctrl and PSMD8 T24 cells were treated with cycloheximide (Chx) for different hours. The cells were subjected to immunoblotting of SLC7A11.

(F) Ubiquitination was ectopically expressed in Ctrl and PSMD8 overexpressed T24 cells. Then the cells were incubated with MG132 (10 μM) for 12 h and subjected to immunoprecipitation with SLC7A11 and immunoblotting of ubiquitination. Cell lysates were also subjected to immunoblotting of PSMD8, SLC7A11, and GAPDH.

(G) Immunoblotting results of PSMD8 and SLC7A11 were checked in the cancer and normal tissues of BLCA patients.

(H) Spearman correlation between PSMD8 and SLC7A11 in BLCA patients. p = 0.0069.

PSMD8 contributes to BLCA cell migration and tumorigenesis through upregulation of SLC7A11

We investigated if SLC7A11 upregulation contributes to PSMD8’s tumor-promoting effects in BLCA. We overexpressed SLC7A11 in BLCA cells with reduced PSMD8 (siPSMD8) and observed restoration of cell proliferation, colony forming ability, cell migration ability under Transwell assay and a significant decrease in lipid peroxidation (Figures 5A, 5B, 5E–5J, and S4). Conversely, knocking down SLC7A11 in cells with high PSMD8 reversed their increased cell proliferation, colony formation, cell migration, and lowered lipid peroxidation levels (Figures 5C–5J).

Figure 5.

PSMD8 exhibits tumor-promoting function in vitro and in vivo through upregulation of SLC7A11

(A and B) 5637 cells transfected with siCtrl, siPSMD8, and siPSMD8 plus SLC7A11 overexpression were subjected to immunoblotting with SLC7A11, and CCK8 assay, and detection of intracellular MDA and GSH level. ∗p < 0.05. ∗∗p < 0.01.

(C and D) T24 cells transfected with Ctrl, PSMD8, and PSMD8 plus siSLC7A11 were subjected to immunoblotting with SLC7A11 and CCK8. ∗p < 0.05. ∗∗p < 0.01.

(E and F) Colony formation was performed in 5657 and T24 cells. ∗p < 0.05. ∗∗p < 0.01.

(G and H) Transwell detection of cell migration ability was conducted in 5657 and T24 cells, the scale bar in Figure 5G is 100 μm. ∗p < 0.05. ∗∗p < 0.01.

(I and J) The cells were subjected to the detection of intracellular MDA and GSH levels. ∗p < 0.05. ∗∗p < 0.01.

(K) Immunoblotting demonstration of PSMD8 and SLC7A11 expression in shCtrl, shPSMD8, and shPSMD8+SLC7A11 5637 cells before implanting the cells into the nude mice.

(L) The macroscopical images of the tumors were derived from nude mice which were implanted with 5×10⁶ shCtrl, shPSMD8, and shPSMD8+SLC7A11 5637 cells.

(M and N) The tumor growth curve and tumor weight were detected. ∗p < 0.05. ∗∗p < 0.01.

We investigated the PSMD8/SLC7A11 axis in vivo using nude mice. Mice implanted with cells containing shRNA targeting PSMD8 (shPSMD8) showed reduced tumor growth compared to controls. Interestingly, overexpression of SLC7A11 in these shPSMD8 cells restored their tumorigenic potential (Figures 5K–5N). These findings suggest PSMD8 promotes BLCA progression by upregulating and stabilizing SLC7A11, which enhances tumor cell migration and growth.

PSMD8 cooperates with USP14 to suppress ferroptosis

The above mass spectrum results also identified that PSMD8 might interact with ubiquitin-specific protease14 (USP14). Previous studies have shown that USP14 acts as an oncogene in various cancer types, including breast cancer, lung cancer, and hepatocellular carcinoma.^25,26,27 Therefore, it is meaningful to illustrate whether USP14 participates in PSMD8-triggering BLCA development. We then overexpressed PSMD8/Flag or USP14/Flag in BLCA cells and subjected them to immunoprecipitation and immunoblotting assays. The results showed that there was a direct interaction between PSMD8 and USP14 in T24 cells (Figure 6A). In addition, USP14 silencing downregulated the expression of SLC7A11, and opposite results were also found in USP14 overexpressing cells (Figure 6B). Rescue experiments also confirmed the interaction between PSMD8 and USP14 in cell proliferation (Figure S5). Furthermore, USP14 was overexpressed in BLCA tissues and there was a positive correlation between PSMD8 and USP14 mRNA levels in BLCA patients from the TCGA database (Figures 6C and 6E). High expression of USP14 could also predict the poor survival of the patients (Figure 6D). To address the function of USP14 in lipid peroxidation and ferroptosis, we knockdown USP14 in 5637 cells and overexpressed USP14 in T24 cells. The results showed that USP14 knockdown induced lipid peroxidation and opposite results were also found in USP14 overexpressing cells (Figure 6F). USP14 knockdown also promoted cell death of 5637 cells, which could be rescued by Ferr-1, but not Z-VAD (Figure 6G). By contrast, ectopic expression of USP14 overcame the cell death induced by erastin, but neither apoptosis activator 2 nor TNF-α (Figure 6G). Collectively, USP14 promotes BLCA cell survival by suppressing ferroptosis. See Figure S6 for the original image of the western blotting in Figure 6.

Figure 6.

PSMD8 collaborates with USP14 to suppress ferroptosis

(A) PSMD8/Flag or USP14/Flag was overexpressed in BLCA

cells. The cell lysates were subjected to immunoprecipitation with IgG or Flag antibody. Immunoblotting was performed with indicated antibodies. Purified GST-PSMD8 was incubated with HIS-USP14, and immunoblotting was performed with indicated antibodies.

(B) Immunoblotting analysis of USP14, PSMD8, and SLC7A11 in siCtrl, siUSP14#1, siUSP14#2 5637 cells and in Ctrl and USP14 overexpressed T24 cells.

(C) Immunoblotting analysis of USP14 in a total of 5 normal and 7 cancer samples from BLCA patients.

(D) Overall survival of BLCA patients was compared in USP14 low and high-expression groups. p = 0.052.

(E) The Spearman correlation between PSMD8 and USP14 in BLCA patients. R = 0.385. p < 0.001.

(F) BLCA cells with USP14 knockdown or overexpression were subjected to MDA and GSH measurement.

(G) siCtrl, siUSP14#1, and siUSP14#2 5637 cells were treated with different cell death inhibitors, including ferrostatin-1 (ferr-1) and Z-VAD-FMK (Z-VAD). Viable cells were examined by AO/PI staining. Ctrl and USP14 T24 cells were treated with different cell death inducers, including erastin, apoptosis activator-2, and TNF-α. Viable cells were examined by AO/PI staining. ∗p < 0.05. ∗∗p < 0.01. Error bars represent the mean values ± SEM. Statistical analyses were performed using Student’s t tests for two groups.

PSMD8/USP14/SLC7A11 axis regulates the sensitivity of cisplatin in BLCA cells

Lastly, we explored the significance of the PSMD8/USP14/SLC7A11 axis in the response of BLCA cells to cisplatin treatment. We checked the IC50 of cisplatin in 5637 and T24 cells and the results were presented in Figure 7A. Consistent with previous reports, cisplatin treatment promoted the lipid peroxidation of BLCA cells (Figure 7B). As expected, downregulation of PSMD8 or USP14 in 5637 cells significantly reduced the IC50 of cisplatin, and opposite findings were shown in T24 cells with overexpression of PSMD8 or USP14 (Figures 7C and 7D). Interestingly, the enhanced IC50 by PSMD8 overexpression could be reversed by knockdown of USP14 or SLC7A11 (Figure 7E). These results suggest that PSMD8 overexpression reduces cisplatin response via SLC7A11 stabilization and USP14 upregulation.

Figure 7.

PSMD8/USP14/SLC7A11 axis regulates the sensitivity of BLCA cells to the treatment of cisplatin

(A) IC50 of cisplatin was detected by CCK8 assay in 5637 and T24 cells.

(B) 5636 and T24 cells were treated with different concentrations of cisplatin and subjected to MDA detection.

(C and D) BLCA cells with knockdown or overexpression of PSMD8 or USP14 were treated with different concentrations of cisplatin and the IC50 was calculated.

(E) Ctrl, PSMD8, PSMD8+siUSP14, PSMD8+siSLC7A11 5637 cells were treated with different concentrations of cisplatin, and the IC50 was calculated. ∗p <0.05. ∗∗p <0.01. Error bars represent the mean values ± SEM. Statistical analyses were performed using Student’s t tests for two groups.

Discussion

The ubiquitination proteasome system is a large and complex system consisting of E1 activating enzymes, E2 ubiquitin-conjugating enzymes, E3 ligase, ubiquitination/deubiquitination enzymes, and proteasome members. Recent studies suggest that malfunctions in ubiquitin-proteasome system enzymes contribute to cancer. However, the role of proteasome subunits in cancer is less understood. PSMD8, a member of the proteasome 26S subunit, has been implicated in cancer progression.^28,29,30,31 Previous studies identified PSMD8 overexpression in ovarian cancer, linking it to aggressive tumor phenotypes.³² PSMD8 upregulation was also observed in BLCA patients,³³ whereas the precise function and underlying mechanisms of PSMD8 during the development of BLCA remain to be determined. Our study found high PSMD8 levels in BLCA correlated with worse patient outcomes and promoted cancer cell growth, migration, and tumor formation. Additionally, lowering PSMD8 increased sensitivity to cisplatin, a common BLCA treatment, while PSMD8 overexpression conferred resistance. These findings suggest PSMD8 acts as an oncogene in BLCA. This study represents the first comprehensive investigation to explore the clinicopathological characteristics and biological properties of PSMD8 in bladder cancer, and it involves the detailed mechanism of PSMD8 regulating protein stability in bladder cancer cells, on this basis, it extends to the effect of PSMD8 on the ferroptosis and cisplatin sensitivity of bladder cancer cells, which makes the study novel and unique. Based on this study, the screening of druggable targets and the exploration of the resistance mechanism of antitumor drugs can be carried out at a later stage.

Cancer development hinges on uncontrolled cell growth and evasion of cell death pathways like apoptosis.^22,34 Apoptosis, regulated by proteins like Bcl-2 and caspases, has been extensively studied, but its direct targeting has yielded limited success in cancer treatment.³⁵ This might be because apoptosis is specific to eliminating cancerous cells. Ferroptosis, another form of cell death, is driven by metabolic imbalances in lipid peroxidation.³⁶ This process is heavily influenced by the balance between reactive oxygen species (ROS) and glutathione (GSH), which are controlled by proteins like SLC7A11, SLC3A2, and GPX4.³⁷ This suggests that ferroptosis might be a more promising target for cancer therapy. SLC7A11 that imports cystine for glutathione production. High SLC7A11 levels suppress lipid peroxidation, a key driver of ferroptosis. Notably, SLC7A11 expression is subject to complex regulatory mechanisms. The tumor suppressor protein p53 can bind to SLC7A11 and limit its activity,²⁴ hindering tumor growth. Conversely, oncogenic proteins like RBMS1 and SOX2 can upregulate SLC7A11, promoting cancer stemness and growth.^38,39 These findings suggest that SLC7A11 is a critical player in cancer development, regulated by opposing forces within the cell. In this study, we demonstrated that PSMD8 contributed to BLCA progression by negatively regulating lipid peroxidation and ferroptosis. Mechanistically, PSMD8 interacted with and promoted the protein stability of SLC7A11, depending on its cooperation with USP14. Interestingly, USP14 also functionally synergistically promotes SLC7A11: similar to PSMD8, USP14 overexpression inhibits the pro-ferroptotic effects of Erastin (an inducer of ferroptosis/targeted inhibitor of SLC7A11), and USP14 also decreases the level of MDA, while increasing intracellular GSH content. We further proved that PSMD8 contributed to BLCA tumorigenesis through SLC7A11 since the knockdown of SLC7A11 abrogated PSMD8-mediated tumor growth in vitro and in vivo.

Cisplatin, a cornerstone chemotherapeutic agent for BLCA, is often rendered ineffective due to the development of drug resistance in some patients.^40,41 Elucidation of the underlying mechanisms driving cisplatin resistance is crucial to enhance the effectiveness and durability of this drug. Recent studies showed that cisplatin-induced cell death via the regulation of ferroptosis.^42,43,44 Therefore, the potential role of the PSMD8/USP14/SLC7A11 axis in modulating BLCA cell sensitivity to cisplatin warrants investigation. Here, we verified that cisplatin-induced lipid peroxidation and ferroptosis in BCLA cells. Importantly, PSMD8 overexpression reduced the sensitivity of BLCA cells to cisplatin treatment, which could be reversed by inhibition of USP14 or SLC7A11. Our results indicate that the PSMD8/USP14/SLC7A11 axis contributes to suppressed cisplatin response in BLCA.

In conclusion, our study provided the evidence that PSMD8 acted as a tumor-promoting protein in BLCA by inhibiting SLC7A11-mediated lipid peroxidation and ferroptosis. Mechanistically, PSMD8 directly interacted with SLC7A11 to enhance the stability of SLC7A11 at the protein level. The cooperation between USP14 and PSMD8 was also required for PSMD8 upregulation of SLC7A11. Targeting PSMD8, USP14, or SLC7A11 increased the sensitivity of BLCA cells to the treatment of cisplatin. Therefore, the PSMD8/USP14/SLC7A11 axis not only contributes to the malignant growth of BLCA cells but also modulates the clinical response of BLCA patients to cisplatin treatment.

Limitations of the study

Despite the identification of the molecular mechanism by which PSMD8 promotes bladder cancer progression through the stabilization of SLC7A11 and subsequent inhibition of ferroptosis, this study is subject to several limitations. Firstly, the analysis primarily relied on retrospective clinical samples with a limited sample size. Future validation of the prognostic significance of PSMD8 necessitates larger-scale, multi-center cohort studies. Secondly, the precise molecular basis underpinning the PSMD8-USP14 complex’s regulation of SLC7A11 protein stability remains incompletely understood. Specifically, the synergistic deubiquitination mechanism of these two proteins and the potential involvement of other cofactors warrant further investigation. Moreover, while this study primarily focused on the ferroptosis pathway, it is recognized that the ubiquitin-proteasome system (UPS) may influence tumor progression through multiple targets. Therefore, a systematic screening of other downstream effector molecules of PSMD8 is warranted in future research. Finally, although targeting SLC7A11/USP14 was observed to enhance cisplatin sensitivity, in vivo experimental validation is required to confirm these findings and to assess their translational medical value in subsequent investigations.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Yunjin Bai (baiyunjin@scu.edu.cn).

Materials availability

The unique materials described in this work are available from the lead contact upon request.

Data and code availability

• •Data: All publicly available datasets analyzed during this study, including those from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas Program (TCGA), have accession numbers listed in the key resources table. These datasets are available at their respective repositories (GEO: https://ncbi.nlm.nih.gov/geo/and TCGA, GDC: https://portal.gdc.cancer.gov/).
• •Code: This paper does not report original code.
• •All other requests: Any additional information required to reanalyze the data or materials reported in this paper is available from the lead contact upon request.

Acknowledgments

The study was supported by the National Nature Science Foundation of China (no.82203298), Natural Science Foundation of Sichuan Province (no. 2022NSFSC1518), Research Projects of Gansu Provincial Hospital (no. 24GSSYB-2) and Joint Research Foundation of Gansu Province (no. 24JRRA892), Sichuan Science and Technology Program (no. 2024YFFK0276).

The Graphical abstract was drawn in Microsoft Office PowerPoint software. The tumor image in the Graphical abstract was adapted from by Servier Medical Art (https://smart.servier.com/), licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).

Author contributions

Experiment design and review and editing, X.W. and Y.B.; data analysis, experiments and writing-original draft, X.W., B.L., and Y.T.; data collection and assistant cell experiments, X.W., B.L., J.W., J.R., R.H., W.C., and D.F.; assistant for animal experiments, X.W., Y.W., and T.G.; review and editing, J.W., X.W., H.L., P.H., and Y.B. All authors read and approved the final manuscript.

Declaration of interests

The authors have no conflict of interest.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti-PSMD8 antibody	Abcam	Cat# ab246883; RRID: AB_2861216
Recombinant Anti-xCT antibody	Abcam	Cat# ab307601; RRID: AB_30945700
Rabbit IgG, monoclonal [EPR25A] Isotype Control	Abcam	Cat# ab172730; RRID: AB_2687931
Anti-USP14/TGT antibody [EPR15943] - C-terminal	Abcam	Cat# ab192618
PSMD8 Antibody (A-1)	Santa Cruz	Cat# sc-398619
Beta Actin Monoclonal antibody	Proteintech	Cat# 66009-1-Ig; RRID: AB_2687938
GAPDH Monoclonal antibody	Proteintech	Cat# 60004-1-Ig; RRID: AB_2107436
GST Tag Monoclonal antibody	Proteintech	Cat# 66001-2-Ig; RRID: AB_2881488
6∗His, His-Tag Monoclonal antibody	Proteintech	Cat# 66005-1-Ig; RRID: AB_11232599
Vinculin Antibody	Cell Signaling Technology	Cat# 4650; RRID: AB_10559207
DYKDDDDK Tag (D6W5B) Rabbit mAb	Cell Signaling Technology	Cat# #14793; RRID: AB_2572291
Ubiquitin (E6K4Y) XP® Rabbit mAb	Cell Signaling Technology	Cat# #20326; RRID: AB_3064918
HRP Conjugated Goat anti-Mouse IgG polyclonal Antibody	HUABIO	Cat# HA1006; RRID: AB_2819167
HRP Conjugated Goat anti-Rabbit IgG polyclonal Antibody	HUABIO	Cat# HA1001; RRID: AB_2819166
Bacterial and virus strains
Lentiviral vector	HANBIO	N/A
BL21 Super Competent Cells	Beyotime	Cat# D1009S
Biological samples
BLCA tissues and adjacent non-tumorous tissues	West China Hospital, Sichuan University	N/A
Chemicals, peptides, and recombinant proteins
RPMI-1640 Medium	GIBCO	Cat# 11875093
McCoy’s 5A Medium	GIBCO	Cat#
Dulbecco’s Modified Eagle Medium	GIBCO	Cat#
Fetal Bovine Serum	GIBCO	Cat#
Penicillin-Streptomycin (10,000 U/mL)	GIBCO	Cat# 15140122
Ready to use immunohistochemical secondary antibody kit (anti rabbit&mouse)	Absin	Cat# abs996-5mL
Lipofectamine™ RNAiMAX Transfection Reagent	Invitrogen	Cat# 13778150
RIPA buffer	Beyotime	Cat# P0013B
Protease inhibitors	Beyotime	Cat# P1005
SuperSignal™ Western Blot Enhancer	Thermo Scientific	Cat# 46641
MG132	MCE	Cat# HY-13259
Cycloheximide	MCE	Cat# HY-12320
TRIzol reagent	Invitrogen	Cat# 15596026CN
ReverTra Ace™ qPCR RT Kit	TOYOBO	Cat# FSQ-301
Cell Counting Kit-8	Abmole	Cat# M4839
Crystal violet stain solution,1%	Solarbio	Cat# G1062
EdU Cell Proliferation Kit	Beyotime	Cat# C0078S
Annexin V-FITC/PI assay kit	4A Biotech	Cat# FXP018
Ferrostatin-1	Selleck	Cat# S7243
Z-VAD-FMK	Selleck	Cat# S7023
Necrostatin-1	Selleck	Cat# S8037
Erastin	Selleck	Cat# S7242
Apoptosis activator-2	Selleck	Cat# S2927
TNF-α	Sigma- Aldrich	Cat# H8916
AO/PI Double Staining	Absin	Cat# abs9727-100T
Lipid Peroxidation MDA Assay Kit	Beyotime	Cat# S0131S
BCA Protein Assay Kit	Beyotime	Cat# P0012
BODIPY 581/591 C11	MCE	Cat# HY-D1301
GSH-Glo™ Glutathione Assay	Promega	Cat# V6911
IPTG	Beyotime	Cat# ST098-1g
Critical commercial assays
SYBR® Green Real-time PCR Master Mix	TOYOBO	Cat# QPK-201
Pierce™ Co-Immunoprecipitation Kit	ThermoFisher Scientific	Cat# 26149
Pierce™ GST Protein Interaction Pull-Down Kit	ThermoFisher Scientific	Cat# 21546
Deposited data
TCGA-BLCA RNA-seq data	UCSC Xena	https://xenabrowser.net/
Gene expression microarray data	Kim WJ45 & Lee JS46	GEO: GSE13507
Gene expression microarray data	Riester M47,48	GEO: GSE31684
Gene expression microarray data	Lindgren D49	GEO: GSE32548
Experimental models: cell lines
HEK293T	Cell Resource Center, Institute of Biochemistry and Cell Biology at the Chinese Academy of Science (Shanghai, China)	Cat# GNHu44
5637	Cell Resource Center, Institute of Biochemistry and Cell Biology at the Chinese Academy of Science (Shanghai, China)	Cat# TCHu 1
T24	Cell Resource Center, Institute of Biochemistry and Cell Biology at the Chinese Academy of Science (Shanghai, China)	Cat# SCSP-536
RT4	Cell Resource Center, Institute of Biochemistry and Cell Biology at the Chinese Academy of Science (Shanghai, China)	Cat# TCHu226
Experimental models: organisms/strains
6-weeks-old Nude (BALB/c-nu) female mice	Beijing HFK Bioscience Co.,Ltd	N/A
Oligonucleotides
See Table S2
Recombinant DNA
pLVX-shRNA2-GFP-Puro-NC	Hanbio	N/A
pLVX-shRNA-PSMD8 -GFP-Puro	Hanbio	N/A
pLVX-EF1α- PSMD8-GFP-Puro	Hanbio	N/A
pLVX-EF1α- SLC7A11-GFP-Puro	Hanbio	N/A
pLVX-EF1α- USP14-GFP-Puro	Hanbio	N/A
pLVX-EF1α-GFP-Puro-Empty	Hanbio	N/A
pcDNA3.1- PSMD8-FLAG	This paper	N/A
pcDNA3.1- SLC7A11-FLAG	This paper	N/A
pcDNA3.1- USP14-FLAG	This paper	N/A
pcDNA3.1-FLAG	This paper	N/A
pcDNA3.1- SLC7A11-HIS	Syngenbio	N/A
pcDNA3.1- USP14-HIS	Syngenbio	N/A
pcDNA3.1-HIS	Syngenbio	N/A
pGEX-4T-1-PSMD8	This paper	N/A
pGEX-4T-1	This paper	N/A
Software and algorithms
R software	The R Foundation	https://www.r-project.org/
GraphPad Prism version 10.0.1	GraphPad Software	https://www.graphpad.com/
PowerPoint software	Microsoft	https://www.microsoft.com/zh-cn/microsoft-365/powerpoint
Excel software	Microsoft	https://www.microsoft.com/zh-cn/microsoft-365/excel
Other
Immobilon® -P PVDF Membrane	Millipore	Cat# IPFL00005

Experimental model and study participant details

Human BLCA samples

BLCA tumor and normal tissue samples were collected from patients (2019–2021) at West China Hospital, Sichuan University. The samples were immediately preserved with liquid nitrogen and 10% formalin for immunoblotting and immunohistochemical staining. All patients provided informed consent, and the study was ethically approved. The clinicopathological information of the BLCA patient included is found in Table S1. In bladder cancer tissues, the PSMD8 expression level showed no significant difference when comparing male and female patient groups.

Ethics statement

This study was approved by the Ethics Committee of West China Hospital, Sichuan University (Approval No.: 2024–604). All procedures involving human tissues and clinicopathologic information were conducted in accordance with the ethical standards of the “Measures of the Ethical Reviews of Life Science and Medical Research Involving Humans” and the Declaration of Helsinki. We collected 14 paired tumor and adjacent normal tissues from BLCA patients after obtaining written informed consent. Patient confidentiality and anonymity were strictly maintained throughout the study.

Informed consent

All included BLCA patients agreed to the study protocol and signed an informed consent.

Cell culture

The HEK293T and human BLCA cell lines (5637, T24, and RT-4) were authenticated and obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Authentication was performed by the supplier via short tandem repeat (STR) profiling. All cell lines were routinely tested and confirmed to be negative for mycoplasma contamination before use in experiments. The T24 and 5637cells were cultured in RPMI-1640 medium (GIBCO), the RT-4 cells were maintained in McCoy’s 5A medium (GIBCO), while HEK293T cells were cultivated in DMEM medium (GIBCO), supplied with 10% fetal bovine serum (FBS, GIBCO) and 1% penicillin-streptomycin solution (GIBCO).

Animal studies

The animal experiments involved in this study were approved by the Animal Ethics Committee of West China Hospital, Sichuan University (Approval No.: 20230303067), and strictly followed the “Guidelines for Ethical Review of Laboratory Animal Welfare” and the “3Rs” principle.

Tumor xenograft model

To investigate PSMD8 and SLC7A11’s function in BLCA progression, we used a xenograft model. We implanted BLCA cells (5637) into female BALB/c nude mice (6 weeks old). First, we reduced PSMD8 expression in these cells using shPSMD8 lentivirus. We then created a second group where PSMD8-reduced cells were further infected with lentivirus to overexpress SLC7A11. Stable cell lines were established using puromycin selection. We then injected these cells (shCtrl, shPSMD8, and shPSMD8+SLC7A11) subcutaneously into the mice. Tumor growth was monitored for 25 days by measuring tumor size weekly. Five weeks later, the mice were euthanatized and the tumor weight was checked. Tumor volume (cm³) = (length × width²)/2. All animal experiments were approved by the Animal Care Committee of Sichuan University.

Method details

Public databases data analysis

The expression of PSMD8 mRNA in BLCA tissues (n = 412) and normal tissues (n = 19), was analyzed on the TCGA database (http://cancergenome.nih.gov/). We also analyzed the GEO database, which included a total of 389 patients with BLCA, including the GSE13507 (165 patients),^45,46 GSE31684 (93 patients),^47,48 and GSE32548 (131 patients)⁴⁹ datasets. To assess the correlation between PSMD8 expression and patients’ stage and grade, the patients were divided into stages I&II, III, and IV, low and high grade. The overall survival (OS) and disease-specific survival (DSS) was also analyzed in BLCA patients with PSMD8 high and low expression. Meta-analysis of OS results from multiple datasets was performed. Lastly, the relationship between PSMD8 expression and patients’ other characteristics (age, smoking, and gender) was analyzed. The above analysis was completed using R (4.4.0) and GraphPad Prism (10.0.1) software.

Immunohistochemistry

Immunohistochemistry (IHC) staining was performed on 4-μm-thick paraffin-embedded sections according to the protocols as described previously.⁵⁰ Tissue sections were first deparaffinized and rehydrated. Three 5-min washes followed this in the PBST buffer. Antigen retrieval pretreatment was then performed, followed by another three 5-min washes in PBST. To eliminate endogenous peroxidase activity and reduce non-specific background staining, 100 μL of endogenous peroxidase blocking agent was added and incubated for 10 min. After three PBST washes, sections were blocked with 100 μL of 5%–10% normal goat serum for 20 min at room temperature. Following PBST washing, the primary antibody was added dropwise and incubated for 30 min at room temperature or 37°C. After three thorough washes with PBST, 100 μL of HRP enzyme-labeled secondary antibody polymer was added and incubated for 30 min in the dark. After PBST rinsing, DAB chromogenic solution was applied for 5 min at room temperature, and the reaction was terminated with tap water. Subsequently, 100 μL of hematoxylin was added dropwise for 3 min, followed by rinsing with tap water until the sections turned blue. Finally, slides were prepared by gradient dehydration, clarification, and mounting with neutral gum. All washing steps involved three 5-min dips in PBST buffer. The primary antibody against PSMD8 was from Abcam. We performed IHC staining for PSMD8 on 14 pairs of bladder cancer tissues and their paracancerous normal tissues, and then scored the results of the IHC staining using a semi-quantitative approach, IHC score = staining depth (0,1,2,3,4) ✕ staining area for each tissue sample (0,1,2,3,4), and the difference in IHC score between cancer and normal group next to cancer was statistically calculated.

Knockdown and overexpression of PSMD8, SLC7A11, or USP14

The knockdown assay was carried out by transfecting the cells with siRNAs targeting PSMD8, SLC7A11, and USP14, which were synthesized from RiboBio Co. Ltd (Guangzhou, China). Lipofectamine RNAiMAX reagent (Invitrogen) was applied for siRNA transfection. Two days later, the cells were harvested for subsequent experiments. For in vivo experiments, the knockdown and overexpression assay were performed by infecting the cells with lentivirus-mediated shRNA of PSMD8 and overexpressing lentivirus of PSMD8, SLC7A11, and USP14 (Hanbio). Stable cell lines were selected by treating the cells with puromycin. The sequences of siRNAs or shRNAs were listed in Table S2.

Immunoblotting

Total protein was extracted from BLCA tissues and cells using RIPA buffer supplemented with protease inhibitors (Beyotime). Following quantification and resuspension, equal amounts were loaded onto SDS-PAGE gels. PVDF membranes (Millipore) were activated and blocked before overnight incubation with primary antibodies (4°C). Secondary antibodies were then applied for 1 h (4°C). Protein levels were visualized using SuperSignal Western Blot Enhancer (Thermo Scientific) and the Tanon-5200 imaging system (Tanon). The antibody information was provided as below: PSMD8 (Santa Cruz), ubiquitin (Abcam), β-actin (Proteintech), and Vinculin (Cell Signaling Technology, CST). Protein expression levels were normalized to β-actin (Proteintech) or vinculin (CST). In addition, the correlation between the expression of PSMD8 and SLC7A11 in clinical samples was analyzed according to the gray value of western blotting.

Real-time quantitative PCR (RT-qPCR) analysis

Total RNA was extracted from BLCA cells by using TRIzol reagent (Invitrogen). After measuring the concentration and quality, the RNA was subjected to reverse transcription with ReverTra Ace qPCR RT Kit (TOYOBO) and qPCR with SYBR Green Real-time PCR Master Mix (TOYOBO) on an ABI Prism 7900 Sequence Detection System. Relative expression of PSMD8 in BLCA cells was calculated according to the 2^− ΔΔCt formula. The sequence of the primers was presented in Table S2 β-actin serves as the internal control.

Cell counting Kit-8 (CCK-8) and colony formation assay

The viability of BLCA cells was assessed by CCK-8 (Abmole) and monoclonal colony-forming ability was detected by colony formation assay. BLCA cells were transfected and incubated for 72 h. Cell viability was measured by CCK-8 assay (OD450) after adding the solution and incubating for 2 h at 37°C. Colony formation was assessed by seeding cells, fixing colonies with methanol, and staining with crystal violet (Solarbio). The images of colonies were collected by the Camera.

Transwell assay

A total of 5× 10⁴ transfected cells were seeded on the upper surface of the Transwell chamber in 24-well plates. One day later, the cells attached to the upper surface were removed and the cells that migrated to the lower surface were fixed with methanol and formalin solution, and stained with crystal violet (Solarbio). Migrated cells were photographed and counted under the microscope.

EdU staining

A 5-ethynyl-2′-deoxyuridine (EdU) cell proliferation kit (Beyotime) was used for the detection of cell proliferation levels. According to the instruction manual, after EdU incubation, cells were fixed, permeabilized, and underwent a click reaction to detect EdU incorporation (new DNA synthesis). Nuclei were stained with Hoechst for visualization under a fluorescence microscope.

Cell apoptosis and cell viability

Apoptosis was checked with the Annexin V-FITC/PI assay kit (4A Biotech), according to the manufacturer’s protocols. Treated cells were harvested with a binding buffer. After incubating with Annexin V-FITC and PI solution, cell apoptosis was detected on flow cytometry (Beckman).

BLCA cells were treated with the inducers or inhibitors, including ferrostatin-1 (3 μM, Selleck), Z-VAD-FMK (10 μg/mL, Selleck), necrostatin-1 (10 μg/mL, Selleck), erastin (7 μM, Selleck), apoptosis activator-2 (5 μM, Selleck), TNF-α (30 ng/mL, Sigma- Aldrich), or their combination. Then the cells were stained by AO/PI (Absin) and viable cells were detected on the Countstar system (BioTech).

Lipid peroxidation measurement

Lipid peroxidation was measured by using the MDA Assay Kit (Beyotime) in BLCA cells. In brief, the cells were collected and subjected to protein quantification by BCA assay kit (Beyotime) and MDA detection, according to the manufacturer’s protocols.

Cells were incubated with DMEM containing 5 μM of C-11 BODIPY dye (MCE) and the cells were returned to the tissue culture incubator for 30 min. Cells were then harvested and washed twice with PBS followed by re-suspending in 500 μl of PBS. ROS levels were analyzed in the flow cytometry system (Beckman).

GSH detection

The intracellular level of GSH was detected by using the GSH-Glo glutathione assay kit (Promega). Cells with different treatments were harvested and incubated with GSH-Glo Reagent and Luciferin Detection Reagent. Luminescence signals were detected using a Fluoroskan luminescence scanner (Thermo Fisher).

Co-immunoprecipitation and mass spectrum

The coding sequence of PSMD8, SLC7A11, or USP14 was cloned into pCDNA3.1 vectors, and tagged with a Flag. For mass spectrum, PSMD8/Flag was overexpressed in HEK293 cells. For other experiments, PSMD/Flag, SLC7A11/Flag, or USP14/Flag were ectopically expressed in BLCA cells. After overexpression, the cells were lysed in RIPA buffer (Beyotime) and then subjected to immunoprecipitation by using Pierce co-Immunoprecipitation kit according to the manufacturer’s instructions (ThermoFisher Scientific) and subsequent mass spectrum assays (Novogene). The antibody information was presented as below: Flag (CST), PSMD8 (Santa Cruz), SLC7A11 (Abcam), and USP14 (CST), IgG (Abcam), GAPDH (Proteintech), Ubiquitin (CST).

PSMD8 regulates the stability of SLC7A11 proteins

In this part of the study, siRNA knockdown and plasmid overexpression of PSMD8 were used to analyze the changes in the stability of PSMD8 to SLC7A11 proteins in HEK293T cells. The knockdown group was transfected with PSMD8 siRNA and then MG132 (25 μM, MCE) was added, and samples were collected at 0, 1, 2, and 4 h, respectively. The overexpression group was transfected with PSMD8 overexpression plasmid, and cycloheximide (Chx, 100 μg/ML, MCE) was added, and samples were taken at 0, 4, 8, and 12 h. The changes in SLC7A11 protein levels were detected by Western blot. The control group and the experimental group were set up in HEK293T cells, both groups were transfected with ubiquitin overexpression plasmid, MG132 (25 μM) was added to the two groups, PSMD8 overexpression plasmid was transfected in the experimental group, immunoprecipitation (IP) was performed with SLC7A11 antibody, and the level of ubiquitination modification was detected by Western blot, and the SLC7A11 ubiquitination differences between the two groups were compared.

GST pull-down assay

We utilized the Pierce GST Protein Interaction Pull-Down Kit (ThermoFisher Scientific) to perform a GST Pull-down assay to measure protein interactions. HIS-SLC7A11 and HIS-USP14 plasmid (Syngenbio) were transfected into HEK293T cells, then cells were lysed with GST lysate, and proteins were collected. The PSMD8 sequence was cloned into the pGEX-4T-1 vector to obtain the GST-PSMD8 plasmid. These plasmids were then transferred into E. coli BL21 (Beyotime) to express the proteins induced by Isopropyl β-D-1-thiogalactopyranoside (IPTG). GST-PSMD8 proteins were purified using GST-Tag purification resin.GST-PSMD8 proteins were arrested by glutathione Sepharose, and then the purified GST-PSMD8 was incubated with HIS-SLC7A11 or HIS-USP14 for 4 h at 4°C on a vertical mixer separately. Finally, the beads were washed and analyzed by Western blotting. The antibody information was presented as below: HIS (Proteintech), and GST(Proteintech).

Quantification and statistical analysis

Statistical data were presented as mean ± SEM. Student’s t test was applied to compare the statistical difference between the two groups. One-way ANOVA followed by Tukey’s post hoc test was used to analyze the statistical difference when more than two groups. The Chi-square test to compare proportions for categorical data. Spearman correlation analysis was used for correlation analysis of gene and protein expression levels. The difference was considered statistically significant when the p-value was less than 0.05.

Published: November 11, 2025