Abstract
Purpose
Our study aims to develop and validate a novel molecular marker for the prognosis and diagnosis of hepatocellular carcinoma (HCC)
Materials & methods
We retrospectively analyzed mRNA expression profile and clinicopathological data of HCC patients fetched from The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO) and The International Cancer Genome Consortium (ICGC) datasets. Univariate Cox regression analysis was performed to collect differentially expressed mRNA (DEmRNAs) from HCC and non-tumor tissues, and YEATS2, a prognostic marker, was identified by further analysis. ROC curve, survival analysis and multivariate Cox regression analysis as well as nomograms were used to evaluate the prognosis of this gene. Finally, the biological function of this gene was preliminarily discussed by using single gene Gene Set Enrichment Analysis (GSEA), and the YEATS2 overexpression and knockdown hepatoma cell line was used to verify the results in vitro and in vivo.
Results
Based on the clinical information of HCC in TCGA, GEO and ICGC databases, the gene YEATS2 with significant differences from HCC was identified. There was a statistical difference in the survival prognosis between the two databases and the ROC curve showed that the survival of HCC in both TCGA, GSE14520 and ICGC groups had a satisfactory predictive effect. Univariate and multivariate Cox regression analysis showed that YEATS2 was an independent prognostic factor for HCC, and Nomograms, which combined this prognostic feature with significant clinical features, provided an important reference for the clinical prognostic diagnosis of HCC. Next, we constructed overexpression and knockdown YEATS2 cell line in Hep3B and LM3 cells, and further proved that overexpression YEATS2 promote the proliferation and migration of HCC cells by CCK8, colony formation experiment, and transwell assays, and knockdown YEATS2 inhibited the proliferation and migration of HCC cells by CCK8, colony formation experiment, and transwell assays. Finally, the biological function of YEATS2 was preliminarily explored through GSEA analysis of a single gene, and it was found that it was significantly correlated with cell cycle and DNA repair, which provided us with ideas for further analysis. Furthermore, the knockdown of YEATS2 promoted radiation-induced DNA damage, enhanced radiosensitivity, and ultimately inhibited the proliferation of hepatocellular carcinoma cells in vitro and in vivo.
Conclusions
Our study identified a promising prognostic marker for hepatocellular carcinoma that is useful for clinical decision-making and individualized treatment.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13402-024-01019-4.
Keywords: YEATS2, Hepatocellular carcinoma, Prognostic marker, Risk score, Biological function, Radiotherapy sensitization
Introduction
Liver cancer is one of the most common malignant tumors in China, with a high degree of malignancy and unsatisfactory prognosis [1–3]. As a country with a high incidence of liver cancer, China ranks third in the world for liver cancer mortality, with approximately 390,000 deaths accounting for 13% of cancer deaths in China, and the mortality rate is second only to lung cancer [4]. Its occurrence and development is a multi-factor, multi-step and multi-gene process involving the abnormal expression of multiple oncogenes and tumor suppressor genes [5–8]. At present, radical treatment of early HCC is mainly surgical resection, ablation, and liver transplantation, but most patients with HCC have poor prognosis due to frequent recurrence or distant metastasis after treatment, with the 5-year survival rate less than 40% [9–11]. Treatment for hepatocellular carcinoma (HCC) has experienced major advancements since the last update of the official Barcelona Clinic Liver Cancer (BCLC) prognosis and treatment strategy published in 2018 [12]. The conventional clinicopathological parameters to detect the prognosis of HCC patients include pathological differentiation, tumor-node-metastasis (TNM) staging, and vascular invasion [3, 13, 14]. However, due to the heterogeneity of liver cancer, their prediction effect is not good. In addition, alpha-fetoprotein is a specific marker in the diagnosis of hepatocellular carcinoma, with a positive rate of about 70% [15]. Currently, it has been widely used in the general survey, diagnosis, evaluation of treatment effect, and prediction of recurrence of liver cancer, but its sensitivity is not high. In about 50% of patients with liver cancer, alpha-fetoprotein does not increase throughout [16]. Therefore, it is of great clinical significance to study the molecular indexes used in the prognosis diagnosis of patients with liver cancer to improve the prognosis of patients.
YEATS2, the protein encoded by this gene is a scaffolding subunit of the ATAC complex, which is a complex with acetyltransferase activity on histones H3 and H4 [17]. Alternative splicing results in multiple transcript variants encoding different isoforms [18]. Recent research has focused on studying YEATS2 as a Selective Histone Crotonylation Reader, mainly involved in epigenetic regulation [19–23]. For instance, YEATS2 exhibits best preference for Lysine benzoylation (Kbz) than lysine acetylation and crotonylation due to its wider ‘tip-sensor’ pocket [23]. It also has been reported that YEATS2 has carcinogenic potential in tumors and can regulate the TAK1/NF-κB signaling pathway, which is crucial for the survival of pancreatic ductal adenocarcinoma cells, can promote pancreatic cancer cell proliferation and migration [24–26]. YEATS2 can also bind to acetylated histone H3 through its Y domain and act as a histone H3K27ac reader, which can regulate the transcription program essential for NSCLC tumorigenesis [17]. However, its expression in different HCC risk factors and its relationship with survival have not been reported.
In recent years, gene examination has also provided great help in the diagnosis of HCC [27]. With the continuous accumulation of genome-wide gene expression information, it is possible to form credible genetic markers in HCC [28–30]. However, novel prognostic biomarkers are urgently needed because no effective and reliable prognostic biomarkers currently exist for HCC patient [31]. Therefore, in-depth mining of publicly available genomic data is an important strategy for evaluating single gene markers with reliable predictive power in the OS of HCC patients, which is expected to improve patient risk stratification and individual therapeutic intervention [32]. In addition, Genetic and epigenetic mutations are common in HCC [33]. In our study, we obtained the mRNA expression profile of HCC from TCGA and GEO, found a prognostic indicator with good reliability for HCC patients, and analyzed and verified it from multiple perspectives. Our current study has successfully identified a new candidate HCC that provides more convincing prognostic evidence for predicting HCC survival in addition to traditional clinicopathological indicators. The whole process is helpful to improve the current detection system for the prognosis diagnosis of liver cancer patients. Furthermore, this study sheds more lights on the HCC-related molecular mechanisms, confirming that knocking down YEATS2 significantly inhibited the proliferation, colony formation experiments, and migration and invasion of liver cancer cells. Additionally, YEATS2 is closely related to DNA damage repair, and knocking down YEATS2 can significantly enhance the radiosensitivity of liver cancer cells and suppress liver cancer growth.
Materials and methods
Cell culture, viruses, stimulation and transfection
In this study, LM3 (CRL-2916) cells were used to construct a stable knockdown YEATS2 cell line. It was derived from the laboratory cell bank. The media DMEM (Gibco, NY, USA) was required for LM3 cell culture. Cells were cultured in a cell incubator at 37℃ with 5% CO2 and the medium containing 10% (v/v) BCS. The sequences of shRNAs targeting YEATS2 used in this paper are presented in Table 1. Plasmid transfection using Lipofectamine max was carried out according to the manufacturer s instructions, and colonies with stable expression were screened by puromycin (1 μg/ml).
Table 1.
Plasmid construction primers
| Name | Sequence | Product Size(bp) |
|---|---|---|
| YEATS2-sh#1 | F: CGTCAGAGTTCAAGTTCATTT | 21 |
| R: AAATGAACTTGAACTCTGACG | ||
| YEATS2-sh#2 | F: GCTGCAAGATTGTGTCAGGTT | 21 |
| R: AACCTGACACAATCTTGCAGC | ||
| YEATS2-sh#3 | F: GCGAGCTACGAACAATGCTAA | 21 |
| R: TTAGCATTGTTCGTAGCTCGC |
Data extraction and manipulation
The original mRNA expression information was obtained from the TCGA and GEO databases, respectively. Firstly, we excluded the information of HCC patients with no differentially expressed mRNA from the clinical information (including follow-up or survival status) and divided the DEmRNAs into tumor tissues and normal samples. Next, we further studied DEmRNAs by using the DESeq vesion 1.38.0 R package in the TCGA dataset and by the Limma version 3.36.2 R package in GEO dataset. Finally, univariate COX regression analysis was used to extract genes significantly associated with the prognosis of HCC patients.
Establishment of nomogram
Nomogram has a robust capacity to predict tumor prognosis. A nomogram was established through incorporating all significant prognostic clinicopathological parameters determined through multivariate Cox regression analysis, thus estimating the probability of 1 -, 3-, and 5 years OS of HCC. We calculated the concordance index (C-index) to identify the discrimination of a nomogram. The calibration curve of a nomogram was utilized to vividly assess the consistency between its prediction probabilities and the actual observation.
Western blot analysis
Collected cells were washed three times in 1x PBS and then lysed on ice in IP lysis buffer containing a protease inhibitor cocktail. An SDS-polyacrylamide gel was used to separate the total proteins obtained by cell lysis, which were then transferred to a polyvinylidene fluoride membrane.
Antibodies reagents
YEATS2(abmart, PA5853), AFP(abclonal, A24997), β-actin(abcam, ab8226), GAPDH(Proteintech, 60,004-1-Ig), p-H2AX(Ser139)(Santa Cruz, sc-517,348), Doxorubicin hydrochloride(Topscience, T1020).
Cell proliferation, colony formation experiment, and transwell assays
In the cell proliferation assay, Cell Counting Kit-8 was used in accordance with its instructions. First, each well of a 96-well plate was seeded with 1000 cells, and each group had 5 sub-wells. Measurement of the OD450 began 2 hr after mixing with CCK8 reagent. In the colony formation, each well of a 6-well plate was seeded with 400 cells, and each group had 3 sub-wells. 15 days later, the cells were fixed with methanol and stained with 0.5% crystal violet. Images were obtained with the light microscope. In the cell transwell assay, dilute with serum-free culture medium with Matrigel (Becton, Dickinson and Company, USA) at a ratio of 1:8, Mix, and add 50 μL into the chamber. Add 3 × 105 cells to each chamber, 48 hours later, the cells were fixed with methanol and stained with 0.5% crystal violet. Images were obtained with the light microscope.
RNA extraction assays
The overall RNAs originated from the lysis of TRizol reagents (Takara, Kusatsu, Japan). cDNAs were obtained through reverse transcription by RNAs with reverse transcription reagents (akara, Kusatsu, Japan). Real-time fluorescent qPCR was performed using Bio-rad CFX Connect real-time fluorescence qPCR instruments.
Tumor formation in nude mice
Female SCID mice aging between 4 to 6 weeks were provided by Hunan SJA Laboratory Animal Co., Ltd (Changsha, Hunan). Animal research was held under the permission of Xiangya Hospital Animal Ethics Committee and in accordance with animal protection legislation and federal criteria. Each mouse was subcutaneously injected with knockdown YEATS2 in LM3 cells, along with corresponding control cells (3 × 106 cells per mice), into the armpits. The tumor volume and mice weight were measured every 3 days. When the tumors reached a volume of around 100 mm3, the mice were irradiated with radiotherapy. The mice were irradiated with 5 Gy three times on the 17, 20, and 23 days. When the tumors were detected to have a volume around 200mm3, the mice were euthanized.
Histology and immunohistochemistry
The Department of Pathology at Xiangya Hospital verified and provided biopsies of lung cancer and associated diseases. The method used for IHC analysis of paraffin sections from lung cancer tissues can be found in previous literature. Images of the paraffin sections were captured with a CX41 microscope (Olympus, Tokyo, Japan) equipped with a DP-72 microscope digital camera system (Olympus, Tokyo, Japan), and differential quantification was performed by two pathologists from Xiangya Hospital, Changsha, China.
Statistical analysis
The criteria of the TNM classification system of malignant tumors (American Joint Committee on Cancer and Union for International Cancer Control, 2009) were used in our study. To evaluate the survival differences between low- and high-risk HCC patients, we performed survival analysis through Kaplan-Meier curve combined with the log-rank test. We also conducted univariate and multivariate Cox regression analyses to identify the association between OS and risk score as well as clinicopathological features. The ROC curve with the corresponding area under the curve (AUC) were rendered to estimate the predictive performance of the prognostic gene signature for HCC survival by the R package “survival ROC”. P < 0.05 was defined statistically significant.
Prediction of biological function
To further explore the biological function of this gene, we used the single-gene GSEA analysis to investigate the potential pathways and biological processes between different gene expression groups in the HCC cohort of GSE14520 and TCGA.
Results
Screening of differentially expressed genes associated with prognosis
We screened 3256 DEmRNAs from HCC samples (n = 374) and noncancerous samples (n = 50) in the TCGA dataset. Figure S1 shows the heatmaps of these DEmRNAs. In order to narrow down the range of genes, we further utilized the univariate COX regression model to identify DEmRNAs that were significantly correlated with the OS of HCC patients., we found that Ada-two-A-containing (ATAC) histone acetyltransferase complex related genes were significantly enriched in the differential genes (Fig. 1B). And then, using the GO database entries ATAC-the inside of the complex gene (POLE4 KAT14, TADA3, YEATS2, MAP3K7, KAT2A, WDR5, SGF29, TADA2A, MBIP, POLE3, DR1, KAT2B, ZZZ3) constructed ATAC-complex signature, using GSVA to score the signature in normal and tumor samples in TCGA database, found that the expression of ATAC-complex was significantly increased in tumor (Fig. 1). Then analyzed each gene of signature, among them, only YEATS2 and MAP3K were differentially expressed genes (adj.P.val < 0.01 & FC > 3). In COX analysis, YEATS2 had the largest HR, so we selected it for further analysis (Table 2). Next, we used SPSS to integrate the clinicopathological data obtained from the TCGA and GEO databases then performed a chi-square test, and finally selected YEATS2 with a large difference (Tables 3–4). After consulting relevant literature, we found that the effect of this gene on tumor specificity has not been reported, so we chose it for further verification, expecting to obtain meaningful results for the clinical treatment of tumors.
Fig. 1.
Validation the prognostic signature of YEATS2. A Heat maps were used to analyze the differential expression of genes in ATAC complexes in clinical samples of HCC. B Volcano plot was used to analyze the differential expression of genes in the ATAC complex in clinical samples of HCC. C The expression of ATAC-complex gene in normal tissues was far lower than that in tumor tissues (P < 0.0001). D, G, J Statistical analysis of YEATS2 expression in normal tissues and tumor tissues of TCGA, GEO and ICGC database samples related to HCC. The expression of YEATS2 in normal tissues was far lower than that in tumor tissues (P < 0.0001). E, H, K Survival analysis of YEATS2 expression in HCC tissues of TCGA, GEO and ICGC database. The survival of patients with YEATS2 high expression was significantly worse than those with YEATS2 low expression (P = 0.0003). F, I, L The ROC curve verification of the prognosis results in the TCGA, GEO and ICGC databases, and the score of AUC (Area Under Curve) in the TCGA, GEO and ICGC databases were 0.893, 0.888 and 0.878, respectively. (P < 0.05)
Table 2.
COX analysis of HR for ATAC-complex genes
| ID | HR | HR.95 L | HR.95 H | P value |
|---|---|---|---|---|
| YEATS2 | 1.09599552038 | 1.053937921359 | 1.13973143613 | 4.404467229309E-06 |
| POLE4 | 1.001753114428 | 0.9957355609757 | 1.007807033911 | 0.5688235335027 |
| KAT14 | 1.061364482780 | 1.027156793550 | 1.09671140022 | 0.0003666496084553 |
| TADA3 | 1.004022283570 | 0.9956939935370 | 1.01242023397 | 0.344884342568236 |
| MAP3K7 | 1.044380336925 | 1.021952972027 | 1.067299883666 | 0.00008833617444030 |
| KAT2A | 1.011620550017 | 1.005759115224 | 1.01751614449 | 0.00009744351625077 |
| WDR5 | 1.013310138845 | 1.006111472075 | 1.02056031164 | 0.0002780257813760 |
| SGF29 | 0.987918362000 | 0.9721714361576 | 1.003920351575 | 0.13815570759895 |
| TADA2A | 1.091435629732 | 1.0219442568 | 1.1656523590 | 0.00914274159968211 |
| MBIP | 1.026538959255 | 0.997272245628 | 1.056664556232 | 0.0759186376417354 |
| POLE3 | 1.015709053593 | 1.007923618639 | 1.023554625045 | 0.00007177559650007 |
| DR1 | 1.045826948766 | 1.02340705234625 | 1.0687380004463 | 0.0000506554807811 |
| KAT2B | 1.007472362519 | 0.9921849256014 | 1.022995345979 | 0.339947416964891 |
| ZZZ3 | 1.056264478896 | 1.022611586928 | 1.091024846226 | 0.0009215337124845 |
Table 3.
Chi-squared statistic—TCGA
| Variable | Groups | N | Low | % | High | % | χ | p |
|---|---|---|---|---|---|---|---|---|
| Age (years) | ≤60 | 177 | 77 | 43.50% | 100 | 56.50% | ||
| >60 | 192 | 108 | 56.25% | 84 | 43.75% | 6 | 0.01442 | |
| Gender | Male | 249 | 132 | 53.01% | 117 | 46.99% | ||
| Female | 120 | 53 | 44.17% | 67 | 55.83% | 2.5 | 0.111399 | |
| Race | White | 183 | 94 | 51.37% | 89 | 48.63% | ||
| Asian | 157 | 72 | 45.86% | 85 | 54.14% | |||
| Black | 17 | 10 | 58.82% | 7 | 41.18% | 1.7 | 0.433277 | |
| NA | 12 | |||||||
| BMI | ≤26 | 202 | 96 | 47.52% | 106 | 52.48% | ||
| >26 | 151 | 81 | 53.64% | 70 | 46.36% | 1.3 | 0.255394 | |
| NA | 16 | |||||||
| Family history | No | 207 | 91 | 43.96% | 116 | 56.04% | ||
| Yes | 111 | 59 | 53.15% | 52 | 46.85% | 2.4 | 0.117544 | |
| NA | 51 | |||||||
| TNM Stage | I | 170 | 97 | 57.06% | 73 | 42.94% | ||
| II | 85 | 40 | 47.06% | 45 | 52.94% | |||
| III | 85 | 35 | 41.18% | 50 | 58.82% | |||
| IV | 5 | 2 | 40% | 3 | 60% | 6.5 | 0.089512 | |
| NA | 24 | |||||||
| T classification | T1 | 180 | 102 | 56.67% | 78 | 43.33% | ||
| T2 | 93 | 42 | 45.16% | 51 | 54.84% | |||
| T3 | 80 | 33 | 41.25% | 47 | 58.75% | |||
| T4 | 13 | 5 | 38.46% | 8 | 61.54% | 7.2 | 0.065714 | |
| TX | 1 | |||||||
| NA | 2 | |||||||
| N classification | N0 | 251 | 124 | 49.40% | 127 | 50.60% | ||
| N1 | 4 | 0 | 0% | 4 | 100% | |||
| NX | 113 | 61 | 53.98% | 52 | 46.02% | 4.7 | 0.09339 | |
| NA | 1 | |||||||
| M classification | M0 | 265 | 136 | 51.32% | 129 | 48.68% | ||
| M1 | 4 | 2 | 50% | 2 | 50% | |||
| MX | 100 | 47 | 47% | 53 | 53% | 0.54 | 0.76254 | |
| Histologic grade | G1 | 55 | 39 | 70.91% | 16 | 29.09% | ||
| G2 | 177 | 101 | 57.06% | 76 | 42.94% | |||
| G3 | 120 | 39 | 32.50% | 81 | 67.50% | |||
| G4 | 12 | 4 | 33.33% | 8 | 66.67% | 29 | 2.06E-06 | |
| NA | 5 | |||||||
| AJCC staging | 4th | 4 | 4 | 100% | 0 | 0% | ||
| 5th | 21 | 12 | 57.14% | 9 | 42.86% | |||
| 6th | 117 | 50 | 42.74% | 67 | 57.26% | |||
| 7th | 227 | 119 | 52.42% | 108 | 47.58% | 7.4 | 0.05941 | |
| Child-Pugh | A | 215 | 120 | 55.81% | 95 | 44.19% | ||
| B | 21 | 11 | 52.38% | 10 | 47.62% | 0.091 | 0.762535 | |
| C | 1 | |||||||
| NA | 132 | |||||||
| Vascular invasion | None | 205 | 112 | 54.63% | 93 | 45.37% | ||
| Micro | 92 | 46 | 50% | 46 | 50% | |||
| Macro | 16 | 6 | 37.50% | 10 | 62.50% | 2 | 0.359366 | |
| NA | 56 | |||||||
| Hepatitis virus infection | HBV | 102 | 49 | 48.04% | 53 | 51.96% | ||
| HCV | 53 | 34 | 64.15% | 19 | 35.85% | |||
| No | 168 | 77 | 45.83% | 91 | 54.17% | 5.5 | 0.062617 | |
| NA | 46 | |||||||
| AFP level | Normal | 114 | 76 | 66.67% | 38 | 33.33% | ||
| Elevated | 147 | 61 | 41.50% | 86 | 58.50% | 16 | 5.37E-05 | |
| NA | 108 | |||||||
| Fibrosis Score | 0–4 | 132 | 74 | 56.06% | 58 | 43.94% | ||
| 5–6 | 78 | 49 | 62.82% | 29 | 37.18% | 0.92 | 0.336609 | |
| NA | 159 | |||||||
| Relapse | No | 173 | 84 | 48.55% | 89 | 51.45% | ||
| Yes | 95 | 54 | 56.84% | 41 | 43.16% | 1.7 | 0.194109 | |
| NA | 101 | |||||||
| YEATS2 | high | 184 | 0 | 0 | 184 | 100% | ||
| low | 185 | 185 | 100% | 0 | 0 | 370 | 3.09E-82 |
Table 4.
Chi-squared statistic—GSE14520
| Variable | Groups | N | Low | % | High | % | χ | p | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (years) | ≤60 | 181 | 87 | 48.1% | 94 | 51.9% | |||||
| >60 | 40 | 24 | 60.0% | 16 | 40.0% | 1.9 | 0.1719073 | ||||
| Gender | Male | 191 | 97 | 50.8% | 94 | 49.2% | |||||
| Female | 30 | 14 | 46.7% | 16 | 53.3% | 0.18 | 0.6748924 | ||||
| TNM Stage | I | 93 | 55 | 59.1% | 38 | 40.9% | |||||
| II | 77 | 41 | 53.3% | 36 | 46.8% | ||||||
| III | 49 | 14 | 28.6% | 35 | 71.4% | 12 | 0.0020013 | ||||
| NA | 2 | ||||||||||
| BCLC stage | A | 148 | 83 | 56.1% | 65 | 43.9% | |||||
| B | 22 | 8 | 36.4% | 14 | 63.6% | ||||||
| C | 29 | 9 | 31.0% | 20 | 69.0% | 8 | 0.0183785 | ||||
| NA | 22 | ||||||||||
| AFP level | >300 | 100 | 27 | 27.0% | 73 | 73.0% | 6 | ||||
| ≤ 300 | 118 | 81 | 68.6% | 37 | 31.4% | 38 | 8.90E-10 | ||||
| NA | 3 | ||||||||||
| Relapse | No | 100 | 51 | 51.0% | 49 | 49.0% | |||||
| Yes | 121 | 60 | 49.6% | 61 | 50.4% | 0.044 | 0.8343372 | ||||
| Tumor size | > 5 cm | 80 | 30 | 37.5% | 50 | 62.5% | |||||
| ≤ 5 cm | 140 | 80 | 57.1% | 60 | 42.9% | 7.9 | 0.005062 | ||||
| NA | 1 | ||||||||||
| YEATS2 | high | 110 | 0 | 0.0% | 110 | 100.0% | |||||
| low | 111 | 111 | 100% | 0 | 0 | 220 | 5.47E-50 | ||||
Validation of the prognostic signature of YEATS2
Based on the TCGA database, we verified the expression of YEATS2 in 374 HCC tissues and 50 normal samples (Fig. 1 D) and also detected the prognosis of 364 HCC patients with high expression of YEATS2 (n = 118) and low expression of YEATS2 (n = 246) (Fig. 1 E). The results show that the expression of YEATS2 in liver cancer tissues was significantly higher than that in normal samples (P < 0.0001), and patients with high expression of YEATS2 in liver cancer patients had a worse prognosis, suggesting that YEATS2 plays the role of oncogene in liver cancer. Similarly, we did the same validation in the GEO database (Fig. 1 G–H) and ICGC database (Fig. 1 J–K). To further verify the accuracy of the results, we performed ROC curve verification of the prognosis results in these databases, and the score of AUC (Area Under Curve) in TCGA, GEO and ICGC databases were 0.893 0.888 and 0.878, respectively (P < 0.05) (Fig. 1 F, I, L), highlighting a substantially effective predictive performance of the signature for HCC prognosis.
The prognostic characteristics of YEATS2 correlated with HCC survival
To further analyze the relationship between the expression of YEATS2 and the prognosis in HCC patients with different clinical-pathological characteristics, we first classified HCC patients according to different characteristics from TCGA databases, including HCV infection, Histological grade, Vital status, Age, Vascular invasion, TNM stage, Fibrosis score and HBV infection (Fig. 2 A–H). According to the TCGA database, we can find that there was no significant difference between the expression of YEATS2 and that of vascular invasion, fibrosis score and HBV infection (Fig. 2 E, G–H), but there was a significant difference between the expression of YEATS2 and that of HCV infection, Histological grade, Age, Vital status, and the TNM stage (Fig. 2 A–D, F). Survival analysis showed that the differences in prognosis of HCC were statistically significant regardless of AFP level, Tumor staging, Patient’s gender, age, sex, and tumor status (Fig. 2 I–X). These results suggest that HCC patients with high expression of YEATS2 have a worse prognosis. We did the same thing with GEO databases as TCGA. Differences were analyzed in different groups including AFP level, Age, Tumor size, Vital status, Histological grade, TNM stage, Gender and HBV infection, showing that there was no significant difference between the expression of YEATS2 and that of the patient’s gender and HBV infection, but there was a significant difference between the expression of YEATS2 and that of the AFP level, tumor stage, tumor size, and vital status (Fig. 3 A–H). The expression of YEATS2 in patients with AFP > 400 was higher than that in patients with AFP ≤ 400, suggesting that YEATS2 could be used as an auxiliary detection indicator of AFP. Survival analysis showed that the differences in prognosis of HCC were statistically significant regardless of gander, tumor stage, age, and tumor size (Fig. 3 I–S). The results of TCGA database are further verified. In addition, it can be seen from Fig. 6 A that the prediction accuracy of our model in TNM staging and pathological differentiation is well.
Fig. 2.
TCGA database was used to analyze the correlation between the prognostic characteristics of YEATS2 and HCC survival rate. A–H Statistical analysis of YEATS2 expression in different clinical pathological features of TCGA database (HCV infection (A), Histological grade (B), Vital status (C), Age (D), Vascular invasion (E), TNM stage (F), Fibrosis score (G) and HBV infection (H)). I–J Survival analysis of YEATS2 expression in HCC patients with elevated and normal AFP of TCGA database. The survival of YEATS2 high expression was no statistical difference with YEATS2 low expression in elevated and normal AFP patients (P = 0.17). K, O Survival analysis of YEATS2 expression in HCC patients with vascular invasion and no vascular invasion of TCGA database. The survival of YEATS2 high expression was worse than those with YEATS2 low expression (P = 0.04). L, P Survival analysis of YEATS2 expression in HCC patients with alcohol consumption and no alcohol of TCGA database. The survival of YEATS2 with high expression was significantly worse than that with YEATS2 low expression (P = 0.01). M–N Survival analysis of YEATS2 expression in HCC patients in relation to gender. The survival of YEATS2 with high expression was significantly worse than that with YEATS2 low expression regardless of the gender of the patient. Q–S Survival analysis of YEATS2 expression in HCC tissues of stage I to III (TNM stage). The survival of patients with YEATS2 high expression was significantly shorter than that of patients with YEATS2 low expression. T, X Survival analysis of YEATS2 expression in HCC patients in relation to age. The survival of YEATS2 with high expression was significantly worse than that with YEATS2 low expression regardless of the age of the patient. U Survival analysis of YEATS2 expression in HCC patients with no HBV infection. The survival of YEATS2 with high expression was significantly worse than that with YEATS2 low expression (P = 0.008). V, W Survival analysis of YEATS2 expression in HCC patients with Hepatitis C and no HCV infection. The survival of YEATS2 with high expression was also significantly worse than that with YEATS2 low expression regardless of the HCV infection
Fig. 3.
GEO database was used to analyze the clinicopathologic characteristics and the correlation between the prognostic characteristics of YEATS2 and HCC survival rate. A–H Statistical analysis of YEATS2 expression in different clinical pathological features of GEO database (AFP level (A), Age (B), Tumor size (C), Vital status (D), Histological grade (E), TNM stage (F), Gender (G) and HBV infection (H)). I–S Survival analysis of YEATS2 expression in HCC patients with different clinical pathological features of GEO database (AFP level (I–J), Age (K–L), Gender (M–N), TNM stage (O–Q) and Tumor size (R–S))
Fig. 6.
YEATS2 deletion slowed cell proliferation, colony formation, and transwell formation. A Western-blot analysis of the protein level of YEATS2 in LM3 knockdown YEATS2 cell lines. B RT-qPCR analysis of the mRNA level of YEATS2 in LM3 knockdown YEATS2 cell lines. C CCK8 assay analysis of the cell viability in LM3 cells with stably knockdown YEATS2. D–E Colony formation assay analysis of the cell colony formation in LM3 knockdown YEATS2 cell lines. F–H Transwell assay analysis of the migration and invasion in LM3 knockdown YEATS2 cell lines
Cox proportional hazards regression analysis
Next, we evaluated the effect of YEATS2 prognostic characteristics on OS in HCC patients by univariate and multivariate Cox regression. For the whole TCGA cohort, univariate and multivariate Cox regression analysis showed that age, gender, race, BMI, family history, TNM stage, AJCC grade, AFP level, and other pathological subclasses were significantly correlated with the prognosis of HCC (P < 0.05) (Table 5). Based on data extraction from the GSE14520 database, univariate Cox regression analysis showed that BCLC Stage, TNM Stage, AFP level, receptors, tumor size, and YEATS2 expression were significantly correlated with HCC survival (all P < 0.05) (Table 6). The corresponding multivariate Cox regression analysis showed that there were differences in the expression of prognosis in patients with high BCLC stage, low expression of falcon, and YEATS2 in liver cancer patients (all P < 0.05) (Tables 5–6). Therefore, YEATS2 can indeed be used as an independent indicator to test the prognosis of HCC.
Table 5.
Univariate multifactor analysis—TCGA
| Variable | Groups | N | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |||
| Age (years) | ≤60 | 180 | Reference | |||
| >60 | 197 | 0.89 (0.63–1.3) | 5.25E-01 | |||
| Gender | Female | 120 | Reference | |||
| Male | 252 | 1.5 (1.0–2.1) | 3.16E-02 | 1.5 (1–2.3) | 4.37E-02 | |
| Race | Asian | 159 | Reference | |||
| White | 184 | 0.39 (0.26–0.58) | 5.51E-06 | |||
| Black | 17 | 0.55 (0.24–1.24) | 1.53E-01 | |||
| NA | 12 | |||||
| BMI | ≤26 | 204 | Reference | |||
| >26 | 151 | 0.83 (0.58–1.2) | 2.99E-01 | |||
| NA | 17 | |||||
| Family history | No | 209 | Reference | |||
| Yes | 111 | 0.83 (0.57–1.2) | 3.23E-01 | |||
| NA | 52 | |||||
| TNM Stage | I | 172 | Reference | |||
| II | 86 | 1.1 (0.81–1.8) | 4.63E-02 | |||
| III | 85 | 1.2 (0.71–1.6) | 3.27E-02 | |||
| IV | 5 | 1.4 (0.34–3.6) | 9.21E-02 | |||
| NA | 24 | |||||
| T classification | T1 | 182 | Reference | |||
| T2 | 94 | 1 (0.64–1.6) | 9.36E-01 | |||
| T3 | 80 | 1.2 (0.76–1.7) | 5.04E-01 | |||
| T4 | 13 | 1.5 (0.77–3.1) | 2.23E-01 | |||
| TX | 1 | 0.95 (0.13–6.9) | 9.59E-01 | |||
| NA | 2 | |||||
| N classification | N0 | 253 | Reference | |||
| N1 | 4 | 0.86 (0.21–3.5) | 8.39E-01 | |||
| NX | 114 | 0.73 (0.51–1.1) | 9.91E-02 | |||
| NA | 1 | |||||
| M classification | M0 | 267 | Reference | |||
| M1 | 4 | 1.1 (0.35–3.5) | 8.53E-01 | |||
| MX | 101 | 0.92 (0.64–1.3) | 6.77E-01 | |||
| Histologic grade | G1 | 55 | Reference | |||
| G2 | 178 | 1.1 (0.63–1.8) | 7.87E-01 | 1.2 (0.7–2.2) | 4.66E-01 | |
| G3 | 122 | 1.8 (1–3.2) | 3.83E-02 | 2 (1.1–3.7) | 2.41E-02 | |
| G4 | 12 | 2.6 (0.95–7.3) | 6.29E-02 | 3.5 (1.2–10) | 1.75E-02 | |
| NA | 5 | |||||
| AJCC stage | 4th | 4 | Reference | |||
| 5th | 21 | 4.5 (1–20) | 4.92E-02 | |||
| 6th | 119 | 6.5 (1.5–28) | 1.17E-02 | |||
| 7th | 228 | 9.2 (2.1–40) | 3.39E-03 | |||
| Child-Pugh | A | 217 | Reference | |||
| B | 21 | 1.5 (0.71–3.2) | 2.84E-01 | |||
| C | 1 | 0.45 (0.061–3.3) | 4.29E-01 | |||
| NA | 132 | |||||
| Vascular invasion | Macro | 16 | Reference | |||
| None | 207 | 0.65 (0.31–1.4) | 2.55E-01 | |||
| Micro | 93 | 0.73 (0.33–1.6) | 4.32E-01 | |||
| Hepatitis virus infection | HBV | 104 | Reference | |||
| HCV | 53 | 0.8 (0.42–1.5) | 4.95E-01 | |||
| No | 169 | 0.86 (0.51–1.4) | 5.55E-01 | |||
| NA | 46 | |||||
| AFP level | Elevated | 147 | Reference | |||
| Normal | 116 | 0.88 (0.54–1.4) | 6.02E-01 | |||
| NA | 109 | |||||
| Fibrosis Score | 0–4 | 133 | Reference | |||
| 5–6 | 80 | 1.6 (0.94–2.9) | 8.24E-02 | |||
| NA | 159 | |||||
| Relapse | No | 173 | Reference | |||
| Yes | 98 | 0.37 (0.22–0.61) | 9.56E-05 | |||
| NA | 101 | |||||
| YEATS2 | High | 185 | Reference | |||
| Low | 187 | 0.55 (0.38–0.78) | 8.63E-04 | 0.51 (0.34–0.78) | 1.66E-03 | |
Table 6.
Univariate multifactor analysis—GSE14520
| Variable | Groups | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | ||
| Age (years) | ≤60 | Reference | |||
| >60 | 0.97 (0.55–1.7) | 0.90342631 | |||
| Gender | Male | Reference | |||
| Female | 0.59 (0.28–1.2) | 0.15328585 | |||
| BCLC Stage | A | Reference | |||
| B | 2.2 (1.1–4.1) | 0.01785427 | 0.83 (0.39–1.7) | 0.62206595 | |
| C | 4.3 (2.5–7.2) | 4.16E-08 | 2.7 (1.2–5.9) | 0.01265396 | |
| TNM | I | Reference | |||
| II | 2.1 (1.2–3.7) | 0.00657822 | 1.5 (0.83–2.7) | 0.17722214 | |
| III | 5.2 (3–9.1) | 8.03E-09 | 1.8 (0.8–4.2) | 0.15488902 | |
| AFP level | >300 | Reference | |||
| ≤ 300 | 0.61 (0.4–0.94) | 0.02504997 | 0.76 (0.45–1.3) | 0.30088401 | |
| Relapse | No | Reference | |||
| Yes | 110 (16–820) | 2.59E-06 | 110 (16–830) | 2.78E-06 | |
| Tumor size | >5 cm | Reference | |||
| ≤5 cm | 0.52 (0.34–0.8) | 0.00279892 | 1.1 (0.58–1.9) | 0.85805588 | |
| YEATS2 | high | Reference | |||
| Low | 0.48 (0.31–0.75) | 0.00114863 | 0.42 (0.25–0.71) | 0.00105884 | |
Establishment and validation of a predictive nomogram
Nomogram was formed by incorporating all significant prognostic factors based on multivariate Cox regression analysis to predict OS in 342 patients of HCC with the TGGA database. Specifically, Relapse and predictive risk scores had the greatest impact on the prognosis of HCC. The c-index of the nomogram related to TCGA is 0.745 (95% CI: 0.676–0.816), which highlights the good predictive value of our nomogram model.
YEATS2 is high expressed in liver cancer clinical sample
In order to confirm the expression of YEATS2 in clinical samples of HCC, we randomly selected 12 pairs of total protein and 24 pairs of total RNA from tumor and adjacent tissues of HCC patients and detected the expression of YEATS2 protein and mRNA by Western blot and RT-qPCR (Fig. 4 A–C). The results showed that the expression of YEATS2 in HCC tissues was significantly higher than that in adjacent tissues, and the protein level of AFP was also significantly higher in tumor tissues than in adjacent tissues, which was consistent with the expression of YEATS2, and the protein level of YEAST2 was more different than that of AFP. In addition, we examined clinical tissue samples from HCC patients using immunohistochemistry, and we found that the IHC score for YEATS2 was higher in HCC tissues compared to adjacent normal tissues (Fig. 4 E–F). Meanwhile, the results are consistent with the immunohistochemical from The Human Protein atlas database(https://www.proteinatlas.org/) (Fig. 4 D). These results indicate that YEATS2 can be used as a risk factor in liver cancer models to promote tumor growth.
Fig. 4.
YEATS2 is a novel oncogene. A Western-blot analysis of the protein level of YEATS2 in HCC patient’s tumor and adjacent tissues. B Western-blot analysis of the protein level of YEATS2 and AFP in HCC patient’s tumor and adjacent tissues. C RT-qPCR analysis of the mRNA level of YEATS2 in HCC patient’s tumor and adjacent tissues. D Two HCC patients of YEATS2 expression in normal and tumor tissues from The Human Protein atlas database. E Histograms of immunohistochemical results about YEATS2 from 20 clinical samples. F Three HCC patients of YEATS2 expression in normal and tumor tissues
Preliminary study on the biological function of YEATS2
In order to verify the proliferation and migration ability of YEATS2 at the cellular level, we constructed a cell line that stably overexpression YEATS2 in LM3 and Hep3B cells. CCK8 assay, colony formation assay, transwell migration and invasion assay confirmed that YEATS2 promoted cell proliferation, colony formation and cell migration ability (Fig. 5A–H). After knockdown YEATS2 in LM3 cells (Fig. 6A–B). Through the results of CCK8 assays, colony formation experiments, and transwell assays, it can be found that after knocking down YEATS2, the proliferation and migration ability of cells and the cells proliferate was reduced compared to the control group (Fig. 6C–H). These results suggest that YEATS2 may be a novel oncogene in HCC. Eventually, we used single gene GSEA to analyze the potential pathways and biological functions of YEATS2. The data were from the HCC queues in the GSE14520 database and the TCGA database (Fig. 7B–C). As shown in Fig. 7, YEATS2 was significantly enriched in biological processes of DNA repair and cell cycle in GEO and KEGG analysis. There were statistically significant differences in cell cycle, cell cycle checkpoint, cell cycle process, recombinational repair, DNA repair, and DNA replication (Fig. 7D–I). It is suggested that the gene may function through these two biological pathways, which provides a very important thought for our further study.
Fig. 5.
YEATS2 overexpression promote cell proliferation, colony formation, and transwell formation. A–B CCK8 assay analysis of the cell viability in LM3 and Hep3B cells with stably overexpression YEATS2. C–D Colony formation assay analysis of the cell colony formation in LM3 and Hep3B overexpression YEATS2 cell lines. E–H Transwell assay analysis of the migration and invasion in LM3 nd Hep3B overexpression YEATS2 cell lines
Fig. 7.
YEATS2 is involved in the DNA damage pathway. A Establishment and validation of a predictive nomogram. B, C KEGG and GO enrichment analysis by GSEA algorithm. D–I GSEA of the whole transcriptome data in YEATS2 overexpressed cells were enriched in cell cycle (D–E, H), and DNA replication (F–G, I) form TCGA and GO database
YEATS2 deficiency increased the cells more sensitive to DNA damage
Previous studies have confirmed that YEATS2 can serve as a new treatment for liver cancer and prognosis markers to promote progress in liver cancer cells, and is associated with DNA damage repair pathways. To further confirm this prediction, we examined the protein expression of γH2AX after YEATS2 overexpression and silencing in YEATS2 overexpression and knockdown HCC cell lines after Doxorubicin -induced DNA damage, western blot results showed that YEATS2 overexpression inhibited Dox-induced DNA damage, while YEATS2 knockdown increased the sensitivity of cells to DNA damage (Fig. 8A–E). We also performed colony formation assays on LM3 cells knockdown YEATS2 after radiation irradiation, results show that YEATS2 deficiency makes the HCC radiotherapy sensitization (Fig. 8F–G). To determine the radiation YEATS2 tumor inhibitory effect of the problem, we use LM3 cells in the nude mice to establish the subcutaneous xenograft model. When the tumor has reached around 100 mm3 for radiotherapy(Fig. 7H–J). Results show that irradiation can inhibit tumor growth, and the effect of YEATS2 deficiency is more significant, which means that the deficiency of YEATS2 is more vulnerable to the effects of radiation therapy. Taken together, YEATS2 can be used as a novel oncogene to inhibit DNA damage in liver cancer cells during radiation-induced DNA damage response, thereby promoting the occurrence and development of tumors. These results suggest that the deletion of YEATS2 enhances the radiotherapy-induced anti-tumor effect, which may provide a new approach to cancer therapy.
Fig. 8.
YEATS2 deficiency made the cells more sensitive to DNA damage. A–B RT-qPCR analysis of the mRNA level of YEATS2 in LM3 and Hep3B overexpression YEATS2 cell lines. C–D Western-blot analysis of the protein level of γ-H2AX after treat with Dox in LM3 and Hep3B overexpression YEATS2 cell lines. E Western-blot analysis of the protein level of γ-H2AX after treat with Dox in LM3 knockdown YEATS2 cell lines. F–G Colony formation assays analysis the colony formation ability after radiation on plates at several time points of LM3 cells that were stably knocking down YEATS2. H–I Subcutaneous tumors of LM3 cells respond to the specified therapies. Each group has six mice. Tumor formation was tracked at the times indicated (H), weight (I), and pictures (J) are shown
Discussion
More and more studies have shown that genetic changes and defects in signaling pathways play a crucial role in the development of HCC [34, 35], suggesting the potential predictive value of molecular markers in the prognosis of HCC. Currently, single-parameter markers based on abnormal mRNA levels have attracted much attention and shown promising prognostic potential for HCC. For instance, DDX17 and PXN-AS1-IR3 may be potential prognostic and immunotherapeutic biomarkers of HCC targeting bio-membrane lipid remodeling [36]. Fucosyltransferase 1 (FUT1) overexpression is a poor prognostic indicator for HCC [37]. A growing number of studies, including epigenetics, metabolism, and immunology, are being conducted based on reducing morbidity and mortality and improving prognosis in HCC patients [38–40].
Our lab has been engaged in the direction of epigenetic regulation for a long time [41–43], epigenetic histone modifiers are key regulators in determining the fate of tumor and normal cells [44, 45]. Therefore, we wanted to find the marker molecules that could regulate the prognosis of liver cancer from the perspective of apparent. To further identify the molecules that were differentially expressed, we first conducted a genome-wide enrichment analysis on the TCGA liver cancer data. We discovered that the differential genes had a high enrichment of genes related to the Ada-two-A-containing (ATAC) histone acetyltransferase complex. Following that, the ATAC-complex gene was created using the GO database entries. The ATAC-complex signature was then scored using GSVA in both normal and tumor samples from the TCGA database. It was discovered that the expression of ATAC-complex was significantly higher in tumors. Each gene of the signature was then examined, first determining the difference, only two of them have differential expression: MAP3K and YEATS2. We chose YEATS2 for additional research because it had the highest HR in the COX analysis.
In conclusion, a novel signature for prognostic prediction in HCC was established, with higher risk scores implying an unfavorable prognosis. A nomogram model integrating the signature with additional significant clinicopathological parameters also yielded promising predictive performance in HCC survival.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgment
This manuscript has been read and approved by all authors for publication and has not been submitted and is not under consideration for publication elsewhere. We would like to thank all laboratory members for their critical discussion of this manuscript and to apologize to those not mentioned due to space limitations.
Abbreviations
- HCC
Hepatocellular carcinoma
- TCGA
The Cancer Genome Atlas
- GEO
Gene Expression Omnibus
- ICGC
The International Cancer Genome Consortium
- DEmRNAs
Differentially expressed mRNAs
- ROC
Receiver operating characteristic
- OS
Overall survival
- TNM
Tumor-node-metastasis
- AFP
Alpha-fetoprotein
- HR
Hazard ratio
- C-index
Concordance index
- AUC
Under the curve
- HBV
Hepatitis B virus
- HCV
Hepatitis C virus
Author contributions
Conception and design: Y. Tao, X. Wang. Development of the methodology: Y. Long, W. Wang. Acquisition of the data: Y. Long, W. Wang. Analysis and interpretation of the data (e.g., statistical analysis, biostatistics, computational analysis): Y. Long, W. Wang, S. Liu. Writing, review, and/or revision of the manuscript: Y. Long, Y. Tao. Study supervision: Y. Tao, X. Wang
Funding
Supported by the Fundamental Research Funds for the Central Universities of Central South University [2024ZZTS0525(Y.Long)].
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
The ethics committee of Cancer Research Institute of Central South University has approved this study.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Yao Long and Wei Wang contributed equally to this work.
Contributor Information
Xiang Wang, Email: wangxiang@csu.edu.cn.
Yongguang Tao, Email: taoyong@csu.edu.cn.
References
- 1.W. Chen, R. Zheng, P.D. Baade, S. Zhang, H. Zeng, F. Bray, A. Jemal, X.Q. Yu, J. He, Cancer statistics in China, 2015. CA Cancer J. Clin. 66(2), 115–132 (2016) [DOI] [PubMed] [Google Scholar]
- 2.R.L. Siegel, K.D. Miller, H.E. Fuchs, A. Jemaal, Cancer statistics, 2021. CA Cancer J. Clin. 71(4), 359 (2021) 10.3322/caac.21654
- 3.F. Bray, J. Ferlay, I. Soerjomataram, R.L. Siegel, L.A. Torre, A. Jemal, Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 68(6), 394–-424 (2018) [DOI] [PubMed] [Google Scholar]
- 4.J.F. Shi, M. Cao, Y. Wang, F.Z. Bai, L. Lei, J. Peng, E. Feletto, K. Canfell, C. Qu, W. Chen, Is it possible to halve the incidence of liver cancer in China by 2050? Int. J. Cancer 148(5), 1051–1065 (2021) [DOI] [PubMed] [Google Scholar]
- 5.R.L. Siegel, K.D. Miller, A. Jemal, Cancer statistics, 2019. CA Cancer J. Clin. 69(1), 7–34 (2019). [DOI] [PubMed] [Google Scholar]
- 6.E. Guttman-Yassky, A.B. Pavel, A. Diaz, N. Zhang, E. Del Duca, Y. Estrada, B. King, A. Banerjee, C. Banfield, L.A. Cox et al., Ritlecitinib and brepocitinib demonstrate significant improvement in scalp alopecia areata biomarkers. J. Allergy Clin. Immunol. (2021) [DOI] [PubMed]
- 7.Y. Yu, X.D. Peng, X.J. Qian, K.M. Zhang, X. Huang, Y.H. Chen, Y.T. Li, G.K. Feng, H.L. Zhang, X.L. Xu et al., Fis1 phosphorylation by Met promotes mitochondrial fission and hepatocellular carcinoma metastasis. Signal Transduct. Target Ther. 6(1), 401 (2021) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hepatocellular carcinoma. Nat. Rev. Dis. Primers. 7(1), 7 (2021) [DOI] [PubMed]
- 9.G. Semmler, E.L. Meyer, K. Kozbial, P. Schwabl, S. Hametner-Schreil, A. Zanetto, D. Bauer, D. Chromy, B. Simbrunner, B. Scheiner et al., HCC risk stratification after cure of hepatitis C in patients with compensated advanced chronic liver disease. J. Hepatol. (2021) [DOI] [PubMed]
- 10.M. Sanduzzi-Zamparelli, Z. Marino, S. Lens, V. Sapena, G. Iserte, A. Pla, N. Granel, C. Bartres, N. Llarch, R. Vilana et al., Liver cancer risk after HCV cure in patients with advanced liver disease without non-characterized nodules. J. Hepatol. (2021) [DOI] [PubMed]
- 11.W. Tang, Z. Chen, W. Zhang, Y. Cheng, B. Zhang, F. Wu, Q. Wang, S. Wang, D. Rong, F.P. Reiter et al., The mechanisms of sorafenib resistance in hepatocellular carcinoma: theoretical basis and therapeutic aspects. Signal Transduct. Target Ther. 5(1), 87 (2020) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.M. Reig, A. Forner, J. Rimola, J. Ferrer-Fabrega, M. Burrel, A. Garcia-Criado, R.K. Kelley, P.R. Galle, V. Mazzaferro, R. Salem, et al., BCLC strategy for prognosis prediction and treatment recommendation Barcelona Clinic Liver Cancer (BCLC) staging system. The 2022 update. J. Hepatol. (2021) [DOI] [PMC free article] [PubMed]
- 13.Z. Zhang, H. Xie, P. Chen, P. Cao: Development and Identification of a Nomogram Prognostic Model for Patients with Primary Clear Cell Carcinoma of the Liver. Med. Sci. Monit. 26, e919789 (2020) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.A. Budhu, H.L. Jia, M. Forgues, C.G. Liu, D. Goldstein, A. Lam, K.A. Zanetti, Q.H. Ye, L.X. Qin, C.M. Croce et al., Identification of metastasis-related microRNAs in hepatocellular carcinoma. Hepatology (Baltimore, Md) 47(3), 897–907 (2008) [DOI] [PubMed] [Google Scholar]
- 15.J. Bruix, M. Reig, M. Sherman, Evidence-based diagnosis, staging, and treatment of patients with hepatocellular carcinoma. Gastroenterology 150(4), 835–853 (2016). [DOI] [PubMed] [Google Scholar]
- 16.M.Y. Chan, W.H. She, W.C. Dai, S.H.Y. Tsang, K.S.H. Chok, A.C.Y. Chan, J. Fung, C.M. Lo, T.T. Cheung, Prognostic value of preoperative alpha-fetoprotein (AFP) level in patients receiving curative hepatectomy- an analysis of 1,182 patients in Hong Kong. Translat. Gastroenterol. Hepatol. 4, 52 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.W. Mi, H. Guan, J. Lyu, D. Zhao, Y. Xi, S. Jiang, F.H. Andrews, X. Wang, M. Gagea, H. Wen et al., YEATS2 links histone acetylation to tumorigenesis of non-small cell lung cancer. Nat. Commun. 8(1), 1088 (2017) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.M. Moustakim, T. Christott, O.P. Monteiro, J. Bennett, C. Giroud, J. Ward, C.M. Rogers, P. Smith, I. Panagakou, L. Díaz-Sáez et al. Discovery of an MLLT1/3 YEATS Domain Chemical Probe. Angew. Chem. (International Ed in English) 57(50), 16302–16307 (2018) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.D. Zhao, H. Guan, S. Zhao, W. Mi, H. Wen, Y. Li, Y. Zhao, C.D. Allis, X. Shi, H. Li, YEATS2 is a selective histone crotonylation reader. Cell Res. 26(5), 629–632 (2016) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.P. Yeetong, M. Pongpanich, C. Srichomthong, A. Assawapitaksakul, V. Shotelersuk, N. Tantirukdham, C. Chunharas, K. Suphapeetiporn, V. Shotelersuk, TTTCA repeat insertions in an intron of YEATS2 in benign adult familial myoclonic epilepsy type 4. Brain 142(11), 3360–3366 (2019) [DOI] [PubMed] [Google Scholar]
- 21.H.C. Beird, C.C. Wu, D.R. Ingram, W.L. Wang, A. Alimohamed, C. Gumbs, L. Little, X. Song, B.W. Feig, C.L. Roland et al., Genomic profiling of dedifferentiated liposarcoma compared to matched well-differentiated liposarcoma reveals higher genomic complexity and a common origin. Cold Spring Harbor Mol. Case Stud. 4(2), (2018) [DOI] [PMC free article] [PubMed]
- 22.Y. Yamada, Y. Yasukochi, K. Kato, M. Oguri, H. Horibe, T. Fujimaki, I. Takeuchi, J. Sakuma, Identification of 26 novel loci that confer susceptibility to early-onset coronary artery disease in a Japanese population. Biomed. Rep. 9(5), 383–404 (2018) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.X. Ren, Y. Zhou, Z. Xue, N. Hao, Y. Li, X. Guo, D. Wang, X. Shi, H. Li, Histone benzoylation serves as an epigenetic mark for DPF and YEATS family proteins. Nucleic Acids Res. 49(1), 114–126 (2021) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Z. Zeng, S. Lei, Z. He, T. Chen, J. Jiang, YEATS2 is a target of HIF1alpha and promotes pancreatic cancer cell proliferation and migration. J. Cell Physiol. 236(3), 2087–2098 (2021) [DOI] [PubMed] [Google Scholar]
- 25.O. Jimenez Martin, A. Schlosser, R. Furtwangler, J. Wegert, M. Gessler, MYCN and MAX alterations in Wilms tumor and identification of novel N-MYC interaction partners as biomarker candidates. Cancer Cell Int. 21(1), 555 (2021) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.H. Sheng, F. Zheng, T. Lan, H.F. Chen, C.Y. Xu, S.W. Wang, Y.Y. Weng, L.F. Xu, F. Zhang. YEATS2 regulates the activation of TAK1/NF-kappaB pathway and is critical for pancreatic ductal adenocarcinoma cell survival. Cell Biol. Toxicol. (2021) [DOI] [PubMed]
- 27.J.C. Nault, P.R. Galle, J.U. Marquardt. The role of molecular enrichment on future therapies in hepatocellular carcinoma. J. Hepatol. 69(1), 237–247 (2018) [DOI] [PubMed] [Google Scholar]
- 28.M. Mao, X. Wang, H. Sheng, Y. Liu, L. Zhang, S. Dai, P.D. Chi, A novel score based on serum apolipoprotein A-1 and C-reactive protein is a prognostic biomarker in hepatocellular carcinoma patients. BMC Cancer 18(1), 1178 (2018) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.B. Yu, Y. Ding, X. Liao, C. Wang, B. Wang, X. Chen, Overexpression of PARPBP correlates with tumor progression and poor prognosis in hepatocellular carcinoma. Dig. Dis. Sci. 64(10), 2878–2892 (2019) [DOI] [PubMed] [Google Scholar]
- 30.C. Luo, J. Pu, F. Liu, X. Long, C. Wang, H. Wei, Q. Tang, MicroRNA-200c expression is decreased in hepatocellular carcinoma and associated with poor prognosis. Clin. Res. Hepatol. Gastroenterol. 43(6), 715–721 (2019) [DOI] [PubMed] [Google Scholar]
- 31.Q. Wu, S. Jiang, T. Cheng, M. Xu, B. Lu, A Novel pyroptosis-related prognostic model for hepatocellular carcinoma. Front. Cell Dev. Biol. 9, 770301 (2021) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.M. Honda, K. Oba, T. Akiyoshi, H. Maeda, K. Kashiwabara, M. Kanda, S. Mayanagi, T. Aoyama, C. Hamada, S. Sadahiro et al., Development and validation of a prognostic nomogram for colorectal cancer after radical resection based on individual patient data from three large-scale phase III trials. Oncotarget 8(58), 99150–99160 (2017) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.G.P. Nagaraju, B. Dariya, P. Kasa, S. Peela, B.F. El-Rayes, Epigenetics in hepatocellular carcinoma. Semin. Cancer Biol. 86(Pt 3), 622–632 (2022) [DOI] [PubMed] [Google Scholar]
- 34.X. Chao, S. Wang, S. Fulte, X. Ma, F. Ahamed, W. Cui, Z. Liu, T. Rulicke, K. Zatloukal, W.X. Zong et al., Hepatocytic p62 suppresses ductular reaction and tumorigenesis in mouse livers with mTORC1 activation and defective autophagy. J. Hepatol. (2021) [DOI] [PMC free article] [PubMed]
- 35.Y. Gao, M. You, J. Fu, M. Tian, X. Zhong, C. Du, Z. Hong, Z. Zhu, J. Liu, G.J. Markowitz et al., Intratumoral stem-like CCR4+ regulatory T cells orchestrate the immunosuppressive microenvironment in HCC associated with hepatitis B. J. Hepatol. (2021) [DOI] [PubMed]
- 36.H.Z. Zhou, F. Li, S.T. Cheng, Y. Xu, H.J. Deng, D.Y. Gu, J. Wang, W.X. Chen, Y.J. Zhou, M.L. Yang et al., DDX17-regulated alternative splicing that produced a novel oncogenic isoform of PXN-AS1 to promote HCC metastasis. Hepatology (2021) [DOI] [PMC free article] [PubMed]
- 37.J.H. Loong, T.L. Wong, M. Tong, R. Sharma, L. Zhou, K.Y. Ng, H.J. Yu, C.H. Li, K. Man, C.M. Lo et al., Glucose deprivation-induced aberrant FUT1-mediated fucosylation drives cancer stemness in hepatocellular carcinoma. J. Clin. Invest. 131(11), (2021) [DOI] [PMC free article] [PubMed]
- 38.T.I. Huo, J.I. Liao, S.Y. Ho, Prognostic prediction for patients with hepatocellular carcinoma receiving immunotherapy: are we there yet? J. Hepatol. (2021) [DOI] [PubMed]
- 39.J. Wang, Y. Wang, Y. Chu, Z. Li, X. Yu, Z. Huang, J. Xu, L. Zheng, Tumor-derived adenosine promotes macrophage proliferation in human hepatocellular carcinoma. J. Hepatol. 74(3), 627–637 (2021) [DOI] [PubMed] [Google Scholar]
- 40.H. Rhee, E.S. Cho, J.H. Nahm, M. Jang, Y.E. Chung, S.E. Baek, S. Lee, M.J. Kim, M.S. Park, D.H. Han et al., Gadoxetic acid-enhanced MRI of macrotrabecular-massive hepatocellular carcinoma and its prognostic implications. J. Hepatol. 74(1), 109–121 (2021) [DOI] [PubMed] [Google Scholar]
- 41.X. Chen, Y. Li, K. Rubio, B. Deng, Y. Li, Q. Tang, C. Mao, S. Liu, D. Xiao, G. Barreto et al. Lymphoid-specific helicase in epigenetics, DNA repair and cancer. Br. J. Cancer 126(2), 165–173 (2022) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.C. Mao, X. Wang, Y. Liu, M. Wang, B. Yan, Y. Jiang, Y. Shi, Y. Shen, X. Liu, W. Lai et al. A G3BP1-interacting lncRNA promotes ferroptosis and apoptosis in cancer via nuclear sequestration of p53. Cancer Res. 78(13), 3484–3496 (2018) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Y. Wu, S. Zhang, X. Gong, S. Tam, D. Xiao, S. Liu, Y. Tao, The epigenetic regulators and metabolic changes in ferroptosis-associated cancer progression. Mol. Cancer 19(1), 39 (2020) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Y. Chen, X. Shen, Y. Tang, Y. Weng, W. Yang, M. Liu, D. Xu, J. Shi, X. Yang, F. Yu et al., The diverse pancreatic tumor cell-intrinsic response to IFNgamma is determined by epigenetic heterogeneity. Cancer Lett. 562, 216153 (2023) [DOI] [PubMed] [Google Scholar]
- 45.J. Ren, B. Ren, X. Liu, M. Cui, Y. Fang, X. Wang, F. Zhou, M. Gu, R. Xiao, J. Bai et al., Crosstalk between metabolic remodeling and epigenetic reprogramming: a new perspective on pancreatic cancer. Cancer Lett. 587, 216649 (2024) [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.