Kaempferol triggers cellular senescence via CDK1 ubiquitination in HCC cells.

Introduction
Hepatocellular carcinoma (HCC) is the most common primary malignant tumor of the liver and the leading cause of cancer-related deaths in China [1, 2]. Although diagnostic and therapeutic approaches have advanced, HCC prognosis remains unfavorable, with a 5-year survival below 20% [3]. Current treatments face limitations due to surgical constraints, inherent resistance to radiotherapy and sorafenib, and frequent recurrence following targeted therapies. These persistent challenges highlight the urgent demand for discovering novel molecular targets and therapeutic strategies against HCC.
Cyclin-dependent kinase 1 (CDK1), a serine/threonine kinase essential for G2/M phase transition, regulates both normal tissue homeostasis and tumorigenesis [4, 5]. While CDK1 drives mitotic progression in proliferating cells, its dysregulation promotes carcinogenesis through mechanisms such as uncontrolled proliferation, genomic instability, and evasion of apoptosis [6–9]. Recent studies have identified CDK1 as hub genes associated with malignancy and adverse clinical outcomes of HCC [10, 11]. CDK1 activation significantly accelerates disease progression in malignant tumors [12]. Knockdown of CDK1 suppresses proliferation, invasion, and migration in HCC cell models [13, 14] suggesting its potential as a therapeutic target.
Cellular senescence is a state of permanent cell cycle arrest that plays roles in both physiological and pathological processes. Recent researches showed that targeting senescence pathways provide therapeutic benefits in malignancies including liver cancer [15, 16] renal cell carcinoma [17]colorectal cancer [18] and lung cancer [19]. CDK1 appears particularly relevant to senescence regulation across multiple cell types - Olaparib triggers senescence in resistant prostate cancer [20] and in mesenchymal stem cells [21] through CDK1 inhibition, while CDK1 knockdown induces senescence in oral squamous cell carcinoma cells [22]. The p21CIP1-CDK1/2 axis mediated temozolomide-induces glioblastoma cell senescence [23] and miR-200b/MYBL2/CDK1 regulates ovine granulosa cell senescence via cell cycle arrest [24]. Although some bioinformatics studies suggested CDK1 may be associated with cellular senescence and the progression of HCC [25, 26] experimental validation remains limited. These collective findings indicate CDK1 likely functions as a senescence regulator in HCC cells.
Kaempferol (KAE), a natural flavonoid abundant in fruits and vegetables, exhibits broad-spectrum antitumor activity against various cancers [27]. Studies indicated that KAE could suppress cell cycle progression by downregulating cyclins and CDKs in HCC, potentially disrupting cell cycle checkpoints [28]. Network pharmacology and molecular docking analyses revealed CDK1 as a direct target of KAE, revealing high-affinity binding between KAE and CDK1’s ATP-binding pocket in Acute Myeloid Leukemia [29] implying KAE may regulate disease progression by directly targeting CDK1. KAE also attenuates senescence-associated secretory phenotype (SASP) production in fibroblasts [30] and induces apoptosis and cellular senescence by inhibiting the PI3K/AKT and hTERT pathway in human cervical cancer cells [31]. However, whether the anti-tumor effects of KAE in HCC is mediated by CDK1-cellular senescence pathway remains unverified.
In this study, we identified CDK1 as a core regulatory protein in the development and progression of HCC, and its correlation with immune cell infiltration. Subsequent GO and KEGG revealed that CDK1 and its associated protein networks are functionally enriched in cellular senescence pathways in HCC. Experimental validation showed elevated CDK1 expression in HCC patient tissues. CDK1 overexpression suppressed cellular senescence, whereas its knockdown promoted this process in HepG2 cells. We also found that KAE has a high affinity for CDK1, triggering its ubiquitination and subsequent proteasomal degradation, reduced CDK1 protein levels, which ultimately induced senescence and suppressing HCC progression. These findings suggested that KAE and CDK1 could serve as promising therapeutic drugs and targets for HCC intervention.

Materials and methods

Data acquisition and analysis
The datasets for this study were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), and the original gene data numbers GSE101728, GSE101685, GSE124535 were downloaded. The differentially expressed genes (DEGs) between the controls and patients were screened and visualized using the “limma” R package (V3.6.3). Statistical cut-off criteria of P < 0.05 and |log FC| ≥ 1.5 was used for screening the significantly DEGs. After that, Venn Software was adopted for screening to obtain overlapping targets.
For GO and KEGG analyses, we conducted Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses using DAVID (https://david.ncifcrf.gov/), with statistical significance set at P < 0.05. The GO analysis categorized gene annotations into biological processes (BPs), cellular components (CCs), and molecular functions (MFs). Results from both GO and KEGG analyses were visualized using the ggplot2 package of R software.
For protein–protein interaction network analysis, the STRING database (V11.0) (https://cn.string-db.org/) was adopted for assessing and integrating the interactions of DEGs. After that, the disconnected nodes in the network were hided. The result was downloaded from STRING (V3.10.3), then imported into Cytoscape V3.10.0, where the CytoHubba plugin identified the top 30 node genes. Intersection analysis refined these to hub genes, with MCC scores prioritizing CDK1, PBK, BUB1B, and NUF2 as the key genes.
GEPIA database (http://gepia.cancer-pku.cn/) was used to evaluate the gene expression differences between tumor and non-tumor tissues using ANOVA, with results visualized as box plots. This dataset also enabled correlation analyses between CDK1, PBK, BUB1B, and NUF2 expression levels and both overall survival (OS) and disease-free survival (DFS) outcomes. P < 0.05 indicated statistical significance.
Images of immunohistochemistry staining for tumor and non-tumor tissues were collected from Human Protein Atlas database (https://www.proteinatlas.org/).
For tumor infiltration analysis, we assessed tumor immune cell infiltration using Tumor Immune Estimation Resource (TIMER) database (https://cistrome.org/TIMER/). The correlation coefficients for gene expression were calculated with Spearman’s correlation and displayed as log2 RSEM. P < 0.05 is statistically correlated.

HCC samples and cell lines
The tissues of 12 patients with hepatocellular carcinoma (tumor group) and paired adjacent tissues (control group) were collected from the Department of Pathology, First People’s Hospital of Zunyi. After isolation, immediately collect the tissue in pre-cooled PBS under sterile conditions. For immunohistochemistry, samples were fixed with 4% paraformaldehyde and embedded in paraffin. For protein and RNA analysis, samples were frozen in liquid nitrogen and stored at −80 °C until use. All experiment procedures were approved by the Human Research Ethics Committee of First People’s Hospital of Zunyi (No. 202503110002). Written informed consent was signed before the study.
The human HCC cell line HepG2 cells were cultured in DMEM medium (Thermo Fisher Scientific, Waltham, MA, USA), supplemented with 10% fetal bovine serum (FBS, Thermo Fisher Scientific, USA), 100 mg/mL penicillin, and 100 mg/mL streptomycin (Sigma-Aldrich, USA) in a humidified incubator with 5% CO2 at 37 °C.

CCK8 assay
The viability of the HepG2 cells was detected by CCK-8 assay (Beyotime Biotechnology, China; Cat#C0037). After the indicated treatments, the cells were washed with PBS and 10 µL CCK-8 solution at a 10% dilution was added to each well, and then the plate was incubated for 2 h at 37 °C. Absorbance at 450 nm was measured using a microplate reader (Bio-Tek, USA). Cell viability (%) = (OD treatment group/OD control group) ×100%.

Cell transfections
For the knockdown of CDK1, two distinct siRNAs of each gene were synthesized and purified by Genepharma (Jiangsu, China). The overexpression plasmid of pcDNA3.1-CDK1 was purchased from Sangon Biotech (Shanghai, China). HepG2 cells were transfected using Lipofectamine 3000 Reagent (Cat. No. L3000001; Thermo Fisher Scientific, USA) following the manufacturer’s protocols. Si-NC (scrambled sequence) or vector (empty vector of pcDNA3.1) served as negative control. These sequences are listed in Table 1. RT-qPCR and Western blotting were used to validate the knockdown or overexpression efficiencies.

RNA isolation and quantitative real-time polymerase chain reaction (RT-qPCR)
Total RNAs were extracted from cells or tissues by Trizol reagent (Cat. No. 12183555; Thermo Fisher Scientific, USA). After cDNA synthesis with the NovoScript All-in-one SuperMix (Novoprotein, China), the CFX96 system (Bio-Rad, Hercules, CA, USA) and SYBR Green Master Mix (Cat. No. 4309155; Thermo Fisher Scientific, USA) were used to perform RT-qPCR assays. Relative mRNA expression was normalized by Actin as gene expression, and calculated by 2−ΔΔCt method. The primer sequences are shown in Table 2.

Protein stability analysis
HepG2 cells were incubated in 4 cm plates and maintained in an incubator overnight at 37 °C. Then, cells were treated with 100 µg/mL cycloheximide (CHX) (Sigma, USA) for 0, 2, 4, 6, or 8 h. Total proteins in each group were extracted and subjected to Western blotting analysis.

Western blotting
Total proteins were isolated from cells or tissues using the RIPA lysis buffer (Beyotime, China). The primary antibodies used in this study included anti-rabbit Actin (Abcam, ab179467, 1:10000), anti-rabbit CDK1 (Abcam, ab265590, 1:1000), anti-rabbit P16 (Abcam, ab185620, 1:1000), anti-rabbit P21 (Abcam, ab109520, 1:2000). The secondary antibodies were goat anti-rabbit IgG H&L (HRP) (Abcam, ab6721, 1:10000). β-actin was the internal control. Protein bands were detected by an ECL kit (EpiZyme, China) and visualized with a ChemiDoc imaging system (BioRad, USA). The intensity of each band was quantified by Image J.

Ubiquitination analysis
Cells were lyzed with ice-cold IP lysis buffer to obtain the supernatant containing total proteins. The cell lysates were incubated with CDK1 antibody (Abcam, ab265590, 20 µL/mg lysate) in a tube overnight at 4 °C to, with homologous anti-IgG as the control antibody. Then, the pre-treated protein A/G beads were added to each tube and incubated at room temperature for 1 h. Free CDK1 and CDK1-Ub were enriched by beads and eluted by elution buffer. 10% cell lysates of total lysates for input control. Subsequently, these proteins were analyzed by Western blotting using anti-rabbit Ub (Abcam, ab7780, 1:1000).

Senescence-associated β-galactosidase (SA-β-gal) assay
SA-β-gal staining was performed using an SA-β-gal staining kit (Beyotime Biotechnology, China; Cat#C0602) according to the manufacturer’s protocol. In brief, cells were fixed in 4% paraformaldehyde for 5 min. After washing with PBS three times, samples were incubated in SA-β-gal solution at 37 °C for 18 h. Then, ice-cold PBS was used to stop the enzymatic reaction. Images were captured using microscope (Olympus, Japan). SA-β-gal positive cells (%) = positive numbers/total number of cells in 5 random fields.

Immunohistochemistry staining
Tissue samples were fixed with 4% paraformaldehyde and embedded in paraffin. Section (4 µm in thickness) were incubated in 10 mM citrate buffer for antigen retrieval. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide. and then, the primary antibodies included anti-rabbit CDK1 (CST, #36169, 1:1000, USA). Sections were rinsed with PBS and incubated with goat anti-rabbit secondary antibody (Abcam, ab6721, 1:5000, UK). The antigens were visualized using 3,3’-diaminobenzidine (Cat. No. D8001; Millipore, USA), and the slides were counter-stained with hematoxylin (Cat. No. H9627; Millipore, USA).

Statistical analysis
SPSS 13.0 software was used for statistical analysis of the data. All data are shown as the mean ± standard deviation as indicated. Independent samples Student’s t-test was applied to compare two groups. One-way ANOVA was performed for comparing three or more groups, and Tukey’s multiple comparison test was employed for post-hoc analysis. Receiver operating characteristic (ROC) curve and area under the curve (AUC) values analyses were performed to evaluate the diagnostic accuracy of the gene expression levels. P < 0.05 was considered statistically significant.

Results

96 DEGs were screened in HCC and involved in the cell cycle and metabolism
The GSE101728, GSE101685, and GSE124535 datasets from GEO were analyzed using the R package “limma,” with results visualized in Figs. 1A-C. 96 overlapping genes (|logFC|≥1.5, P < 0.05) were obtained from Venn diagram in the three datasets (Fig. 1D). GO analysis indicated these genes were enriched in chromosome segregation and mitotic nuclear division (BP, Fig. 1E); chromosomal region and condensed chromosome, among others (CC, Fig. 1E); and G protein − coupled receptor binding, peptidase inhibitor activity and DNA replication origin binding (MF, Fig. 1E). KEGG pathway analysis highlighted involvement in cell cycle, p53 signaling and TNF signaling and metabolic pathways (Fig. 1F). These results suggested that the 96 DEGs contributed to HCC pathogenesis by modulating critical cell cycle and metabolic pathways.

CDK1, PBK, BUB1B, and NUF2 are highly expressed and serve as indicators of unfavorable prognosis in HCC
To investigate gene-gene regulatory networks in HCC progression, we analyzed correlation analysis of 96 DEGs using Cytoscape (Fig. 2A). Our analysis found that CDK1, PBK, BUB1B, and NUF2 emerged as hub genes with the higher connectivity (Fig. 2B). Comparison of mRNA expression between HCC (N = 369) and non-tumor tissues (N = 160) from the GEPIA showed significantly elevated levels of these four genes in tumor samples (Figs. 2C-F, P < 0.0001). Higher expression of CDK1, PBK, BUB1B, and NUF2 correlated with reduced overall survival (OS) in HCC patients (Figs. 2G-J). ROC curve analyses yielded AUC values of 0.977 (CDK1), 0.966 (PBK), 0.953 (BUB1B), and 0.983 (NUF2), indicating strong diagnostic potential (Figs. 2K-N). These findings collectively implicated CDK1, PBK, BUB1B and NUF2 as both prognostic indicators and candidate diagnostic biomarkers for HCC.

CDK1 was positively correlated with immune cell infiltration and related to the prognosis of HCC patients
CDK1 exhibited the highest node count in gene-gene regulatory network analyses (Fig. 2B), implicating its central role in HCC progression. However, the role and mechanism of CDK1 in HCC are still unclear. Therefore, we continued to investigate the role of CDK1 in HCC.
Analysis of CDK1 mRNA and protein levels in tumor tissues using the TCGA-LIHC cohort and Human Protein Atlas revealed significantly elevated expression in HCC tissues (Figs. 3A, B). RT-qPCR, WB, and IHC analyses confirmed CDK1 overexpression in HCC patient samples compared to adjacent non-tumor tissues (Figs. 3C-E). Survival analysis demonstrated that elevated CDK1 expression correlated with reduced overall survival (OS), recurrence-free survival (RFS), and progression-free survival (PFS) in HCC patients (Figs. 3F-H). In addition, TCGA-LIHC population data further associated CDK1 expression with multiple clinicopathological parameters, including TNM stage, vascular invasion, gender, age, weight and serum AFP levels (Fig. 3I). Finally, TIMER analysis revealed positive correlations between CDK1 expression and tumor infiltration by various immune cell subtypes, including B cells, CD4+ T cell, CD8+ T cell, macrophage, neutrophil, and dendritic cell (Figs. 3J, K). Collectively, these findings established CDK1 as both a prognostic biomarker and potential immunomodulatory factor in HCC.

CDK1 regulated HCC progression by inducing cellular senescence
To further investigate the biological function of CDK1, we constructed protein-protein interaction network (PPI) and found that 10 proteins (CCNB1/2, CDC20, CCNA1/2, CDC25, CKS1B, BUB1B, and CKS2) closely associated with CDK1. GO analysis indicated these genes involved in mitotic cell cycle phase transition, nuclear division, and cyclin-dependent protein kinase regulation (Figs. 4A, B). KEGG analysis revealed that CDK1 influenced HCC progression through key pathways such as the cell cycle, cellular senescence, the p53 signaling pathway (Fig. 4C).

While CDK1 has been linked to cellular senescence in previous studies [32, 33]its role in HCC remains unknown. To investigate CDK1’s role in HCC, we transfected cells with siRNA or pcDNA3.1 plasmid to achieve knockdown or overexpression in HepG2 cells. CDK1 knockdown reduced both mRNA and protein levels (Fig. 4D), whereas overexpression elevated them in HepG2 cells (Fig. 4E). We then assessed cellular senescence markers - senescence-related β - galactosidase (SA-β-gal activity), and cell cycle arrest related proteins (P16 and P21). In HepG2 cells, CDK1 knockdown increased SA-β-gal activity, up-regulated the protein levels of P16 and P21 (Figs. 4F, G), while overexpression produced the opposite effects (Figs. 4H, I). These results suggested CDK1 suppressed senescence, implicating it as a potential therapeutic target. Our findings established CDK1’s involvement in senescence regulation and its possible influence on HCC progression.

KAE-induced cellular senescence by targeting CDK1
Kaempferol (KAE, molecular structures in Fig. 5A) is a flavonoid found in various plants and has antioxidant, anti-inflammatory, and anti-tumor effects. Recent studies showed that CDKs (cyclin-dependent kinases) and TP53 may serve as key molecular targets for KAE’s anti-HCC activity [34]. However, whether KAE exerted its anti-HCC effect by targeting CDK1 remains unknown.

Molecular docking analysis revealed interactions between KAE and the core target protein CDK1, with autodock simulations demonstrating high-affinity binding (affinity = −9.7 kcal/mol). These results suggested KAE spontaneously forms stable complexes with CDK1, supporting its therapeutic potential for HCC. Three-dimensional modeling of the CDK1-KAE complex identified the lowest-energy binding conformations and potential hydrogen bonds (Fig. 5B). KAE treatment (0, 20, or 40 µM) significantly decreased HepG2 cell viability in a concentration-dependent manner (Fig. 5C). Moreover, KAE induced cellular senescence by increasing SA-β-gal activity, and upregulating P16 and P21 protein levels in HepG2 cells (Figs. 5D, E). Taken together, these data implied that KAE modulated cellular senescence by targeting CDK1.

KAE downregulated CDK1 protein levels, promoted its ubiquitination and degradation
KAE significantly reduced CDK1 protein levels in HepG2 cells in a concentration-dependent manner (0, 20, or 40 µM; Fig. 6A). To investigate the mechanism underlying this effect, we first assessed CDK1 protein stability using cycloheximide (CHX, a protein translation inhibitor) to inhibit protein translation. It was found that the protein degradation of CDK1 was increased in KAE-treated HepG2 cells (Fig. 6B), showing that KAE promotes CDK1 protein degradation. In general, proteins are degraded through two pathways: the ubiquitin-proteasome pathway and the autophagy-lysosome pathway, which could be specifically blocked by treatment cells with MG132 or chloroquine (CQ), respectively. MG132 treatment accumulated CDK1 protein levels in KAE-treated HepG2 cells, whereas CQ had no effect (Fig. 6C), indicating proteasomal degradation mediates CDK1 reduction in HepG2 cells. To further confirm this pathway, the protein level of CDK1-Ub was directly determined. KAE decreased total CDK1 protein level (in input) but increased CDK1-Ub level in KAE-treated HepG2 cells (Fig. 6D, lane 2 VS 1). Furthermore, MG132 co-treatment further elevated CDK1 protein level (in input) and CDK1-Ub level in KAE-treated HepG2 cells (Fig. 6D, lane 4 VS 2), demonstrating proteasomal blockade reverses KAE-induced degradation. Taken together, these findings demonstrated that KAE reduced CDK1 level by promoting its ubiquitin-proteasome-mediated degradation in HepG2 cells.

Overexpression of CDK1 enhanced cell viability, an effect reversed by KAE treatment (Fig. 6E). Moreover, KAE additionally induced cellular senescence in HepG2 cells, as evidenced by elevated SA-β-gal activity and upregulated P16/P21 protein expression, while these changes were all abolished by overexpression CDK1 (Figs. 6F-G). In summary, these results demonstrated that KAE induced cell senescence by CDK1 inhibition (Fig. 6H), suggesting therapeutic potential for HCC.

Discussion
In this study, we identified CDK1 as a hub gene associated with hepatocarcinogenesis, with its elevated expression showing positive correlations with poor prognosis, tumor cell infiltration, and tumor pathological stage. The findings of are highly consistent with those of other researchers [13, 35]. Cyclin-dependent kinases (CDKs), a highly conserved Ser/Thr protein kinase, play an important role in the process of cell cycle [36]. The targeted therapies that inhibit CDKs promote cell growth arrest and cell senescence in pancreatic cancer [37]with specific CDK4/6 inhibitors like palbociclib, ribociclib, and amebaciclib demonstrating senescence-inducing effects [37]. However, the potential role of CDK1 regulates cellular senescence in HCC remains unexplored, GO and KEGG analysis showed that CDK1 and its related proteins predominantly enriched in cellular senescence. Experimental validation in HepG2 cells confirmed CDK1-mediated senescence induction, supporting CDK1’s involvement in senescence-driven tumor suppression.
KAE, a tetrahydroxy-flavonoid, exhibits anticancer properties by promoting cancer cell apoptosis, inhibiting cancer metastasis, and reducing drug resistance and side effects [27, 38]. Network pharmacological studies demonstrated that KAE could interact with CDK1 [39, 40], and induce cell death via ER stress and CHOP-autophagy signaling pathway in HepG2 or Huh7 cells [41]. KAE also inhibited HCC progression by inducing cellular autophagy in Hep3B cells [42]. These studies suggest that KAE may also execute anti-tumor effects by cellular autophagy. Autophagy conversely also regulates cellular senescence, as evidenced by bleomycin induced senescence through lysosomal membrane permeabilization and ROS-mediated autophagy blockade in HT22 cells [43]. This interplay is further supported by METTL3-dependent m6A modification of ATG7, which elevates SASP and autophagy to drive senescence and osteoarthritis progression [44]. However, in this study, we only found preliminary evidence that KAE induced cellular senescence and reduced CDK1 level by promoting CDK1 ubiquitination-mediated degradation in HepG2 cells, the potential CDK1-autophagy regulatory axis in this process requires further investigation.
CDK1, a key kinase at cell cycle checkpoint serves as an effective potential therapeutic target for cancers. Dinaciclib, a CDK1 specific inhibitor, induces senescence by promoting mitochondrial dysfunction and increasing mitochondrial ROS in patient-derived 2D and 3D glioblastoma cell models [45]. And another CDK inhibitor p21CIP1 similarly interacts with CDK1, arresting cell cycle in the G2 phase and potentiating TMZ-induced senescence in glioblastoma [23]. In addition, inhibition of CDK1 also could enhance DNA damage, induce cell senescence, and restore Olaparib sensitivity in Olaparib-Resistant Prostate Cancer [20]. Moreover, knocking-down or inhibiting CDC2/CDK1 could prevent the escape from accelerated cellular senescence (ACS) in human lung cancers [46]. Our experiments demonstrated that CDK1 knockdown via siRNA increased senescence markers (the levels or activity of P16, P21, and SA-β-gal), and induced cellular senescence in HepG2 cells. Taken together, inhibition of CDK1 by siRNA or inhibitors as an effective method in anti-cancer therapy.
Cellular senescence has a “double-edged sword” effect in HCC progression, exerting both anti-tumor and pro-tumor effects. In the early stages of HCC, senescence in damaged hepatocytes prevents malignant transformation, whereas in advanced preclinical models, the senescence-associated secretory phenotype (SASP) remodels the tumor microenvironment to facilitate progression. lncRNA NEAT1 deficiency increased senescence in HCC tissue and HepG2 cells, accelerating HCC progression [47]. PD-0332991 (CDK4/6 inhibitor) effectively induced cellular senescence in hepatic tumor initiating cells (hTICs), elevating SASP factors CCL2 and CXCL10 while suppressing HCC initiation [48]. Conversely, obesity-induced HCC models demonstrated that senescent hepatic stellate cells (HSCs) secrete abundant IL-33 through SASP; subsequent IL-33/ST2 signaling activates regulatory T cells, impairing anti-tumor immunity and fostering HCC progression [49]. Therefore, establishing accurate senescence detection methods is crucial for the research and evaluation of senescence interventions.
The degradation of CDK1, a subunit of the CDK1/cyclin B kinase complex, is essential for kinase inactivation and mitotic exit. In eukaryotic cells, selective protein degradation primarily occurs through the autophagy-lysosome and ubiquitin-proteasome systems [50, 51]. Galindo et al., demonstrated that chemotherapeutic agents and proteolytic stress induced CDK1 degradation in human breast cancer MCF7 cells through p62/HDAC6-mediated selective autophagy [52]. Pectolinarigenin promoted the degradation of CDK1 protein dependent on the auto-lysosomal pathway, thereby inhibiting the proliferation of glioblastoma cells by inducing G2/M phase cell cycle arrest [53]. Moreover, CUL4-DDB1-DCAF1 complexes facilitated CDK1 ubiquitination and subsequent proteasomal degradation in a murine model of spontaneous breast cancer [54]. In addition, OTUD4 directly interacted with CDK1 and stabilized CDK1 by removing its K11, K29, and K33-linked polyubiquitination, thereby promoting the progression of glioblastoma [55]. In this study, we found that KAE treatment reduced CDK1 protein levels in KAE-treated HepG2 cells by promoting ubiquitin-dependent degradation, suggesting a proteasomal rather than autophagic pathway.
This study lacks sufficient experimental evidence to confirm KAE-CDK1 binding or identify their interaction sites. The E3 ubiquitin ligase mediating KAE-induced CDK1 ubiquitination also remains unidentified. In addition, further study is required to identify the dominant cell death pathway responsible for KAE-induced cytotoxicity, whether apoptosis, autophagy, senescence or ferroptosis. Subsequent research will investigate the therapeutic effects and mechanisms of KAE through animal experiments.

Conclusions
Here, we demonstrate that KAE promotes CDK1 ubiquitination and degradation, thereby reducing CDK1 levels and inducing cellular senescence, which suggests its therapeutic potential for HCC.