
==== Front
Transl Oncol
Transl Oncol
Translational Oncology
1936-5233
Neoplasia Press

S1936-5233(24)00218-3
10.1016/j.tranon.2024.102091
102091
Original Research
Disulfidptosis-related gene expression reflects the prognosis of drug-resistant cancer patients and inhibition of MYH9 reverses sorafenib resistance
Zhang Kangnan a1
Zhu Zhenhua b1
Zhou Jingyi c1
Shi Min a
Wang Na WN1885@shtrhospital.com
a⁎
Yu Fudong fdyu@fudan.edu.cn
d⁎
Xu Ling xuling82@shsmu.edu.cn
a⁎
a Department of Gastroenterology, Shanghai Tongren Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, 200336, China
b Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, Department of Pathophysiology, Shanghai Jiao Tong University School of Medicine (SJTU-SM), Shanghai, 200001, China
c Department of Oncology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, 200080, China
d NHC Key Laboratory of Reproduction Regulation (Shanghai Institute of Planned Parenthood Research), Public Health School, Fudan University, Shanghai, 200030, China
⁎ Corresponding authors. WN1885@shtrhospital.comfdyu@fudan.edu.cnxuling82@shsmu.edu.cn
1 Zhang, Zhu and Zhou contributed equally to this work.

14 8 2024
11 2024
14 8 2024
49 10209118 10 2023
3 5 2024
11 8 2024
© 2024 The Authors. Published by Elsevier Inc.
2024

https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Highlights

• Disulfidptosis-related genes are significantly associated with somatic alteration patterns, immune infiltration, prognosis, and drug sensitivity in various tumors.

• Disulfidptosis-related genes are associated with drug sensitivity to various drugs, and MYH9, a disulfidptosis-related gene, plays a vital role in sorafenib resistance in hepatocellular carcinoma.

• Inhibiting the expression of MYH9 to a certain extent will increase the expression of SLC7A11. At the same time, it will cause cell shrinkage and F-actin contraction, leading to disulfidptosis-like changes, thereby increasing the drug sensitivity of sorafenib in liver cancer.

The onset of drug resistance in advanced cancer patients markedly diminishes their prognosis. Recently, disulfidptosis, a novel form of cell death, has been identified, triggered by excessive disulfide formation leading to cell shrinkage and F-actin contraction. Previous studies have identified 15 essential genes (FLNA, FLNB, MYH9, TLN1, ACTB, MYL6, MYH10, CAPZB, DSTN, IQGAP1, ACTN4, PDLIM1, CD2AP, INF2, SLC7A11) associated with disulfidptosis. This study sourced pan-cancer mRNA expression data from Xena to thoroughly evaluate the molecular and clinical characteristics of disulfidptosis-related genes. Through unsupervised clustering, mRNA expression data identified the expression levels of disulfidptosis-related genes and potential clusters related to this form of cell death. Kaplan-Meier survival curves illustrated the correlation between different clusters and overall survival. The findings reveal that high expression of disulfidptosis-related genes is linked to poor survival in liver cancer. The GDSC database was utilized to analyze the relationship between disulfidptosis-related genes and the AUC of 198 drugs. The results demonstrate that 12 disulfidptosis-related genes influence sorafenib resistance, as revealed by the intersection of differential genes related to sorafenib resistance from the GSE109211 dataset. Among them, the MYH9 gene was found to play a crucial role in both. Finally, experimental evidence confirmed that MYH9 mitigates sorafenib resistance in hepatocellular carcinoma through disulfidptosis-like changes. This study identifies disulfidptosis as a promising avenue for enhancing the sensitivity of tumor cells to drugs, offering new therapeutic perspectives for future research on disulfidptosis and drug resistance in cancer patients.

Graphical abstract

Disulfidptosis is the discovery of a new mode of cell death. There are not many studies on disulfidptosis in tumors. Our study found that there is also a correlation between disulfidptosis-related genes and drug sensitivity. Our experiments showed that inhibiting MYH9 expression in hepatocellular carcinoma attenuates sorafenib resistance. However, previous studies have focused on how MYH9 enhances sorafenib resistance by promoting stemness in hepatocellular carcinoma cells. Our study identified disulfidptosis as a breakthrough that could mitigate sorafenib resistance in cancer by enhancing disulfidptosis-like changes. This has important implications for developing personalized drugs for cancer treatment and overcoming drug resistance.Image, graphical abstract

Keywords

Hepatocellular carcinoma
Sorafenib resistant
Disulfidptosis
MYH9
Abbreviations

ACC adrenocortical carcinoma

ACTB actin beta

ACTN4 actinin alpha 4

ANOVA analysis of variance

AUC area under curve

BLCA bladder urothelial carcinoma

BRCA breast invasive carcinoma

CAPZB capping protein (actin filament) muscle Z-line, beta

CD2AP CD2 associated protein

CESC cervical squamous cell carcinoma and endocervical adenocarcinoma

CHOL cholangiocarcinoma

CNV copy number variation

COAD colon adenocarcinoma

DFI disease free interval

DLBC diffuse large B-cell lymphoma

DNAss tumor stemness based on DNA methylation

DSS disease special survival

DSTN destrin, actin depolymerizing factor

ESCA esophageal carcinoma

FLNA filamin A
==== Body
pmcIntroduction

The issue of drug resistance in advanced cancer patients is a significant challenge for patient survival [1]. Despite substantial advancements in cancer treatment over the past few decades with targeted therapies, immunotherapies, and chemotherapy, drug resistance frequently diminishes the effectiveness of these therapies [2,3]. The mechanisms of drug resistance are complex and diverse, necessitating further research to address this challenge from multiple perspectives in clinical settings [4,5].

Previously, we explored the relationship between ferroptosis and sorafenib resistance [6]. A new cell death mechanism, disulfidptosis, characterized by excessive disulfide formation in the actin cytoskeleton leading to the collapse of actin filaments (F-actin), has recently been identified [7]. Thus, there appears to be a certain correlation between disulfidptosis and drug sensitivity. Gan et al. found that SLC7A11 is overexpressed in several cancers, rendering tumor cells susceptible to cell death triggered by glucose deficiency [8]. Further research revealed that SLC7A11-mediated cystine uptake and metabolism led to intracellular disulfide accumulation and NADPH depletion under glucose starvation, ultimately resulting in cell death [9]. While this mode of cell death and ferroptosis both involve GSH and NADPH depletion, cystine deficiency promotes ferroptosis while inhibiting disulfidptosis, indicating that disulfidptosis is a distinct form of cell death [10,11]. Fifteen key disulfidptosis-related genes (FLNA, FLNB, MYH9, TLN1, ACTB, MYL6, MYH10, CAPZB, DSTN, IQGAP1, ACTN4, PDLIM1, CD2AP, INF2, SLC7A11) were identified by Liu et al. [12,13]. It was found that utilizing disulfidptosis signatures improved survival and drug sensitivity in bladder cancer patients [14]. This suggests that disulfidptosis could be highly effective for tumors with this metabolic profile.

Current research in liver cancer has found it to represent one of the cancers with thCurrent research has shown that liver cancer has one of the fastest-growing mortality rates, high recurrence rates, and low five-year survival rates over the past few decades [15]. Although early-stage hepatocellular carcinoma (HCC) can be treated with resection, liver transplantation, and other surgical therapies, over 50 % of patients are diagnosed at an advanced stage, with a 70 % recurrence rate within five years of treatment [16]. Clinically, sorafenib can prolong the median survival of patients with advanced HCC [17], but the majority of patients develop resistance to sorafenib, limiting its efficacy. Sorafenib resistance has become a significant barrier in the clinical management of advanced HCC patients.

Non-muscle myosin heavy chain IIA (MYH9) has been linked to liver cancer development in previous studies [3]. Additionally, MYH9 is dysregulated as an oncogene in many cancers [18,19]. Several studies have demonstrated that pathways involving MYH9 enhance liver cancer cell stemness, leading to sorafenib resistance [3,20]. However, research has primarily focused on how MYH9 promotes sorafenib resistance by enhancing liver cancer cell stemness. Whether MYH9 modulates sorafenib resistance in this cancer through disulfidptosis remains unexplored.

In this study, we identified the critical gene MYH9 and verified that inhibiting MYH9 could alleviate sorafenib resistance in hepatocellular carcinoma through disulfidptosis-like changes. This provides new insights and strategies for clinical immunotherapy planning and patient management.

Methods

Analysis of SNV and CNV of disulfidptosis-related genes in pan-cancer

SNV and CNV data for 33 tumor types from the TCGA database were downloaded via Xena Functional Genomics Explorer (https://xenabrowser.net/datapages/). The SNV data were analyzed for Missense_Mutation, Nonsense_Mutation, Nonstop_Mutation, Frame_Shift_Del, Splice_Site, Frame_Shift_Ins, and In_Frame_Del variants. Based on prior studies [21,22], CNV values of 2 were considered amplifications, while values of −2 were classified as deep deletions. We computed SNV and CNV ratios for each tumor type and visualized them using the R package "Complex-Heatmap" [23].

Analysis of mRNA for disulfidptosis-related genes in pan-cancer

Pan-cancer mRNA expression data from the TCGA dataset were obtained from Xena to analyze the differential expression of disulfidptosis-related genes in different tissues. The data were normalized to adjust for batch effects and transformed to log2(x + 1). The network of interactions among disulfidptosis-related genes was examined using the STRING database (https://string-db.org/). For the 17 tumor types with more than five pairs of tumor and normal samples, differential expression analysis was conducted. The fold change was calculated as the ratio of the mean expression of tumor samples to that of normal samples, and the p-value was determined using a t-test. Disulfidptosis-related clusters were identified using unsupervised clustering based on the PAM algorithm. Bootstrap runs were conducted 1000 times with 80 % of patient data, and cluster numbers from 2 to 9 were set to identify the optimal clusters [24].

Survival analysis of disulfidptosis-related genes

Clinical data were acquired from Xena. We integrated the expression data of disulfidptosis-related genes with clinical information to calculate risk ratios for OS, DFI, DSS, and PFI using Cox regression analyses, which enabled the classification of samples into high or low risk. For each tumor type, samples were split into two groups using the median value of each variable as a cutoff, and the p-value was calculated using a log-rank test. Kaplan-Meier survival curves for DSS and DFI were plotted for disulfidptosis-related genes significant to LIHC and analyzed using a log-rank test with the R package "survival".

Drug sensitivity analysis

CellMiner Data Analysis: Data from 60 cancer cell lines across 9 tumor types were acquired from the CellMiner database [25]. mRNA expression levels for disulfidptosis-related genes and Z-scores for cell sensitivity (GI50) were obtained via the CellMiner interface (https://discover.NCI.nih.gov/CellMiner/) and analyzed using Pearson correlation.

GDSC Data Analysis: Normalized gene expression data for 809 tumor cell lines and their response to 198 compounds were obtained from the GDSC database [26]. The disulfidptosis-related genes were analyzed to identify specific drug resistance.

Analysis of liver cancer microenvironment

Immunoscore was used to analyze the level of stromal cell infiltration in different tumors, and the stromal score was used to analyze the level of stromal cell infiltration in different tumors [27]. The analysis was based on TCGA-related gene expression profiling data (http://bioinformatics.mdanderson.org/estimate/) [27] . Estimated scores from the immunoscores were used to describe tumor purity. Spearman correlation tests were used to analyze the correlation between disulfidptosis-related genes expression and these scores. Six immune subtypes were defined to measure immune infiltration in the tumor environment [28]. The association between the expression of disulfidptosis-related genes in the tumor microenvironment and the type of immune infiltration was tested by ANOVA modeling using immune subtypes obtained from TCGA pan-cancer data. Tumor stem cell features extracted from transcriptomics, and epigenetics of TCGA tumor samples were used to measure stem cell-like features of tumor cells [29], and correlation analysis was performed using Spearman.

Cell lines and animals

Two cell lines, HCCLM3 and HepG2, were used in this study. They were obtained from the American Type Culture Collection (ATCC) and authenticated using STR fingerprinting. Mycoplasma-free cells were used.

For the xenograft experiment, 1 × 10^6 cells were injected subcutaneously into 7-week-old BALB/c nu/nu mice (Shanghai Institute of Physical Medicine, Chinese Academy of Sciences). Tumor growth was monitored weekly. After two weeks, the mice received 30 mg kg-1/day of Sorafenib orally for two weeks before being euthanized. Tumor volume was calculated using the formula V (cm^3) = 1/2 × Length × Width^2.

For mouse orthotopic tumor experiments, 1 × 10^6 cells were injected into the liver of 7-week-old BALB/c nu/nu mice. The mice were allowed to develop tumors naturally and then euthanized. Their livers were excised for examination. All experimental procedures adhered to the ethical guidelines of the Animal Committee of Tongren Hospital (approval number: A2023–119–01). The animal experiments complied with regulations set by the Ministry of Science and Technology of the People's Republic of China.

Culture of Sorafenib resistant cell lines

Cells were revived, and their status was adjusted before detecting the IC50 values over 72 h (SPSS software for statistics). HCCLM3 and HepG2 cells were cultured to approximately 60 % confluence before adding a specific concentration of Sorafenib (IC10) (S7397, Selleck, USA) to the medium. The cells were co-cultured for 48–72 h, followed by drug withdrawal, PBS washing, and replacement with normal medium. After cell recovery, cells were trypsinized and seeded into new culture flasks at appropriate cell densities. When they reached 50 %−60 % confluence, Sorafenib (IC10 concentration) was added for co-culture again. This co-cultivation process with the drug was repeated. If the cells could grow normally at this concentration, the concentration of Sorafenib (IC20, IC30, IC40, or higher) was increased during drug resistance testing, while repeatedly freezing, thawing, and observing the stability of the cell line.

Drug treatment during cell lines culture process

Cells were revived, and their status was adjusted before adding the MYH9 inhibitor Blebbistatin (MCE, HY-13,813, USA) to HCCLM3R and HepG2R at concentrations ranging from 1 μM to 10 μM. The cells were incubated with the 10 μM concentration for 10 days while adjusting for cell status.

Western blot analysis

Total proteins were extracted using a Whole Protein Extraction Kit (KGP200, KeyGen BioTECH, China). Protein concentrations were determined using a BCA protein assay kit (KGPBCA, KeyGen BioTECH, China). Equal amounts of protein samples were separated by SDS-PAGE and transferred to PVDF membranes. After blocking with 5 % skim milk in TBST for 2 h at room temperature, the membranes were incubated overnight at 4 °C with primary antibodies against MYL6 (1:1000, 68,142–1-Ig, Proteintech, USA), MYH9 (1:1000, 11,128–1-AP, Proteintech, USA), and GAPDH (1:1000, sc-365,062, Santa Cruz Biotechnology, USA). The membranes were then incubated with the secondary antibody for 1 hour at room temperature. Protein bands were visualized using Immobilon Western HRP Substrate (WBKLS0500, Millipore, USA) and imaged using a Tanon-5200 chemiluminescence detection system (Tanon Science, Shanghai, China). Protein densitometry was performed using ImageJ software.

Fluorescent staining of actin filaments and cellular membrane

Inoculate 2 × 10^5 cells in a Glass Bottom Cell Culture Dish (801,001, NEST, China), incubate for 12 h-24 h until the cell wall unfolds, and then wash with PBS. Then, it was fixed for 10 minutes at room temperature with 4 % paraformaldehyde in 1×PBS. For co-staining of actin filaments and cellular membrane, the abovementioned fixed cells were incubated at room temperature in the dark for 30 min with Deep red actin tracking stain dye (Thermo Fisher Scientific, A57245) and CellMask green plasma membrane stain dye (Thermo Fisher Scientific, C37608) [12]. The fluorescence images were captured using a confocal laser-scanning microscope (Leica, Germany).

Tissue microarray and immunohistochemistry

Tissue microarrays were constructed by Shanghai Zhuoli Biotechnology Co., Ltd (ZL-LVC1609, Zhuoli Biotechnology Co, Shanghai, China). Two tissue microarray (TMA) chips containing 80 pairs of tumors and matched adjacent tissues. These samples were attained under the hospital's approval.

The tissue samples were first deparaffinized for 1 hour and dehydrated in ethanol. Endogenous peroxidase was quenched by 3 % H2O2. The arrays were blocked using the 10 % normal goat serum for 20–30 min and incubated with primary antibody (MYH9, 11,128–1-AP, Proteintech; MYL6, 68,142–1-Ig, Proteintech; Ki67,HA721115,HUABIO; PCNA, abs158226, Absin) overnight at 4 °C. After washed in PBS, the tissue samples were incubated with HRP-conjugated secondary antibody for 1 hour at room temperature. The positive signals were then visualized by the DAB kit by reacting with the HRP substrate. After being washed in PBS and dehydrated in ethanol, the tissue slides were sealed and mounted under microscopy.

Gene set enrichment analysis (GSEA) analysis

The Gene Set Enrichment Analysis (GSEA) technique was used to analyze the enrichment of KEGG and GO pathways associated with MYH9, yielding pathways related to its expression.

Statistical analysis

Univariate Cox proportional risk regression models were used to test the correlation between gene expression and patient overall survival and progression-free intervals. Spearman correlation was used to test the correlation between gene expression at the mRNA and protein levels, as well as the correlation between gene expression and key transducers, stemness scores, stromal scores, immune scores, estimation scores, and drug sensitivity scores. Analysis of variance (ANOVA) was used to test correlations between gene expression, immune infiltration subtypes, and cancer subtypes. Plots were created using R and the packages ggplot2, pheatmap, corrplot, or surviminer. Statistical analyses (two-tailed Student's t-test) of bar graphs and scatter plots were performed using GraphPad Prism software. All statistical data are presented as the mean ± s.d. All experiments were repeated at least twice independently with similar results. P values of <0.05 were considered significant.

Results

The multifaceted role of the disulfidptosis-related genes in pan-cancer

Disulfidptosis, a newly identified form of cell death, has a strong association with various aspects of cancer. We analyzed the abnormal expression of disulfidptosis regulatory factors in multiple cancers (Fig. 1E). Moreover, disulfidptosis activity predicted overall survival (OS) in pan-cancer. Cox regression analysis of these regulatory factors across different tumor types revealed varying prognostic effects (Figure S1A). To clarify the impact of disulfidptosis on patient survival, tumors were grouped into two based on the median level of disulfidptosis regulatory factors, and significance was assessed using a log-rank test. As illustrated in Fig. 1A, these factors exhibited distinct prognostic effects in various tumor types. High expression in ESCA, GBM, HNSC, READ, SKCM, and STAD correlated with improved OS, while high expression in PGCP, CHOL, DLBC, ACC, MESO, UCS, and KICH was associated with poorer OS.Fig. 1 Survival analysis of disulfidptosis in pan-cancer.

A-D: Prognostic analysis of disulfidptosis regulatory factors for OS(A), PFI (B), DFI (C), and DSS (D). The size of the dots represents the significance of the impact of disulfidptosis regulatory factors on the survival rate for each cancer type, with p-values derived from the log-rank test. Only tumor types with significant results are displayed. Red dots represent high gene expression, and blue dots represent an association between high gene expression and better survival rates. E: The mRNA differences between tumor samples and adjacent normal samples. Red indicates high expression in tumors, and blue indicates low expression. F: Consistency matrix plot in LIHC with consistency clustering between samples when κ takes the optimal value. The heatmap shows 2 different clusters, with each row representing a disulfidptosis regulator and each column representing a patient. Red color indicates high expression and blue color indicates low expression. The Kaplan–Meier curve shows the difference in OS between Clusters 1 and 2. Cluster 1 is indicated in the orange line and Cluster 2 in green. *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.

Fig 1:

In addition to OS, we investigated the relationship between disulfidptosis and progression-free interval (PFI), disease-specific survival (DSS), and disease-free interval (DFI) in various cancer types. Across 19 cancer types, disulfidptosis regulatory factor levels and activity scores closely correlated with DFI, DSS, and PFI, depending on the specific tumor type (Fig. 1B-D). Unsupervised clustering of mRNA expression divided tumors into two clusters. Eight tumors within clusters 1 and 2 exhibited statistically significant differences in OS (P < 0.05) [13,14,30]. The expression of disulfidptosis-related genes impacted survival rates differently across tumors, with high expression improving survival in KIRC and LAML but posing a risk in LIHC, BLCA, LGG, MESO, KICH, and PAAD (Figure S2). For example, high disulfidptosis gene expression in LIHC was associated with worse survival (Fig. 1F-H).

To understand the genomic changes in the 15 disulfidptosis regulatory factors across tumors, we analyzed single nucleotide variations (SNVs) and copy number variations (CNVs) in all cancer samples from the Xena database (Figure S1B,C). We outlined the landscape of SNVs, CNVs, and somatic gene changes across 33 tumor types (Figure S1D-F). Next, we examined the relationship between disulfidptosis regulatory gene expression and tumor purity (evaluated by the matrix score) in tumors (Figure S1H) [28]. We also analyzed the correlation between the disulfidptosis gene and tumor stemness using the RNA stemness score (RNAss) based on mRNA expression and the DNA stemness score (DNAss) based on DNA methylation patterns (Figure S1I,J) [29,31]. Furthermore, we investigated the correlation between their expression and tumor infiltration using TCGA pan-cancer data and identified six immune infiltration subtypes: C1 (wound healing), C2 (INF-γ dominant), C3 (inflammation), C4 (lymphocyte-depleted), C5 (immune quiescent), and C6 (TGFB dominant) (Figure S1G) [28]. Among these subtypes, C3 and C5 exhibited significantly higher survival rates, while C4 and C6 had the lowest survival rates [28,32]. In conclusion, the unique expression patterns of the 15 disulfidptosis regulatory genes across different immune subtypes suggest that each gene's function depends on the immune subtype.

Therefore, disulfidptosis-related genes are not only closely linked to tumor prognosis but also to various aspects, including tumor somatic changes, immunity, and tumor stem cells.

Correlation of disulfidptosis-related genes with chemosensitivity of cancer cells

Our analysis indicates that the expression levels of disulfidptosis-related genes affect the survival of different tumors. A critical factor influencing survival in advanced cancer is drug resistance. We subsequently investigated the expression of these genes in a human cancer cell line (NCI-60) and systematically analyzed the correlation between their expression and the sensitivity of NCI-60 cell lines to over 200 chemotherapeutic agents (Supplementary File S1). Similar to observations in patient tumors, the expression levels of disulfidptosis-related genes exhibited significant heterogeneity across different cell lines. In the NCI-60 cell line data, drug sensitivity was measured by a Z score; a higher score indicates greater sensitivity to drug treatment [25]. A positive correlation between increased gene expression and drug sensitivity scores suggests that higher expression correlates with a favorable drug response.

Based on the correlation between gene expression and drug response Z-scores, we found that members of the disulfidptosis-related genes showed strong correlations with drug response. The top 16 analyses with an absolute correlation value of |r| ≥ 0.45 and p < 0.001 (Fig. 4C) were linked to genes such as ACTN4, MYH9, FLNB, MYL6, FLNA, and SLC7A11. We also observed that disulfidptosis-related genes were predominantly linked to increased drug resistance and associated with several drugs; for instance, FLNB was linked to resistance to 71 drugs, ACTN4 to 69, and MYH9 to 68 (P < 0.05). Some drugs showed both positive and negative correlations with disulfidptosis-related genes: MYL6, MYH9, and CAPZB were all negatively correlated with sensitivity to Cobimetinib (isomer 1), while IQGAP1 was positively correlated with it. Some drugs exhibited positive correlations between cellular sensitivity and disulfidptosis-related genes, such as Lenvatinib and Irofulven, while others like Imexon and Carfilzomib also demonstrated similar positive correlations (Fig. 2A and Supplementary File S1).Fig. 2 Association of disulfidptosis gene expression with drug sensitivity. A: Scatter plots to show the association between disulfidptosis gene expression and drug sensitivity (Z-score from CellMiner interface) tested with Pearson Correlation using NCI-60 cell line data. B: The GDSC database has statistical results of the relationship between drugs and disulfidptosis-related genes. The circle size represents the size of the P-value and the color depth indicates the size of the correlation.

Fig 2

To better understand whether disulfidptosis-related genes impact patients' responses to chemotherapy and targeted therapies, we integrated gene expression and drug sensitivity data from cancer cell lines available in the Genomics of Drug Sensitivity in Cancer (GDSC) database. We analyzed the correlation between disulfidptosis-related genes and the AUC of 198 drugs (Supplementary File S2). Fig. 2B presents a correlation plot of disulfidptosis-related genes across pan-cancer with highly relevant drugs in the GDSC database. Up to 13 disulfidptosis-related genes were found to be associated with vorinostat sensitivity, while sorafenib, mirin, and oxaliplatin followed closely, each showing associations with 12 disulfidptosis-related genes for their drug sensitivity.

These findings suggest that the selection of cancer patients with high expression of these genes could guide treatment choices and dosage through drug cell sensitivities.

Disconfidptosis-related gene MYH9 is associated with sorafenib drug sensitivity in hepatocellular carcinoma

Previous unsupervised consistent clustering of messenger ribonucleic acid (mRNA) regulator expression in hepatocellular carcinoma classified samples into two clusters. Survival analysis of these two clusters revealed that the cluster with high expression of disulfidptosis-related genes was associated with poorer survival. Chemotherapy drugs play a significant role in the survival of advanced cancer patients; however, drug resistance is a common challenge in these patients. Our analysis suggests that sorafenib is associated with 12 disulfidptosis-related genes (Fig. 2B).

Sorafenib is widely recognized as a first-line treatment for patients with advanced hepatocellular carcinoma [33,34]. However, resistance to sorafenib can pose a challenge [35]. To investigate the potential link between disulfidptosis and drug resistance, we analyzed sorafenib's impact on liver cancer (LIHC) sensitivity.

Using the GSE109211 dataset from the Gene Expression Omnibus (GEO), we examined the differences in sorafenib resistance between responders (n = 21) and non-responders (n = 46). A volcano plot (Fig. 3A) was plotted based on P < 0.05 and |log2FC| ≥ 1. Differentially expressed genes in GSE109211 overlapped with the 12 genes related to sorafenib resistance in disulfidptosis (Fig. 2B). Among these genes, MYH9, MYH10, and MYL6 were most strongly associated with disulfidptosis and sorafenib (Fig. 3B). The expression levels of these genes in the sorafenib-resistant (non-responder) and sensitive (responder) groups from the GSE109211 dataset [36,37] are shown in Fig. 3F. MYH9, MYH10, and MYL6 were highly expressed in the sorafenib-resistant group (P < 0.001).Fig. 3 Association of disulfidptosis gene expression with sorafenib resistance in LIHC. A: GSE109211 Analysis of differences between Sorafenib-treated responder and non-responder in the dataset, volcano plots based on P < 0.05, |log2FC|≥1. B: Wayne plot of the intersection of sorafenib resistance differential genes from the GSE109211 dataset and disulfidptosis-related genes related to sorafenib drug sensitivity (P < 0.05) from the GDSC database. C: AUC correlation plot of MYH9 with sorafenib in liver cancer cell lines from the GDSC database. D: AUC correlation plot of MYL6 with sorafenib in liver cancer cell lines from the GDSC database. E: AUC correlation plot of SLC7A11 with sorafenib in liver cancer cell lines from the GDSC database. F: Expression levels of MYH9, MYL6 and SLC7A11 in sorafenib-resistant (non-responder) and sensitive groups (responder) in the GSE109211 dataset. *P<0.05, **P<0.01, ***P<0.001, ns, not significant.

Fig 3

Additionally, we analyzed 15 disulfidptosis-related genes for sorafenib resistance in liver cancer cell lines using data from 14 liver cancer cell lines in GDSC2 (Figure S3). Although MYH10′s significance was not substantial (P = 0.085), MYH9 (Fig. 3C) and MYL6 (Fig. 3D) showed significant correlations with sorafenib AUC in hepatocellular carcinoma, with MYH9 being more significant (P = 0.004) compared to MYL6 (P = 0.047), warranting further exploration.

Inhibition of MYH9 expression increases sorafenib drug sensitivity in hepatocellular carcinoma by disulfidptosis-like changes

Previous analyses have revealed that elevated expression of disulfidptosis-related genes in certain tumors correlates with drug resistance (e.g., sorafenib) and poor prognosis (e.g., hepatocellular carcinoma). To confirm the role of disulfidptosis in tumor cell drug sensitivity, we experimentally validated MYH9, which was identified through our analyses. MYH9 exhibited a stronger correlation and a more significant p-value compared to MYL6 regarding sorafenib resistance in hepatocellular carcinoma. Therefore, we focused on MYH9 for further exploration, briefly validating MYL6.

We established sorafenib-resistant cell lines, HCCLM3R and HepG2R, derived from HCCLM3 and HepG2 cell lines (details of the construction process are provided in Supplementary File S3). Western blot assays confirmed that MYH9 expression in HCCLM3R and HepG2R cells was higher than in normal HCCLM3 and HepG2 cell lines, which aligns with our previous analysis (Fig. 3F). This increase in MYH9 expression was associated with sorafenib resistance (Fig. 6A). Comparable results were observed for MYL6. We then treated these cells with Blebbistatin, an MYH9 inhibitor [[38], [39], [40]], and generated two cell lines, HCCLM3R+Bl and HepG2R+Bl, which exhibited reduced MYH9 expression, as confirmed by Western blot (Fig. 6B).

Recent research emphasizes that elevated SLC7A11 expression is a crucial characteristic of disulfidptosis, leading to cysteine accumulation, disulfide stress, glucose depletion, blockage of the PPP pathway, reduced NADPH production, and aberrant disulfide bonds in actin cytoskeletal proteins, resulting in cell contraction and death [7,41]. Our next step is to experimentally validate the relationship between MYH9, disulfidptosis, and sorafenib in liver cancer.

Initially, we performed CellMask and F-actin immunofluorescence staining (Fig. 4) and observed that various cells demonstrated different degrees of disulfidptosis-like changes following sorafenib treatment [12,42]. In the absence of sorafenib, HCCLM3, HepG2, HCCLM3R, HepG2R, HCCLM3R+Bl, and HepG2R+Bl cells exhibited normal cell membranes and cytoskeletons. HCCLM3R and HepG2R cell membranes and cytoskeletons were not significantly altered after 7 h of stimulation with a low concentration of 2 µM sorafenib. HCCLM3, HepG2, HCCLM3R+Bl, and HepG2R+Bl cells displayed varying degrees of cell shrinkage and F-actin contraction, which were more pronounced in HCCLM3 and HepG2 than in HCCLM3R+Bl and HepG2R+Bl, but the cells remained viable. When stimulated with a higher concentration of 5 µM sorafenib for 7 h, the HCCLM3R and HepG2R cytoskeleton showed only a slight contraction, and the cells were still viable. The cytoskeleton of HCCLM3 and HepG2 cells disintegrated, and cell death occurred, while the cytoskeleton of HCCLM3R+Bl and HepG2R+Bl cells exhibited significant cytoskeletal contraction and became disorganized, compared to HCCLM3R and HepG2R.Fig. 4 F-actin contraction during the action of sorafenib. A-F: Fluorescent staining of F-actin with phalloidin in cells cultured in normal complete culture medium, normal complete culture medium with 2uM sorafenib, and normal complete culture medium with 5uM sorafenib for 7 h, respectively.

Fig 4

Secondly, SLC7A11 plays a pivotal role in disulfidptosis. Western blot experiments exhibited elevated SLC7A11 expression in MYH9 inhibitor-treated cells (HCCLM3R+Bl and HepG2R+Bl), indicating a higher propensity for disulfidptosis in these cells (Fig. 6B). Immunofluorescence staining further confirmed significant cytoskeletal constriction and disulfidptosis-like alterations in these cells. Analyzing the GSE109211 dataset highlighted reduced SLC7A11 expression in drug-resistant samples (Fig. 3F). Data from the GDSC database illustrated a negative correlation between SLC7A11 and AUC, suggesting that drug-resistant cell lines are less susceptible to disulfidptosis (Fig. 3E). Experimental validation demonstrated that SLC7A11 content was higher in HCCLM3 and HepG2 than in HCCLM3R and HepG2R cell lines (Fig. 6A), indicating a lower likelihood of disulfidptosis in sorafenib-resistant cell lines. Inhibiting MYH9 partially mitigated sorafenib resistance in liver cancer through disulfidptosis-like changes.

Thirdly, tissue microarray analysis (Fig. 5A,F) revealed significantly elevated MYH9 expression in hepatocellular carcinoma samples compared to adjacent normal tissues (Fig. 5B,C), consistent with previous research [3]. A similar pattern was observed for MYL6 (Fig. 5G,H). Based on the clinical information from the tissue microarray (Supplementary File S4, S5), we classified the samples as resistant to sorafenib based on their recurrence after treatment. Analysis indicated that MYH9 expression was higher in the sorafenib-resistant group compared to the sorafenib-sensitive group (Fig. 5D,E), aligning with our GEO database analysis (Fig. 3F). MYL6 microarray sorafenib resistance analysis was also consistent with the GSE109211 results (Fig. 3F, Fig. 5I,J). Kaplan-Meier curves based on the clinical information revealed better survival in the sorafenib-sensitive group (Fig. 5K).Fig. 5 Tissue chips immunohistochemical analysis. A and F: MYH9 (A) and MYL6 (F) tissue chips immunohistochemical staining. B and G: Representative pictures of MYH9 (B) and MYL6 (G) expressions in HCC and adjacent normal tissues by immunohistochemical. C and H: The statistical results of the difference of MYH9 (C) and MYL6 (H) expressions in HCC and adjacent normal tissues were analyzed by paired t-test, n = 80. D and I: The samples were divided into groups of sorafenib-resistance and sorafenib sensitivity according to whether or not they relapsed after treatment with sorafenib. Representative images of MYH9 (D) and MYL6 (I) expressions in the sorafenib-resistance and sorafenib sensitivity groups by immunohistochemical. E and J: The statistical results of the difference between MYH9 (E) and MYL6 (J) expressions in the sorafenib-resistance and sorafenib-sensitivity groups. n = 80. K: KM survival curve according to whether or not they are resistant to sorafenib based on tissue microarray clinical information. *P<0.05, **P<0.01, ***P<0.001, ns, not significant.

Fig 5

Lastly, we validated our findings using BALB/c nude mice with four cell lines: HCCLM3R, HepG2R, HCCLM3R+Bl, and HepG2R+Bl (Fig. 6C,D). Tumor diameters were measured every seven days, and sorafenib treatment at a dose of 30 mg kg-1/day [[43], [44], [45]] was started 14 days after liver cancer cell injection and concluded on day 28 (Fig. 6C). Sorafenib significantly reduced tumor volume in the (LM3R+Bl)+Sor and (HepG2R+Bl)+Sor groups compared to the respective vehicle groups, demonstrating that inhibiting MYH9 enhances sorafenib sensitivity (Fig. 6D,E). To further explore whether MYH9 silencing in HCCLM3 cells induces changes in tumor growth rate, local metastasis, and growth characteristics in nude mice, we injected different cell groups (HCCLM3R, HepG2R, HCCLM3R+Bl, and HepG2R+Bl) into the livers of nude mice to establish orthotopic tumors. Sorafenib was administered at a dose of 30 mg kg-1/day. Sorafenib treatment significantly reduced tumor volume in the (LM3R+Bl)+Sor and (HepG2R+Bl)+Sor groups compared to their respective vehicle groups. Furthermore, tumors derived from HCCLM3R and HepG2R cells were generally larger, regardless of sorafenib treatment. H&E staining indicated lung metastasis in all nude mice injected with HCCLM3R cells, likely due to the high invasiveness of HCCLM3 liver cancer cells (Fig. 6F). To better observe orthotopic tumor proliferation in each group, we used immunohistochemical staining for biological markers like Ki67 and Proliferating Cell Nuclear Antigen (PCNA) [46], evaluating the proliferative activity of orthotopic tumors in nude mice (Fig. 6F). Tumors derived from HCCLM3R and HepG2R cells exhibited significantly higher proliferative activity than those in the HCCLM3R+Bl and HepG2R+Bl groups, indicating that MYH9 suppression indeed reduces liver cancer proliferative activity. Interestingly, following sorafenib treatment, orthotopic tumor growth in the HCCLM3R+Bl and HepG2R+Bl groups slowed significantly, with a noticeable decrease in proliferative activity. Thus, MYH9 suppression effectively enhances sorafenib's therapeutic efficacy in liver cancer.Fig. 6 Validation of MYH9 associated with sorafenib resistance in hepatocellular carcinoma. A: Western Blot detected the expression of MYH9, MYL6 and SLC7A11 in liver cancer cell lines lysate (3 cases randomly selected in each group). B: Western Blot detected the expression of MYH9 and SLC7A11 in liver cancer cell lines lysate (3 cases randomly selected in each group). C: Workflow of in vivo drug sensitivity assay. D: Samples of HCCLM3R, HCCLM3R+Bl, HepG2R, and HepG2R+Bl cells (1 × 10^6) were inoculated into nude mice subcutaneously. E: Quantitative analysis of tumor volume of xenografts. Tumor volume was compared at indicated time points, and tumor weight was measured at the endpoint. F: Gross photographs of mouse liver orthotopic tumors and lung metastases, as well as H&E staining. Immunohistochemical staining of Ki67 and PCNA in mouse liver orthotopic tumors. * p < 0.05, **p < 0.01, *** p < 0.001, ns, not significant.

Fig 6

In short, our four-step validation suggests that inhibiting MYH9 leads to SLC7A11 accumulation, causing disulfidptosis-like changes in liver cancer cells. This enhances sensitivity to sorafenib, emphasizing the importance of disulfidptosis in tumor cell drug sensitivity.

MYH9′s role in pathways of sorafenib resistance

Exploring the molecular mechanisms by which MYH9 impacts sorafenib resistance in HCC cells, we employed GSEA to identify MYH9-associated KEGG and GO pathways. The analysis revealed specific signaling pathways directly associated with sorafenib resistance.

Using the TCGA database with 374 HCC patients, our study found that pathways such as collagen-activated tyrosine kinase receptor signaling and regulation of protein tyrosine kinase activity were significantly linked with MYH9 (Fig. 7D-F). This underscores MYH9′s influence on tyrosine kinase activity, which directly relates to sorafenib's mechanism of action, inhibiting multiple tyrosine kinases crucial for HCC proliferation and angiogenesis [17].Fig. 7 MYH9 can influence tyrosine kinase-related pathways to promote sorafenib resistance. A-F: Sorafenib-related MYH9 pathways were obtained by enriching MYH9-related pathways using the GSEA technique.

Fig 7

Moreover, MYH9 was found to be enriched in pathways like positive regulation of cell migration (Fig. 7C) and regulation of vascular endothelial cell proliferation (Fig. 7A), both of which are critical for tumor progression [47]. Elevated MYH9 expression fosters a pro-tumorigenic environment (Fig. 7B), enhancing cellular invasion and resistance to therapeutic agents like sorafenib. The findings also point to the transmembrane receptor protein serine/threonine kinase signaling pathway, highlighting MYH9′s role in cellular processes vital for cancer growth and resistance.

In summary, the data suggest that MYH9, through its regulatory impact on tyrosine kinase signaling, plays a significant role in facilitating sorafenib resistance. This observation aligns with the understanding that sorafenib targets several kinases influenced by MYH9, contributing to drug resistance in HCC cells. These findings highlight MYH9 as a potential target for novel therapeutic strategies to overcome sorafenib resistance in HCC.

Discussion

In this study, we conducted a comprehensive analysis of disulfidptosis-related genes in 33 tumors, providing a valuable reference for future clinical research. These genes showed associations with genomic alterations and immune infiltration across various tumors. FLNA, MYH9, TLN1, MYL6, CAPZB, and IQGAP1 primarily appear to function as tumor promoters, being linked to more aggressive cancer phenotypes. On the other hand, SLC7A11 and MYH10 were correlated with favorable immune infiltration phenotypes, suggesting their roles as tumor suppressors and potential therapeutic targets. Clustering analysis revealed associations with survival rates and drug sensitivity in cancer patients. We examined the role of disulfidptosis in sorafenib resistance in liver cancer and identified MYH9 as a key gene. Inhibiting MYH9 mitigated sorafenib resistance in hepatocellular carcinoma, potentially through enhancing disulfidptosis-like changes.

Yang et al. [48] previously found that the upregulation of MYH9 and activation of the NOTCH pathway could promote liver cancer cell stemness and lenvatinib resistance, revealing the mechanism by which MYH9 is linked to lenvatinib resistance in liver cancer. Furthermore, several studies have indicated that the pathways involving MYH9 can enhance liver cancer cell stemness, leading to sorafenib resistance in liver cancer [3,20]. This confirms that the MYH9 gene we analyzed is connected to sorafenib sensitivity in liver cancer. However, prior studies focused on how MYH9 enhances sorafenib resistance by promoting liver cancer cell stemness. Our study found that disulfidptosis is a key mechanism that can alleviate sorafenib resistance in liver cancer by enhancing disulfidptosis-like changes.

However, it's important to note that MYH9′s role in enhancing sorafenib sensitivity through disulfidptosis is still inconclusive. MYH9, a cytoskeletal protein involved in cell adhesion and migration, binds to F-actin filaments [49]. MYH9 inhibitors might disrupt cytoskeletal proteins, potentially facilitating sorafenib's entry into cells and exerting anti-tumor effects. We refer to cellular changes consistent with disulfidptosis as disulfidptosis-like changes. MYH9′s impact on disulfidptosis occurrence points to its pivotal role in determining drug sensitivity in tumor cells. Although our validation was specific to liver cancer, it provides novel insights into the relationship between drug resistance in cancer patients.

In summary, our research will significantly contribute to future laboratory studies by revealing and validating the role of disulfidptosis-related genes in tumor development, immune response, tumor microenvironment, and drug resistance. This is crucial for developing personalized cancer treatment drugs and overcoming drug resistance.

Funding

This work was financed by grants received from the 10.13039/501100001809 National Natural Science Foundation of China (Grant NO. 81970530 ; 82170638 ; 82072642 ).

Availability of data and materials

Data will be made available on request. All the data sets used in this study including TCGA dataset, GEO dataset, GDSC dataset and NCI60 cell line data are publicly available.

Ethics approval and consent to participate

All procedures complied with the Animal Committee of Tongren Hospital guidelines.

CRediT authorship contribution statement

Kangnan Zhang: Data curation, Formal analysis, Software, Writing – original draft, Writing – review & editing. Zhenhua Zhu: Data curation, Formal analysis, Methodology, Software. Jingyi Zhou: Conceptualization, Formal analysis, Methodology. Min Shi: Methodology. Na Wang: Validation. Fudong Yu: Conceptualization, Project administration, Supervision. Ling Xu: Funding acquisition, Project administration, Supervision, Visualization, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no conflict of interests.

Appendix Supplementary materials

Image, application 1

Image, application 2

Image, application 3

Acknowledgments

Not applicable.

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.tranon.2024.102091.
==== Refs
References

1 Vasan N. Baselga J. Hyman D.M. A view on drug resistance in cancer Nature 575 7782 2019 299 309 31723286
2 Gao L. Wu Z.-X. Assaraf Y.G. Chen Z.-S. Wang L. Overcoming anti-cancer drug resistance via restoration of tumor suppressor gene function Drug Resist. Updates 57 2021 100770
3 Lin X. Li A.M. Li Y.H. Luo R.C. Zou Y.J. Liu Y.Y. Silencing MYH9 blocks HBx-induced GSK3β ubiquitination and degradation to inhibit tumor stemness in hepatocellular carcinoma Signal. Transduct. Target. Ther. 5 1 2020 13 32296025
4 Valente A. Podolski-Renić A. Poetsch I. Filipović N. López Ó. Turel I. Metal- and metalloid-based compounds to target and reverse cancer multidrug resistance Drug Resist. Updates 58 2021 100778
5 Das T. Anand U. Pandey S.K. Ashby C.R. Assaraf Y.G. Chen Z.-S. Therapeutic strategies to overcome taxane resistance in cancer Drug Resist. Updates 55 2021 100754
6 Zhang K. Zhang Q. Jia R. Xiang S. Xu L. A comprehensive review of the relationship between autophagy and sorafenib-resistance in hepatocellular carcinoma: ferroptosis is noteworthy Front. Cell Dev. Biol. 11 2023 1156383
7 Meng Y. Chen X. Deng G. Disulfidptosis: a new form of regulated cell death for cancer treatment Mol. Biomed. 4 1 2023 18 37318660
8 Koppula P. Zhang Y. Shi J. Li W. Gan B. The glutamate/cystine antiporter SLC7A11/xCT enhances cancer cell dependency on glucose by exporting glutamate J. Biol. Chem. 292 34 2017 14240 14249 28630042
9 Liu X. Olszewski K. Zhang Y. Lim E.W. Shi J. Zhang X. Cystine transporter regulation of pentose phosphate pathway dependency and disulfide stress exposes a targetable metabolic vulnerability in cancer Nat. Cell Biol. 22 4 2020 476 486 32231310
10 Zheng P. Zhou C. Ding Y. Duan S. Disulfidptosis: a new target for metabolic cancer therapy J. Exp. Clin. Cancer Res. 42 1 2023 103 37101248
11 Liu X. Zhuang L. Gan B. Disulfidptosis: disulfide stress-induced cell death Trends. Cell Biol. 2023
12 Liu X. Nie L. Zhang Y. Yan Y. Wang C. Colic M. Actin cytoskeleton vulnerability to disulfide stress mediates disulfidptosis Nat. Cell Biol. 25 3 2023 404 414 36747082
13 Wang T. Guo K. Zhang D. Wang H. Yin J. Cui H. Disulfidptosis classification of hepatocellular carcinoma reveals correlation with clinical prognosis and immune profile Int. Immunopharmacol. 120 2023 110368
14 Chen H. Yang W. Li Y. Ma L. Ji Z Leveraging a disulfidptosis-based signature to improve the survival and drug sensitivity of bladder cancer patients Front. Immunol. 14 2023 1198878
15 Siegel R.L. Miller K.D. Fuchs H.E. Jemal A. Cancer statistics, 2022 CA Cancer J. Clin. 72 1 2022 7 33 35020204
16 Forner A. Reig M. carcinoma Bruix J.Hepatocellular Lancet 391 10127 2018 1301 1314 29307467
17 Abdelgalil A.A. Alkahtani H.M. Sorafenib Al-Jenoobi FI. Profiles. Drug Subst. Excip. Relat. Methodol. 44 2019 239 266 31029219
18 Li Y. Liu X. Lin X. Zhao M. Xiao Y. Liu C. Chemical compound cinobufotalin potently induces FOXO1-stimulated cisplatin sensitivity by antagonizing its binding partner MYH9 Signal. Transduct. Target. Ther. 4 2019 48 31754475
19 Derycke L. Stove C. Vercoutter-Edouart A.S. De Wever O. Dollé L. Colpaert N. The role of non-muscle myosin IIA in aggregation and invasion of human MCF-7 breast cancer cells Int. J. Dev. Biol. 55 7–9 2011 835 840 22161839
20 Hou R. Li Y. Luo X. Zhang W. Yang H. Zhang Y. ENKUR expression induced by chemically synthesized cinobufotalin suppresses malignant activities of hepatocellular carcinoma by modulating β-catenin/c-Jun/MYH9/USP7/c-Myc axis Int. J. Biol. Sci. 18 6 2022 2553 2567 35414777
21 Beroukhim R. Mermel C.H. Porter D. Wei G. Raychaudhuri S. Donovan J. The landscape of somatic copy-number alteration across human cancers Nature 463 7283 2010 899 905 20164920
22 Wu C. Tan J. Wang X. Qin C. Long W. Pan Y. Pan-cancer analyses reveal molecular and clinical characteristics of cuproptosis regulators Imeta 2 1 2023 e68 38868340
23 Gu Z. Complex heatmap visualization Imeta 1 3 2022 e43 38868715
24 Wilkerson M.D. Hayes D.N. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking Bioinformatics. 26 12 2010 1572 1573 20427518
25 Reinhold W.C. Sunshine M. Liu H. Varma S. Kohn K.W. Morris J. CellMiner: a web-based suite of genomic and pharmacologic tools to explore transcript and drug patterns in the NCI-60 cell line set Cancer Res. 72 14 2012 3499 3511 22802077
26 Yang W. Soares J. Greninger P. Edelman E.J. Lightfoot H. Forbes S. Genomics of drug sensitivity in cancer (GDSC): a resource for therapeutic biomarker discovery in cancer cells Nucleic Acids Res. 41 D1 2012 D955 DD61 23180760
27 Yoshihara K. Shahmoradgoli M. Martínez E. Vegesna R. Kim H. Torres-Garcia W. Inferring tumour purity and stromal and immune cell admixture from expression data Nat. Commun. 4 2013 2612 24113773
28 Thorsson V. Gibbs D.L. Brown S.D. Wolf D. Bortone D.S. Ou Yang T.-H. The immune landscape of cancer Immunity. 48 4 2018 812-30.e14
29 Malta T.M. Sokolov A. Gentles A.J. Burzykowski T. Poisson L. Weinstein J.N. Machine learning identifies stemness features associated with oncogenic dedifferentiation Cell 173 2 2018 338-54.e15
30 Li X.M. Liu S.P. Li Y. Cai X.M. Zhang S.B. Xie Z.F. Identification of disulfidptosis-related genes with immune infiltration in hepatocellular carcinoma Heliyon. 9 8 2023 e18436 37520990
31 Neufeld G. Mumblat Y. Smolkin T. Toledano S. Nir-Zvi I. Ziv K. The semaphorins and their receptors as modulators of tumor progression Drug Resist. Updat. 29 2016 1 12 27912840
32 Zhang X. Klamer B. Li J. Fernandez S. Li L. A pan-cancer study of class-3 semaphorins as therapeutic targets in cancer BMC. Med. Genomics. 13 Suppl 5 2020 45 32241267
33 Man S. Luo C. Yan M. Zhao G. Ma L. Gao W. Treatment for liver cancer: from sorafenib to natural products Eur. J. Med. Chem. 224 2021 113690
34 Tang W. Chen Z. Zhang W. Cheng Y. Zhang B. Wu F. The mechanisms of sorafenib resistance in hepatocellular carcinoma: theoretical basis and therapeutic aspects Signal. Transduct. Target. Ther. 5 1 2020 87 32532960
35 Cheng Z. Wei-Qi J. Jin D. New insights on sorafenib resistance in liver cancer with correlation of individualized therapy Biochim. Biophys. Acta Rev. Cancer 1874 1 2020 188382
36 Jiang H. Ning G. Wang Y. Lv W. Identification of an m6A-related signature as biomarker for hepatocellular carcinoma prognosis and correlates with sorafenib and anti-PD-1 immunotherapy treatment response Dis. Markers 2021 2021 5576683
37 Zheng L. Gu X. Zheng G. Li X. He M. Liu L. Prediction of early recurrence and response to adjuvant Sorafenib for hepatocellular carcinoma after resection PeerJ. 9 2021 e12554 34900444
38 Han X. Clark J.J. Sharma P. Bentley E.G. Kipar A. Alsayer M. Amino acids 1811-1960 of myosin heavy chain 9 is involved in murine gammaherpesvirus 68 infection Virology. 587 2023 109849
39 Li L. Sun W. Hu Q. Wang T. Zhu G. Zhao Q. Identification of MYH9 key domain involved in the entry of prrsv into permissive cells Front. Microbiol. 13 2022 865343
40 Gong S. Cao G. Li F. Chen Z. Pan X. Ma H. Endothelial conditional knockdown of NMMHC IIA (nonmuscle myosin heavy chain IIA) attenuates blood-brain barrier damage during ischemia-reperfusion injury Stroke 52 3 2021 1053 1064 33588591
41 Zheng T. Liu Q. Xing F. Zeng C. Wang W. Disulfidptosis: a new form of programmed cell death J. Exp. Clin. Cancer Res. 42 1 2023 137 37259067
42 Machesky L.M. Deadly actin collapse by disulfidptosis Nat. Cell Biol. 25 3 2023 375 376 36918690
43 Liu L. Cao Y. Chen C. Zhang X. McNabola A. Wilkie D. Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell apoptosis in hepatocellular carcinoma model PLC/PRF/5 Cancer Res. 66 24 2006 11851 11858 17178882
44 Wu C.X. Wang X.Q. Chok S.H. Man K. Tsang S.H.Y. Chan A.C.Y. Blocking CDK1/PDK1/β-Catenin signaling by CDK1 inhibitor RO3306 increased the efficacy of sorafenib treatment by targeting cancer stem cells in a preclinical model of hepatocellular carcinoma Theranostics. 8 14 2018 3737 3750 30083256
45 Li Y. Yan J. Zhao Q. Zhang Y. Zhang Y. ATF3 promotes ferroptosis in sorafenib-induced cardiotoxicity by suppressing Slc7a11 expression Front. Pharmacol. 13 2022 904314
46 Juríková M. Danihel Ľ. Polák Š. Varga I. Ki67, PCNA, and MCM proteins: markers of proliferation in the diagnosis of breast cancer Acta Histochem. 118 5 2016 544 552 27246286
47 Feng Y. Luo S. Fan D. Guo X. Ma S The role of vascular endothelial cells in tumor metastasis Acta Histochem. 125 6 2023 152070
48 Yang S.B. Zhou Z.H. Lei J. Li X.W. Chen Q. Li B. TM4SF1 upregulates MYH9 to activate the NOTCH pathway to promote cancer stemness and lenvatinib resistance in HCC Biol. Direct. 18 1 2023 18 37069693
49 Ren X. Zhu H. Deng K. Ning X. Li L. Liu D. Long noncoding RNA TPRG1-AS1 suppresses migration of vascular smooth muscle cells and attenuates atherogenesis via interacting with MYH9 protein Arterioscler. Thromb. Vasc. Biol. 42 11 2022 1378 1397 36172865
