Sage Journals: Discover world-class research

Abstract

Hepatocellular carcinoma (HCC) is a widespread, common type of cancer in Asian countries, and the need for biomarker-matched molecularly targeted therapy for HCC has been increasingly recognized. However, the effective treatment for HCC is unclear. Therefore, identifying additional hub genes and pathways as novel prognostic biomarkers for HCC is necessary. In this study, the expression profiles of GSE121248, GSE45267 and GSE84402 were obtained from the Gene Expression Omnibus (GEO), including 132 HCC and 90 noncancerous liver tissues. Differentially expressed genes (DEGs) between HCC and noncancerous samples were identified by GEO2 R and Venn diagrams. In total, 109 DEGs were identified in these datasets, including 24 upregulated genes and 85 downregulated genes. Subsequently, Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) preliminary analyses of the DEGs were performed using DAVID. The protein–protein interaction (PPI) network of the DEGs was constructed with the Search Tool for the Retrieval of Interacting Genes (STRING) and visualized in Cytoscape. Module analysis of the PPI network was performed using MCODE to get hub genes. Moreover, the influence of the hub genes on overall survival was determined with Kaplan–Meier plotter. All hub genes were analyzed by Gene Expression Profiling Interactive Analysis (GEPIA) and KEGG. Overall, the hub genes DTL, CDK1, CCNB1, RACGAP1, ECT2, NEK2, BUB1B, PBK, TOP2A, ASPM, HMMR, RRM2, CDKN3, PRC1, and ANLN were upregulated in HCC, and the survival rate was lower for HCC with increased expression of these hub genes. CCNB1, CDK1, and RRM2 were enriched in the p53 signaling pathway, and CCNB1, CDK1, and BUB1B were enriched in the cell cycle. In brief, we screened 15 hub genes and pathways to identify potential prognostic markers for HCC treatment. However, the specific occurrence and development of HCC with expression of the hub genes should be verified in vivo and in vitro.

Keywords

hepatocellular carcinoma bioinformatics analysis hub genes prognostic biomarkers

Introduction

Hepatocellular carcinoma (HCC) is a lethal malignancy of the liver. Similar to other cancers, HCC is associated with potentially alterable risk factors, such as excess body weight, alcohol abuse, hepatitis B virus (HBV), hepatitis C virus, nonalcohol fatty liver disease, and certain genes^1,2. Most cases of HCC (80%) occur in Asian countries because of chronic HBV infection and exposure to aflatoxin³. It is estimated that by 2030 China will have the largest number of HCC patients in the world, an increase of 82% from 2016⁴. Therefore, we need to determine the mechanism of HCC to detect and treat early HCC at the molecular level.

To date, the common treatment methods for HCC include radiofrequency ablation or resection, transplantation, and radioembolization. Patients with tumors of different stages choose different treatment methods, but intrahepatic spread and recurrence rates are still very high after 5 years⁵. In 2013, Nault et al. discovered potential biomarkers, which shifted the research focus to genes related to the pathogenesis of HCC⁶. Gores proposed that HCC needed an individualized treatment scheme and the stratification of patients according to a five-gene score to adopt different molecularly matched targeted treatments⁷. Subsequently, genomic mutations, such as those in the TERT promoter, TP53, CTNNB1, FGF, and PTEN anomalies, which are used in guiding biomarker-matched molecularly targeted therapy of HCC, were detected in HCC, and research on these biomarkers has recently shown significant progress⁸. Therefore, we need to identify additional dysregulated genes to find new treatment strategies to improve prognosis and to realize precision medicine.

In recent years, microarrays have rapidly developed and have become the most successful tool to allow multiple combinatorial chemistry, genomics, and proteomics assays to be carried out in parallel⁹. Currently, a large number of microarray datasets have been disclosed, but the amount of data thoroughly analyzed is lacking. However, bioinformatics analysis can be used to illustrate large and complex datasets. In our study, we examined three HCC datasets in the Gene Expression Omnibus (GEO) and identified differentially expressed genes (DEGs) through a comparison of human HCC tissue with the corresponding noncancerous liver tissue, and applied bioinformatics analysis to identify hub genes and conduct a series of functional analyses.

Materials and Methods

Datasets from the GEO

Three microarray datasets from the GPL570 platform ([HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array) were downloaded from the GEO (http://www.ncbi.nlm.nih.gov/geo/). The GSE121248 expression profile contains 70 HCC samples and 37 noncancerous samples derived from the National Cancer Centre Singapore¹⁰. The GSE45267 expression profile contains 48 HCC samples and 39 noncancerous samples from National Yang-Ming University in Taiwan¹¹. The GSE84402 expression profile contains 14 HCC samples and 14 noncancerous samples from the Shanghai Cancer Institute in China¹².

Inclusion Criteria of the DEGs

GEO2 R is a free download system for online data analysis in the GEO; thus, the DEGs between HCC and noncancerous liver tissues in the GSE121248, GSE45267 and GSE84402 datasets could be obtained¹³. We established the following inclusion criteria for the DEGs: upregulated genes must have a log2 fold change (logFC) ≥ 2 and an adjusted p-value < 0.05, while downregulated genes must have a logFC ≤ –2 and an adjusted p-value < 0.05. Subsequently, Venn diagrams of the up- and downregulated genes were drawn for the different databases.

Functional Annotation of the DEGs

The Database for Annotation, Visualization and Integrated Discovery (DAVID, https://david.ncifcrf.gov/version 6.8) was used to perform a preliminary analysis of the obtained DEGs with systematic and comprehensive biological function notes. The Functional Annotation Tool is the core of DAVID, which includes Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses. Through GO enrichment analysis, we can roughly compare and classify DEGs to better understand their biological characteristics¹⁴. The KEGG helps us to study the functional interpretation of genes and genomes as a whole network¹⁵. In our paper, the threshold p-value < 0.05 was considered statistically significant.

Protein–Protein Interaction Network of the DEGs

To identify the hub regulatory genes and to examine the interactions between the DEGs, a protein–protein interaction network (PPI) was generated with the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING, https://string-db.org/)¹⁶. These genes required an interaction score ≥ 0.4 and a maximum number of interactors = 0, and the genes were imported into Cytoscape (version 3.6.1) with the Molecular Complex Detection (MCODE) app (version 1.5.1)¹⁷ to screen the modules of hub genes with a degree cut-off = 2, haircut on, node score cut-off = 0.2, k-core = 2, and max. depth = 100.

Validation of the Hub Genes

Kaplan–Meier plotter (http://kmplot.com/analysis) is a website that offers an online validation of survival biomarkers and analyzes the overall survival (OS) of patients with high and low expression of certain genes. In our study, hub genes were detected, and a survival curve was drawn. The log-rank p-value (p < 0.05) and hazard ratio with the 95% confidence interval were also calculated. Next, Gene Expression Profiling Interactive Analysis (GEPIA) (http://gepia.cancer-pku.cn/), an open source cancer big data analysis website, was used to analyze the differential expression of 369 HCC and 160 normal tissues from The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) portal. All hub genes were analyzed singly, and |log2FC| =1 and p-value = 0.01 were used as cut-off criteria¹⁸.

Results

Identification of the DEGs between HCC and Noncancerous Tissues

In our study, we chose 132 HCC and 90 noncancerous liver tissues from three datasets, GSE121248, GSE45267, and GSE84402, which were analyzed by GEO2 R and on the basis of filter criteria; 176, 480, and 585 DEGs were obtained, respectively (Additional files 1, 2, and 3). However, we found 109 DEGs with repeated emergence in these datasets, including 24 upregulated genes (logFC ≥ 2) and 85 downregulated genes (logFC ≤ –2), as shown in the Venn diagram (Table 1 and Fig. 1).

Table 1.

Identification of the Up- and Downregulated DEGs in HCC Tissues Compared with Noncancerous Liver Tissues.

DEGs	Gene names
Upregulated	CDK1, CAP2, DTL, PEG10, RACGAP1, CTHRC1, IGF2BP3, RRM2, CCNB1, TOP2A, ASPM, HMMR, CDKN3, PBK, GPC3, SULT1C2, NEK2, ANLN, ACSL4, DUXAP10, CRNDE, BUB1B, ECT2, PRC1
Downregulated	CYP4A22///CYP4A11, CYP26A1, BBOX1, CYP2A6, SERPINE1, PGLYRP2, LINC01093, CXCL14, SLC22A1, IGF1, CYP39A1, HAO2, IGHM, FAM134B, MT1F, SLC25A47, MFSD2A, ZG16, FLJ22763, HHIP, KCNN2, SLCO1B3, CYP1A2, CNDP1, BCO2, ACSM3, FCN3, GBA3, PDGFRA, TTC36, CLEC4G, C3P1, CDH19, CYP2B6, GYS2, FOLH1B, KMO, CD5 L, LPA, GHR, CLEC1B, CXCL2, FOSB, LIFR, FAM65C, CYP2C9, CLRN3, CYP2A7, LCAT, CLEC4 M, ESR1, FOS, LOC101928916///NNMT, PLAC8, ALDOB, HAMP, DNASE1L3, DCN, NAT2, BCHE, AKR1D1, TMEM27, CRHBP, THRSP, IDO2, HGFAC, IGHA2///IGHA1///IGH, C7, ADH4, GPM6A, OIT3, MT1 M, HGF, GLYAT, CYP2B7P///CYP2B6, JCHAIN, GLS2, SRD5A2, ADRA1A, EGR1, APOF, C9, SRPX, FCN2, LINC00844

DEGs

Gene names

Upregulated

CDK1, CAP2, DTL, PEG10, RACGAP1, CTHRC1, IGF2BP3, RRM2, CCNB1, TOP2A, ASPM, HMMR, CDKN3, PBK, GPC3, SULT1C2, NEK2, ANLN, ACSL4, DUXAP10, CRNDE, BUB1B, ECT2, PRC1

Downregulated

CYP4A22///CYP4A11, CYP26A1, BBOX1, CYP2A6, SERPINE1, PGLYRP2, LINC01093, CXCL14, SLC22A1, IGF1, CYP39A1, HAO2, IGHM, FAM134B, MT1F, SLC25A47, MFSD2A, ZG16, FLJ22763, HHIP, KCNN2, SLCO1B3, CYP1A2, CNDP1, BCO2, ACSM3, FCN3, GBA3, PDGFRA, TTC36, CLEC4G, C3P1, CDH19, CYP2B6, GYS2, FOLH1B, KMO, CD5 L, LPA, GHR, CLEC1B, CXCL2, FOSB, LIFR, FAM65C, CYP2C9, CLRN3, CYP2A7, LCAT, CLEC4 M, ESR1, FOS, LOC101928916///NNMT, PLAC8, ALDOB, HAMP, DNASE1L3, DCN, NAT2, BCHE, AKR1D1, TMEM27, CRHBP, THRSP, IDO2, HGFAC, IGHA2///IGHA1///IGH, C7, ADH4, GPM6A, OIT3, MT1 M, HGF, GLYAT, CYP2B7P///CYP2B6, JCHAIN, GLS2, SRD5A2, ADRA1A, EGR1, APOF, C9, SRPX, FCN2, LINC00844

DEGs: differentially expressed genes; HCC: hepatocellular carcinoma.

Figure 1.

Identification of the DEGs between HCC and noncancerous liver tissues in three datasets via a Venn diagram. The blue circle indicates GSE121248, the green circle indicates GSE45267, and the red circle indicates GSE84402. A. Twenty-four upregulated genes had a logFC ≥ 2 and an adjusted p-value < 0.05. B. Eighty-five downregulated genes had a logFC ≤ –2 and an adjusted p-value < 0.05.

GO and KEGG Enrichment Analysis of the DEGs

To extract biological information from the 109 DEGs, we used DAVID for analysis (GO enrichment and KEGG analyses). The results of the GO analysis are presented from three aspects (Table 2): the biological process (BP) terms of the upregulated genes were mitotic nuclear division, regulation of attachment of spindle microtubules to kinetochore and G2/M transition of mitotic cell cycle, while the BP terms of the downregulated genes were epoxygenase P450 pathway, oxidation-reduction process, exogenous drug catabolic process, xenobiotic metabolic process, monocarboxylic acid metabolic process, and drug metabolic process; the cellular component (CC) terms of the upregulated genes were midbody, cytoplasm, mitotic spindle, cytosol, and nucleus, while the CC terms of downregulated genes were organelle membrane, extracellular region, blood microparticle, and extracellular space; the molecular function (MF) terms of the upregulated genes were heme binding, iron ion binding, oxidoreductase activity, acting on paired donors, with incorporation or reduction of molecular oxygen, reduced flavin or flavoprotein as one donor, and incorporation of one atom of oxygen, arachidonic acid epoxygenase activity, oxygen binding and monooxygenase activity, while the MF terms of downregulated genes were histone kinase activity, protein binding, protein kinase activity, and protein serine/threonine kinase activity. The results of the KEGG analysis are shown in Table 3. The upregulated genes were significantly enriched in the p53 signaling pathway and the cell cycle. The downregulated genes were significantly enriched in retinol metabolism, caffeine metabolism, drug metabolism – cytochrome P450, metabolism of xenobiotics by cytochrome P450, and chemical carcinogenesis.

Table 2.

GO Enrichment Analysis of the Up- and Downregulated DEGs.

Expression	Category	Term	Count	%	p-value	FDR
Upregulated	GOTERM_BP_DIRECT	GO:0007067∼mitotic nuclear division	6	25	1.13E-05	0.0147121
Upregulated	GOTERM_BP_DIRECT	GO:0051988∼regulation of attachment of spindle microtubules to kinetochore	3	12.5	3.12E-05	0.040502664
	GOTERM_BP_DIRECT	GO:0000086∼G2/M transition of mitotic cell cycle	4	16.67	6.34E-04	0.821644638
	GOTERM_CC_DIRECT	GO:0030496∼midbody	6	25	3.05E-07	3.20E-04
	GOTERM_CC_DIRECT	GO:0005737∼cytoplasm	17	70.83	8.67E-06	0.009070333
	GOTERM_CC_DIRECT	GO:0072686∼mitotic spindle	3	12.5	0.001009	1.05139872
	GOTERM_CC_DIRECT	GO:0005829∼cytosol	11	45.83	0.001941	2.012816159
	GOTERM_CC_DIRECT	GO:0005634∼nucleus	14	58.33	0.002201	2.280305423
	GOTERM_MF_DIRECT	GO:0035173∼histone kinase activity	2	8.33	0.004967	5.022871579
	GOTERM_MF_DIRECT	GO:0005515∼protein binding	18	75	0.006112	6.147950372
	GOTERM_MF_DIRECT	GO:0004672∼protein kinase activity	4	16.67	0.009546	9.450412521
	GOTERM_MF_DIRECT	GO:0004674∼protein serine/threonine kinase activity	4	16.67	0.010824	10.6521653
	GOTERM_MF_DIRECT	GO:0005524∼ATP binding	6	25	0.033186	29.48034821
Downregulated	GOTERM_BP_DIRECT	GO:0019373∼epoxygenase P450 pathway	5	6.17	1.01E-06	0.001505249
Downregulated	GOTERM_BP_DIRECT	GO:0055114∼oxidation-reduction process	14	17.28	1.65E-06	0.0024543
	GOTERM_BP_DIRECT	GO:0042738∼exogenous drug catabolic process	4	4.94	1.76E-05	0.026056805
	GOTERM_BP_DIRECT	GO:0006805∼xenobiotic metabolic process	6	7.41	2.38E-05	0.035330156
	GOTERM_BP_DIRECT	GO:0032787∼monocarboxylic acid metabolic process	3	3.70	5.73E-05	0.084993128
	GOTERM_BP_DIRECT	GO:0017144∼drug metabolic process	4	4.94	2.23E-04	0.329889509
	GOTERM_CC_DIRECT	GO:0031090∼organelle membrane	8	9.88	6.75E-08	7.45E-05
	GOTERM_CC_DIRECT	GO:0005576∼extracellular region	23	28.40	2.98E-07	3.29E-04
	GOTERM_CC_DIRECT	GO:0072562∼blood microparticle	7	8.64	3.82E-05	0.042171314
	GOTERM_CC_DIRECT	GO:0005615∼extracellular space	15	18.52	0.001068	1.172144989
	GOTERM_MF_DIRECT	GO:0020037∼heme binding	8	9.88	1.72E-06	0.002191179
	GOTERM_MF_DIRECT	GO:0005506∼iron ion binding	8	9.88	3.58E-06	0.004576749
	GOTERM_MF_DIRECT	GO:0016705∼oxidoreductase activity, acting on paired donors, with incorporation or reduction of molecular oxygen	6	7.41	4.03E-06	0.005148689
	GOTERM_MF_DIRECT	GO:0016712∼oxidoreductase activity, acting on paired donors, with incorporation or reduction of molecular oxygen, reduced flavin or flavoprotein as one donor, and incorporation of one atom of oxygen	4	4.94	3.13E-05	0.039961043
	GOTERM_MF_DIRECT	GO:0008392∼arachidonic acid epoxygenase activity	4	4.94	3.13E-05	0.039961043
	GOTERM_MF_DIRECT	GO:0019825∼oxygen binding	5	6.17	4.47E-05	0.05708859
	GOTERM_MF_DIRECT	GO:0004497∼monooxygenase activity	5	6.17	1.03E-04	0.131099338

DEGs: differentially expressed genes; GO: Gene Ontology; BP: biological process; CC: cellular component; MF: molecular function; FDR: false discovery rate.

Table 3.

KEGG Enrichment Analysis of the Up- and Downregulated DEGs.

Expression	Pathway ID	Name	Count	%	p-value	Genes
Upregulated	hsa04115	p53 signaling pathway	3	12.5	0.001367	CCNB1, CDK1, RRM2
Upregulated	hsa04110	Cell cycle	3	12.5	0.004611	CCNB1, CDK1, BUB1B
Downregulated	hsa00830	Retinol metabolism	6	7.41	4.88E-05	CYP2B6, CYP2C9, ADH4, CYP26A1, CYP2A6, CYP1A2
	hsa00232	Caffeine metabolism	3	3.70	3.95E-04	NAT2, CYP2A6, CYP1A2
	hsa00982	Drug metabolism – cytochrome P450	5	6.17	8.80E-04	CYP2B6, CYP2C9, ADH4, CYP2A6, CYP1A2
	hsa00980	Metabolism of xenobiotics by cytochrome P450	5	6.17	0.001210	CYP2B6, CYP2C9, ADH4, CYP2A6, CYP1A2
	hsa05204	Chemical carcinogenesis	5	6.17	0.001617	CYP2C9, ADH4, NAT2, CYP2A6, CYP1A2

DEGs: differentially expressed genes; KEGG: Kyoto Encyclopedia of Genes and Genomes.

PPI Network Analysis of the DEGs and the Identification of Hub Genes

To explore the functions of these genes, we looked for proteins that interact with the DEGs in STRING and constructed a PPI network that included 69 nodes and 209 edges (Fig. 2A). Then, the hub genes of the DEGs were identified with the MCODE app from Cytoscape (Fig. 2B). In this network, we obtained 15 nodes and 105 edges; these nodes represent 15 genes (all upregulated hub genes: denticleless E3 ubiquitin protein ligase homolog (DTL), cyclin-dependent kinase 1 (CDK1), cyclin B1 (CCNB1), Rac GTPase-activating protein 1 (RACGAP1), epithelial cell transforming 2 (ECT2), NIMA-related kinase 2 (NEK2), BUB1 mitotic checkpoint serine/threonine kinase B (BUB1B), PDZ-binding kinase (PBK), DNA topoisomerase II alpha (TOP2A), abnormal spindle microtubule assembly (ASPM), hyaluronan-mediated motility receptor (HMMR), ribonucleotide reductase regulatory subunit M2 (RRM2), cyclin-dependent kinase inhibitor 3 (CDKN3), protein regulator of cytokinesis 1 (PRC1), and anillin actin-binding protein (ANLN) (Table 4).

Figure 2.

Identification of hub genes from the DEGs by STRING and MCODE. A. A PPI network was constructed; red nodes represent upregulated genes, and blue nodes represent downregulated genes. B. The hub genes (yellow nodes) with a degree cut-off = 2, haircut on, node score cut-off = 0.2, k-core = 2, and max. depth = 100 were screened with MCODE.

Table 4.

Identification of the Hub Genes in the Protein–protein Interaction Network.

Expression	Genes
Upregulated	DTL, CDK1, CCNB1, RACGAP1, ECT2, NEK2, BUB1B, PBK, TOP2A, ASPM, HMMR, RRM2, CDKN3, PRC1, ANLN

Kaplan–Meier Plotter and GEPIA of the Hub Genes

After screening the 15 hub genes, we performed an OS analysis of the target genes in Kaplan–Meier plotter. The results showed that mutations in these hub genes could cause poor OS in HCC patients (Fig. 3). Subsequently, using GEPIA, we also found that the expression of the hub genes was significantly increased in tumor tissues (Fig. 4).

Figure 3.

Overall survival analysis of the hub genes to validate survival biomarkers by Kaplan–Meier plotter. p < 0.05 was considered statistically significant.

Figure 4.

Expression of the hub genes was validated in 369 HCC tissues and 160 normal tissues with GEPIA. |log2FC| > 1 and p-value < 0.01 were considered statistically significant. Tumor tissue is shown in red, and normal tissue is shown in gray.

KEGG Analysis of the Hub Genes

After the verification of 15 hub genes using Kaplan–Meier plotter and GEPIA, KEGG pathways of these hub genes were re-analyzed via DAVID to better understand their functions; we identified four pathways associated with the 15 hub genes (Table 5). CCNB1, CDK1, and RRM2 were enriched in the p53 signaling pathway, CCNB1, CDK1, and BUB1B were enriched in the cell cycle (Fig. 5), and CCNB1 and CDK1 were enriched in Progesterone-mediated oocyte maturation and in Oocyte meiosis (p > 0.05), respectively. Studied have proved that p53 signaling pathway and cell cycle play important roles in the progression of HCC^19

–24. Therefore, CCNB1, CDK1, RRM2 and BUB1B may play important roles in the occurrence and development of HCC.

Table 5.

KEGG Enrichment Analysis of the Hub Genes.

Pathway ID	Name	Count	%	p-value	Genes
hsa04115	p53 signaling pathway	3	20	5.54E-04	CCNB1, CDK1, RRM2
hsa04110	Cell cycle	3	20	0.001888704	CCNB1, CDK1, BUB1B
hsa04914	Progesterone-mediated oocyte maturation	2	13.33	0.049647728	CCNB1, CDK1
hsa04114	Oocyte meiosis	2	13.33	0.063012175	CCNB1, CDK1

KEGG: Kyoto Encyclopedia of Genes and Genomes

Figure 5.

KEGG enrichment analysis of the hub genes. A. CCNB1, CDK1, and RRM2 are enriched in the p53 signaling pathway (Cyclin B represents CCNB1, Cdc2 represents CDK1, and p53R2 represents RRM2). B. CCNB1, CDK1, and BUB1B enriched in the cell cycle (BubR1 represents BUB1B, and CycB represents CCNB1).

Discussion

HCC-related genes have yet to be identified to elucidate the underlying molecular mechanisms of cancer susceptibility, progression and prognosis. That is, a novel therapy targeting a direct regulatory mechanism that is correlated with a poor prognosis in HCC patients still needs to be identified. In this study, we selected three public microarray datasets from Asian countries, GSE121248, GSE45267, and GSE84402, and 109 DEGs were screened by GEO2 R according to the inclusion criteria between 132 HCC samples and 90 noncancerous samples. A PPI network of the DEGs was used to identify hub genes, and then the roles of these hub genes in HCC were validated via an OS analysis of patients in Kaplan–Meier plotter. The expression of these hub genes was compared between HCC and noncancerous tissues from the TCGA and GTEx. Finally, 15 hub genes related to a poor prognosis in HCC were identified: CCNB1, CDK1, RRM2, BUB1B, DTL, RACGAP1, ECT2, NEK2, PBK, TOP2A, ASPM, HMMR, CDKN3, PRC1, and ANLN.

In our study, among the 15 hub genes that were significantly enriched in the p53 signaling and cell cycle pathways were four potential therapeutic targets: CCNB1, CDK1, RRM2, and BUB1B. Studied have proved that cell cycle and p53 signaling pathways play important roles in development of HCC19-24. The mammalian cell cycle is controlled by regulators of the G1 to S transition such as p53, retinoblastoma, and cyclin D1 proteins. Many reports have shown that disruption of these cell cycle-related genes results in the progression of HCC19-21. And the p53 pathway is composed of a set of genes and their proteins that respond to a wide variety of stress signals. These responses to stress include cell cycle arrest, cellular senescence, or apoptosis. Moreover, the p53-regulated genes produce proteins that communicate these stress signals to adjacent cells, prevent and repair damaged DNA, and create feedback loops that regulate p53 activity and communicate with other signal transduction pathways, such as the Wnt/β-catenin, RB/INK4a, and p38 MAP pathways. Thus, the disruption of the p53 pathway has been reported in almost every type of cancer including HCC^22–24. Based on these studies, CCNB1, CDK1, RRM2, and BUB1B may play important roles in the occurrence and development of HCC. Previous studies have reported that the CCNB1–Cdk1 complex is a key regulator of mitotic entry²⁵. Importantly, Chai et al. noted that CCNB1 is highly expressed in HCC and is closely related to the poor prognosis of HCC patients, consistent with our results²⁶. Gu et al. showed that CCNB1 is directly suppressed by miR-144 as a therapy targeting HCC²⁷. Thus, the expression of CCNB1 is often used to estimate prognosis after treatment with anticancer drugs. CDK1 is a serine/threonine kinase and plays an important role in cell cycle progression. The inhibition of CDK1 suppresses cellular proliferation^28,29. Its expression is also significantly higher in HCC tissue and cells; however, the real mechanism underlying the correlation between CDK1 and HCC remains unclear. RRM2 catalyzes the production of deoxynucleotide ribonucleotides in DNA synthesis and is a potential prognostic biomarker in glioma³⁰, breast cancer³¹, and prostate cancer³². BUB1B, as a key mitotic spindle checkpoint, plays an important role in the development of many tumors. For example, the expression of BUB1B is increased in adrenocortical carcinomas³³, and BUB1B promotes tumor proliferation and induces radioresistance in glioblastoma³⁴.

We also identified other hub genes, namely, ANLN, ASPM, CDKN3, DTL, ECT2, HMMR, NEK2, PBK, PRC1, RACGAP1, and TOP2A, all of which encode proteins with high degrees in the PPI network. Most of these genes are closely related to HCC. For instance, ANLN is an actin-binding protein that is essential for assembly of the cleavage furrow during cytokinesis. Lian et al.³⁵ and Zhang et al.³⁶ have shown that ANLN promotes tumor growth by decreasing apoptosis and DNA damage, and the inhibition of ANLN in liver cells blocks cytokinesis and inhibits the development of liver tumors. ASPM overexpression is a molecular marker that predicts the enhanced invasive/metastatic potential of HCC³⁷. CDKN3 is involved in the cell cycle, and Xing et al. found that CDKN3 is frequently overexpressed in both HCC cell lines and samples and that the overexpression of CDKN3 is correlated with poor tumor differentiation and advanced tumor stage by promoting cell proliferation³⁸. Chen et al. showed that PRC1 promotes early HCC recurrence and poor patient outcomes in association with the Wnt/β-catenin signaling pathway³⁹. Yang et al. demonstrated that PBK promotes the metastasis of hepatocellular carcinoma by activating the ETV4-uPAR signaling pathway⁴⁰. In addition, TOP2A contributes to the early detection and targeted therapy of a variety of cancers^41
–43, including HCC⁴⁴. However, the relationship between some of these genes, such as NEK2, and HCC remains unclear. NEK2 participates in the progression of multiple types of cancer, such as glioblastoma⁴⁵, adrenal cortical carcinoma⁴⁶ and myeloma⁴⁷. Interestingly, Li et al. reported that NEK2 was overexpressed in HCC tissue and cells⁴⁸. HCC patients with high NEK2 expression had an unfavorable prognosis, in accord with our finding. However, Fu et al. proposed that low NEK2 expression was related to a poor prognosis in HCC⁴⁹. Therefore, the effect of NEK2 on HCC is controversial, and its elucidation requires further evidence.

In brief, our study found that high expression of 15 hub genes was closely related to poor survival in HCC patients, indicating that their antagonism may improve the prognosis of HCC. However, the detailed mechanisms underlying the possible effects of these genes still need to be characterized in vivo and in vitro. Moreover, whether there are differences in the expression of these genes in different stages of HCC or different diseases also needs to be further studied and discussed.

Supplemental Material

Supplemental Material, Complete_list_of_differentially_expressed_genes - Screening Hub Genes as Prognostic Biomarkers of Hepatocellular Carcinoma by Bioinformatics Analysis

Supplemental Material, Complete_list_of_differentially_expressed_genes for Screening Hub Genes as Prognostic Biomarkers of Hepatocellular Carcinoma by Bioinformatics Analysis by Zengyuan Zhou, Yuzheng Li, Haiyue Hao, Yuanyuan Wang, Zihao Zhou, Zhipeng Wang and Xia Chu in Cell Transplantation

Supplemental Material

Supplemental Material, Complete_list_of_differentially_expressed_genes_2 - Screening Hub Genes as Prognostic Biomarkers of Hepatocellular Carcinoma by Bioinformatics Analysis

Supplemental Material, Complete_list_of_differentially_expressed_genes_2 for Screening Hub Genes as Prognostic Biomarkers of Hepatocellular Carcinoma by Bioinformatics Analysis by Zengyuan Zhou, Yuzheng Li, Haiyue Hao, Yuanyuan Wang, Zihao Zhou, Zhipeng Wang and Xia Chu in Cell Transplantation

Supplemental Material

Supplemental Material, Complete_list_of_differentially_expressed_genes_3 - Screening Hub Genes as Prognostic Biomarkers of Hepatocellular Carcinoma by Bioinformatics Analysis

Supplemental Material, Complete_list_of_differentially_expressed_genes_3 for Screening Hub Genes as Prognostic Biomarkers of Hepatocellular Carcinoma by Bioinformatics Analysis by Zengyuan Zhou, Yuzheng Li, Haiyue Hao, Yuanyuan Wang, Zihao Zhou, Zhipeng Wang and Xia Chu in Cell Transplantation

Footnotes

Ethical Approval

Ethical Approval is not applicable for this article.

Statement of Human and Animal Rights

This article does not contain any studies with human or animal subjects.

Statement of Informed Consent

There are no human subjects in this article and informed consent is not applicable.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by grants from the National Natural Science Foundation of China (No. 81673153), University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province (UNPYSCT-2016184), Postdoctoral Science Foundation of Heilongjiang province (LBH-Z15142) and Outstanding Youth Fundation of the Fourth Affiliated Hospital of Harbin Medical University (HYDSYJQ201508).

ORCID iD

Zengyuan Zhou

Supplemental Material

Supplemental material for this article is available online.

References

Islami

Goding Sauer

Miller

Siegel

Fedewa

Jacobs

McCullough

Patel

Soerjomataram

Flanders

, et al. Proportion and number of cancer cases and deaths attributable to potentially modifiable risk factors in the United States. CA Cancer J Clin. 2018;68(1):31–54.

Degasperi

Colombo

. Distinctive features of hepatocellular carcinoma in non-alcoholic fatty liver disease. Lancet Gastroenterol Hepatol. 2016;1(2):156–164.

Forner

Llovet

Bruix

. Hepatocellular carcinoma. Lancet. 2012;379(9822):1245–1255.

Estes

Anstee

Arias-Loste

Bantel

Bellentani

Caballeria

Colombo

Craxi

Crespo

Day

Eguchi

, et al. Modeling NAFLD disease burden in China, France, Germany, Italy, Japan, Spain, United Kingdom, and United States for the period 2016-2030. J Hepatol. 2018;69(4):896–904.

Kulik

El-Serag

. Epidemiology and management of hepatocellular carcinoma. Gastroenterology. 2019;156(2):477–491.e1.

Nault

De Reyniès

Villanueva

Calderaro

Rebouissou

Couchy

Decaens

Franco

Imbeaud

Rousseau

Azoulay

, et al. A hepatocellular carcinoma 5-gene score associated with survival of patients after liver resection. Gastroenterology. 2013;145(1):176–187.

Gores

GJ.

Decade in review-hepatocellular carcinoma: HCC-subtypes, stratification and sorafenib. Nat Rev Gastroenterol Hepatol. 2014;11(11):645–647.

Khemlina

Ikeda

Kurzrock

. The biology of hepatocellular carcinoma: implications for genomic and immune therapies. Mol Cancer. 2017;16(1):149.

Díaz-Mochón

Tourniaire

Bradley

. Microarray platforms for enzymatic and cell-based assays. Chem Soc Rev. 2007;36(3):449–457.

10.

Wang

Ooi

Hui

. Identification and validation of a novel gene signature associated with the recurrence of human hepatocellular carcinoma. Clin Cancer Res. 2007;13(21):6275–6283.

11.

Chen

Tsai

Huang

Liang

Sun

Chau

Yeh

Chang

. Lymphoid enhancer factor 1 contributes to hepatocellular carcinoma progression through transcriptional regulation of epithelial-mesenchymal transition regulators and stemness genes. Hepatol Commun. 2018;2(11):1392–1407.

12.

Wang

Huo

Yang

Cheng

Wang

Deng

Jin

Wang

Zhao

, et al. STAT3-mediated upregulation of lncRNA HOXD-AS1 as a ceRNA facilitates liver cancer metastasis by regulating SOX4. Mol Cancer. 2017;16(1):136.

13.

Barrett

Wilhite

Ledoux

Evangelista

Kim

Tomashevsky

Marshall

Phillippy

Sherman

Holko

Yefanov

, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res. 2013;41(Database issue):D991–D995.

14.

Mazandu

Chimusa

Mulder

. Gene ontology semantic similarity tools: survey on features and challenges for biological knowledge discovery. Brief Bioinform. 2017;18(5):886–901.

15.

Kanehisa

Furumichi

Tanabe

Sato

Morishima

. KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res. 2017;45(D1):D353–D361.

16.

Franceschini

Szklarczyk

Frankild

Kuhn

Simonovic

Roth

Lin

Minguez

Bork

von Mering

Jensen

. STRING v9.1: protein-protein interaction networks, with increased coverage and integration. Nucleic Acids Res. 2013;41(Database issue):D808–D815.

17.

Bader

Hogue

CWV

. An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinformatics. 2003;4:2.

18.

Tang

Kang

Gao

Zhang

. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017;45(W1):W98–W102.

19.

Farazi

DePinho

. Hepatocellular carcinoma pathogenesis: from genes to environment. Nat Rev Cancer. 2006;6(9):674–687.

20.

Teufel

Staib

Kanzler

Weinmann

Schulze-Bergkamen

Galle

. Genetics of hepatocellular carcinoma. World J Gastroenerol. 2007;13(16):2271–2282.

21.

Joo

Kang

Kim

Lee

Jang

. Cyclin D1 overexpression in hepatocellular carcinoma. Liver. 2001;21(2):89–95.

22.

Levine

Feng

. The P53 pathway: what questions remain to be explored? Cell Death Differ. 2006;13(6):1027–1036.

23.

Nishida

Goel

. Genetic and epigenetic signatures in human hepatocellular carcinoma: a systematic review. Curr Genomics. 2011;12(2):130–137.

24.

Meng

Franklin

Dong

Zhang

. MDM2-p53 pathway in hepatocellular carcinoma. Cancer Res. 2014;74(24):7161–7167.

25.

Nakayama

Yamaguchi

. Role of cyclin B1 levels in DNA damage and DNA damage-induced senescence. Int Rev Cell Mol Biol. 2013;305:303–337.

26.

Chai

Xie

Yin

Guo

Wang

Liu

Zhang

Yin

Nie

, et al. FOXM1 promotes proliferation in human hepatocellular carcinoma cells by transcriptional activation of CCNB1. Biochem Biophys Res Commun. 2018;500(4):924–929.

27.

Liu

. MicroRNA-144 inhibits cell proliferation, migration and invasion in human hepatocellular carcinoma by targeting CCNB1. Cancer Cell Int. 2019;19:15.

28.

Xiao

Liu

Xing

. E2F8 is a potential therapeutic target for hepatocellular carcinoma. J Cancer. 2017;8(7):1205–1213.

29.

Zhang

Huang

Ran

Xiong

Zhong

Fan

Wang

miR-582-5p inhibits proliferation of hepatocellular carcinoma by targeting CDK1 and AKT3. Tumour Biol. 2015;36(11):8309–8316.

30.

Sun

Yang

Zhang

Song

Zhang

Xing

Yang

Wei

, et al. RRM2 is a potential prognostic biomarker with functional significance in glioma. Int. J Biol Sci. 2019;15(3):533–543.

31.

Chen

Yang

Cheng

Qian

Sun

Zhu

. Bioinformatics analysis revealing prognostic significance of RRM2 gene in breast cancer. Biosci Rep. 2019;39(4):BSR20182062.

32.

Mazzu

Armenia

Chakraborty

Yoshikawa

Coggins

Nandakumar

Gerke

Pomerantz

Qiu

Zhao

Atiq

, et al. A novel mechanism driving poor-prognosis prostate cancer: overexpression of the DNA repair gene, ribonucleotide reductase small subunit M2 (RRM2). Clin Cancer Res. 2019;25(14):4480–4492.

33.

Passaia

BDS

Dias

Kremer

Antonini

SRR

de Almeida

Fragoso

MCBV

Lotfi

CFP

. TCF21/POD-1, a transcritional regulator of SF-1/NR5A1, as a potential prognosis marker in adult and pediatric adrenocortical tumors. Front Endocrinol (Lausanne). 2018;9:38.

34.

Liu

Shang

. The FOXM1/BUB1B signaling pathway is essential for the tumorigenicity and radioresistance of glioblastoma. Oncol Rep. 2017;38(6):3367–3375.

35.

Lian

Huang

Wang

Deng

Xia

Zeng

Chen

Wang

Huang

. Anillin is required for tumor growth and regulated by miR-15a/miR-16 -1 in HBV-related hepatocellular carcinoma. Aging (Albany NY). 2018;10(8):1884–1901.

36.

Zhang

Nguyen

Zhou

Sehgal

Nassour

Gopal

Goodman

Singal

Yopp

, et al. Knockdown of anillin actin binding protein blocks cytokinesis in hepatocytes and reduces liver tumor development in mice without affecting regeneration. Gastroenterology. 2018;154(5):1421–1434.

37.

Lin

Pan

Liu

Jeng

Peng

Lai

Hsu

. ASPM is a novel marker for vascular invasion, early recurrence, and poor prognosis of hepatocellular carcinoma. Clin Cancer Res. 2008;14(15):4814–4820.

38.

Xing

Xie

Zhou

Zhang

Ding

Wei

Zheng

. Cyclin-dependent kinase inhibitor 3 is overexpressed in hepatocellular carcinoma and promotes tumor cell proliferation. Biochem Biophys Res Commun. 2012;420(1):29–35.

39.

Chen

Rajasekaran

Xia

Zhang

Kong

Sekar

Seshachalam

Deivasigamani

Goh

Ooi

Hong

, et al. The microtubule-associated protein PRC1 promotes early recurrence of hepatocellular carcinoma in association with the Wnt/β-catenin signalling pathway. Gut. 2016;65(9):1522–1534.

40.

Yang

Zhong

Jia

Tang

Cheng

Ren

Zhou

Ren

, et al. PBK overexpression promotes metastasis of hepatocellular carcinoma via activating ETV4-uPAR signaling pathway. Cancer Lett. 2019;452:90–102.

41.

Labbé

Sweeney

Brown

Galbo

Rosario

Wadosky

Sjöström

Alshalalfa

Erho

Davicioni

, et al. TOP2A and EZH2 provide early detection of an aggressive prostate cancer subgroup. Clin Cancer Res. 2017;23(22):7072–7083.

42.

Chen

Chien

Chen

Hwang

Chen

Yeh

Hsieh

Chang

Hsu

Hua

. Amplification of HER2 and TOP2A and deletion of TOP2A genes in a series of Taiwanese breast cancer. Medicine (Baltimore). 2017;96(2): e5582.

43.

Aumayr

Klatte

Neudert

Birner

Shariat

Schmidinger

Susani

Haitel

. HER2 and TOP2A gene amplification and protein expression in upper tract urothelial carcinomas. Pathol Oncol Res. 2018;24(3):575–581.

44.

Panvichian

Tantiwetrueangdet

Angkathunyakul

Leelaudomlipi

. TOP2A amplification and overexpression in hepatocellular carcinoma tissues. Biomed Res Int. 2015;2015:381602.

45.

Wang

Cheng

Pavlyukov

Zhang

Kim

Minata

Mohyeldin

Xie

Chen

Goidts

, et al. Targeting NEK2 attenuates glioblastoma growth and radioresistance by destabilizing histone methyltransferase EZH2. J Clin Invest. 2017;127(8):3075–3089.

46.

Mete

Gucer

Kefeli

Asa

. Diagnostic and prognostic biomarkers of adrenal cortical carcinoma. Am J Surg Pathol. 2018;42(2):201–213.

47.

Franqui-Machin

Hao

Bai

Zhan

Habelhah

Jethava

Qiu

Frech

Tricot

Zhan

. Destabilizing NEK2 overcomes resistance to proteasome inhibition in multiple myeloma. J Clin Invest. 2018;128(7):2877–2893.

48.

Zhong

Shen

Zhou

Deng

Chen

Zhou

. NEK2 serves as a prognostic biomarker for hepatocellular carcinoma. Int J Oncol. 2017;50(2):405–413.

49.

Liu

Wang

Mou

Tong

. Low expression of NEK2 is associated with hepatocellular carcinoma progression and poor prognosis. Cancer Biomark. 2017;20(1):101–106.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.84 MB

0.79 MB

0.45 MB