Introduction

Non-communicable diseases include brain haemorrhage, severe pulmonary illnesses, cardiovascular disease, insulin resistance, and carcinoma. These are the leading causes of death globally1. The cause of death might be due to both internal and external reasons. In the recent report2, among all the cancers, hepatic carcinoma is among the most probable causes of cancer. Regarding the total deaths reported globally, HCC (hepatocellular carcinoma) is the third major cause of death. A total of 906,000 cancer cases were reported worldwide. In these cases, mortality occurred in 830,000 cases2. Among the liver cancer patients, 75–85% of patients are affected by HCC, while 10–15% of patients are suffering from intra-hepatic cholangiocarcinoma. Viral hepatitis B or C infection, aflatoxin-based food contaminants, excessive alcohol consumption, weight gain, type 2 diabetes, and tobacco consumption are the key leading causes of HCC. The leading causes of HCC in the world are region-dependent2. Liver cirrhosis is a major hepatic disease. It is also identified as a major health risk for HCC3. Clinically, in liver cirrhosis, the progressive deterioration of the hepatic functions takes place, which leads to HCC development4.

Despite the availability of modern clinical facilities such as surgical removal, organ transplants, catheter-based arterial chemoembolization, and radiofrequency ablation, patients with HCC have a low chance of survival5. Modern therapeutic techniques for liver cancer have improved incidence and morbidity. However, HCC survival rates are quite poor in the early stages6. In general, clinically significant signs and symptoms of HCC are not recognized in the early stages of cancer. Thus, in these stages, treatment was missed for cancer complications7. Patients with HCC can benefit from timely and successful therapy in terms of both wellbeing and life longevity8,9. Consequently, the emergence of new prognostic potential signatures and therapeutic targets are crucial concerns in the treatment of HCC. Previous studies have shown that valuable prognostic biomarkers, like functional genes, were recognized through bioinformatics analysis, which is based on the high-throughput data of HCC10,11,12. Computational approaches could be used to analyze the gene expression datasets, which revealed the important mechanisms of action of the genes involved in HCC treatment13.

In the present work, we have identified the prevalent HCC genes from the Gene Card and DisGeNET datasets. Further, protein–protein interaction (PPI) with these prevalent genes was performed. In this study, hub genes were selected based on ontological characteristics such as degree of closeness, betweenness, etc. Then, Gene Ontology (GO) evaluation and the Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway linked to HCC were investigated. The TIMER2.0 database is utilized for the evaluation of potential links between each hub gene and the infiltrated immune cells. Databases, such as GEPIA214, HPA15, and TIMER2.016, were used to visualize the prognostic scenario of candidate hub genes during the present investigation. This technique was employed to ensure a stable platform for exploring new biomarkers for disease diagnosis and to find potential treatment targets for HCC. Hub genes were screened in this work by merging the results from the PPI and GO datasets. Finally, expression and survival analysis and immune infiltration analysis were used to validate the hub genes of HCC. Thus, in the present study, hub genes of the HCC were identified through an in-silico approach.

Results

Identification of target gene

From the datasets, a total of 1569 HCC target genes from Gene Card and 361 genes from DisGenet were taken. In the present study, 98 prevalent genes were retrieved and further recognized from datasets for network design (Fig. 1).

Figure 1
figure 1

Venn diagram showing the target genes for HCC based on different dataset.

Full size image

PPI network building and hub gene identification

Using the STRING database, a PPI network was created based on 98 common genes. Results revealed that 91 nodes as well as 1068 edges with a 0.695 clustering coefficient along with an enrichment p-value < 1.0E−16 were recorded (Fig. 2A). Based on the degrees of connectivity, the top ten genes were screened through CytoNCA. These genes were represented as hub genes. Based on the degree, five interlinked genes (TP53, ESR1, AKT1, CASP3, and JUN) were selected for further investigation (Fig. 2B).

Figure 2
figure 2

PPI network showing the distribution of nodes and edges (A, B) which attributes the presence of hub gene Network (https://string-db.org/).

Full size image

Functional annotation of hub genes

On the basis of the GO study and gene enrichment analysis with the KEGG pathway, it is clear that these hub genes regulate the different bioactivities as shown in Table S1. These bioactivities are concerned with different molecular functions (MF), biological processes (BP), as well as cellular components (CC) (Table S1). KEGG analysis revealed the involvement of the hub genes in different signaling pathways such as estrogen, FoxO, IL-17, cancer pathways, mTOR, JAK-STAT, VEGF, PI3K-Akt, Ras, and Toll-like receptor (Table S2). Information based on the bioactivities and signaling pathways, the possible role of these hub genes in cancer regulation processes is shown in Fig. 3.

Figure 3
figure 3

The possible role of hub gene in the regulation of HCC pathways (https://www.genome.jp/kegg).

Full size image

Assessment of hub genes survival

An overall survival study of hub genes was carried out to describe the identification of prognostic variables in selected five hub genes for HCC. The Kaplan–Meier plotter and GEPIA2 analysis revealed significantly (p > 0.05) higher expressions of hub genes TP53 and JUN in patients with HCC. Similarly, the ESR1 hub gene displayed significantly (p < 0.05) lower expression, as confirmed during GEPIA2 and Kaplan–Meier Plotter analysis, in patients suffering with HCC (Fig. 4). Thus, we observed that these hub genes were poorly linked with prognosis in HCC patients. However, hub genes AKT1 and CASP3 were not utilized for further analysis due to variables showing violation (p > 0.05) in the GEPIA2 database as well as results obtained through Kaplan–Meier Plotter analysis.

Figure 4
figure 4

The overall survival of patients (A–E) and (F–J) with HCC by Kaplan–Meier plots (https://kmplot.com/analysis).

Full size image

Verification of hub gene mRNA and protein expression

The GEPIA2 database was used to validate the mRNA expression levels of selected five hub genes in normal and HCC samples (Fig. 5A–E). ESR1 expression was significantly higher in HCC thanin normal hepatic tissues (Fig. 5C). Similar results were reported by several researchers17.

Figure 5
figure 5

The mRNA expression and immunohistochemistry of the (A) TP53, (B) AKT1, (C) ESR1, (D) JUN and (E) CASP3 in normal liver tissues and HCC tissues from the GEPIA2 (http://gepia2.cancer-pku.cn/#index) and HPA database (https://www.proteinatlas.org/). The gray bars in box plots represent normal samples; the red bars in box plots represent tumor samples(*p < 0.01).

Full size image

HPA immune histochemistry data revealed that TP53 and CASP3 protein expressions were positively up-regulated in HCC tissues compared to normal (Fig. 5A,D), whereas hubgenes AKT1, ESR1, and JUN protein expressions were negatively down-regulated in HCC tissues compared to normal (Fig. 5B,C,E). Thus, the mRNA expressions of genes as well as protein expression of the selected five hub genes were regulated in patients suffering with HCCin comparison to normal liver tissue.

The UALCAN database was used for the expression correlation analysis of selected five hub genes linked with HCC. The results of the correlation revealed the significant expression of the selected genes at different stages of HCC (Fig. 6A–E). Based on the findings, hub genes TP53, ESR1, AKT1, CASP3, and JUN were compactly regulated with the risk of HCC development and its progression chronically.

Figure 6
figure 6

The mRNA expression of the TP53 (A), AKT1 (B), ESR1(C), JUN(D) and CASP3 (E) in individual cancer(*p < 0.05, ***p < 0.001) (http://ualcan.path.uab.edu).

Full size image

Hub gene expression in various immune cells

The TIMER database results revealed that the TP53 and ESR1 genes have a significant correlation with the tumor purity parameter. The expression of hub genes (AKT1, ESR1, and JUN) was down regulated and correlated against tumor-infiltrating levels negatively into LIHC; in contrast, up regulated hub genes (TP53 and CASP3) expression was correlated against all TIICs subsets positively into HCC (Fig. 7 A,E). In LIHC, AKT1 gene expression (Fig. 7B)was significantly correlated with five TIICs except CD8+ cells. The infiltration of B cells and macrophages in ESR1 gene expression (Fig. 7C) was significantly observed. JUN gene expression was significant against dendritic cells (DCs), CD4+ T cells, neutrophils, and macrophage infiltration levels (Fig. 7E). In comparison to DCs, neutrophils, and CD8+ T cells (Fig. 8), the hub genes TP53 and CASP3 (Fig. 8A,E) were significantly more expressed in macrophages, CD4+ T cells, and B cells. Results suggested that macrophages have a preferential corner in the immune infiltration study in HCC with reference to selected five hub genes.

Figure 7
figure 7

Analysis of hub gene expression in various immune cells in HCC.TP53 (A), AKT1(B), ESR1(C), CASP3 (D) and JUN (E) (http://timer.cistrome.org).

Full size image
Figure 8
figure 8

Hub gene expression analysis in different immune cells (http://timer.cistrome.org).

Full size image

Discussion

The International Agency for Research on Cancer (IARC) documented that liver cancer diseases are the 6th most commonly diagnosed cancer, while they were the 3rd major cause of global mortality during 2019–202018. Non-alcoholic steatohepatitis is also one of the most prevalent and predominant liver cancers19,20. The identification of hub genes of HCC significantly expanded the research in the areas of bioactivities, survival analysis, expression, and poor prognosis values, as well as immune infiltration using bioinformatics approaches.

The expression of 91 overlapping genes varied in HCC tissues. Furthermore, the hub genes were chosen for their high degree of connectivity. Hub gene expression verification, overall survival, as well as immune infiltration were analyzed. According to the DAVID database, compromised genes were significantly associated with various signaling pathways that were linked with HCC. The GEPIA database was used to validate that TP53 and CASP3 mRNA expression were significantly higher. TP53 is an essential mediator of apoptosis in numerous cell types. It is evidenced by an increase in TP53 expressions. TP53 has the ability to regulate the expression of target genes that are involved in multiple activities, including cell cycle arrest, DNA repair, apoptosis, and preventing deformation processes. TP53 is also important for mitochondrial trans-membrane stability, and it increases apoptosis without transcriptional regulation. The BAX gene promoter area can be bound with wild-type TP53, which regulates the expression of the BAX gene21,22. It was observed that AKT1, ESR1, and JUN mRNA thresholds were significantly lower in expression in HCC patients.

BAX is a BCL2 family member that causes apoptosis23. BAX is a TP53 target. The BAX constructs complexes of BCL2 and inhibits its function while promoting the function of cytochrome c in the mitochondria initially and later into the cytoplasmic matrix. Thus, it aids caspase activation and apoptosis implementation. CASP3 is a key player in apoptotic and necrotic cell death through intrinsic and extrinsic pathways. ROS production and apoptotic cell detachment are also inhibited by CASP7 and CASP9.These caspases support the apoptotic pathways of CASP3. The TP53-BAX-caspase3 signaling pathway induced apoptosis of HCC cells by up-regulating TP53 and activating BAX and CASP3 protein expression24.

The outcome of the hub gene results revealed that abnormal miRNAs play a pivotal function in the development and progression of HCC. Application of miRNAs has recently been introduced as a diagnostic, curative, and prognostic biomarker25. AKT1 is believed to play an important role in HCC oncogenesis, and it has been identified as a key target of most miRNAs in the disease. miRNAs suppressed the expression of key hub genes related to cancer. It suggests that miRNAs may be useful in the detection and therapeutic strategies for the control of HCC. AKT1 is a critical receptor for the activation of the highly cancer-causing Wnt/β-catenin signaling pathway, which is linked with cancer development. This signalling pathway is also connected to HCC uncontrolled cell proliferation26.

Activating protein 1 is a protein family, incorporating JUN, which is a dimer and also functions as a transcription factor. Multiple downstream genes are regulated by such proteins in genetic transcription processes after attachments with the DNA strand. It is responsible for the regulation of cell proliferation, differentiation, invasion, and metastasis etc. Thus, it was clear that activating protein1 plays an important role in cancer development27. JUN is found in both homodimerized and heterodimerized forms in the AP-1 family27. The heterodimers of JUN have more stable and relatively stronger DNA-binding behaviour than homodimers of JUN28. According to accumulating evidence, v-JUN has been one of the 17 modifying oncogenes of the bird malignancy virus and is cellularly homologous to JUN29. JUN appears to be important in the tumour development and growth of HCC28. Its higher expression has been reported in human hepatocellular carcinoma cells27. Furthermore, data from different experimental models with reduced expression of JUN revealed that the involvement of JUN in HCC signaling pathways, which influenced cancer cell proliferation and migration30.

The observation of HCC tissues revealed a significantly lower expression of ESR1 mRNA. It caused the down-regulated expression of ESR1. Inhibition of ESR1 expression could significantly neutralize the inhibition activity of miRNA, which leads to suppression of HCC cell development and progression. The function of miRNAs is important in promoting cell proliferation, migration, and invasion. Thus, targeting ESR1-based treatment of patients suffering from HCC can successfully bring wellness to these patients31.

ABCC532, hub gene, and immune subtype prognostic biomarkers33 have all been linked to HCC. ABCC5 is involved inseveral mechanisms liketrans-membrane transport as well as immune cell filtration of HCC.

In-silico analysis revealed the association of five hub genes (TP53, ESR1, AKT1, CASP3, and JUN) with HCC. These hub genes may be used as liver cancer prognostic biomarkers. The role of these hub genes as biomarkers is based on the analysis of PPI, KEGG pathways, miRNAs, protein expression, and immune cell filtration. This information is based on bioinformatics approaches, which may require additional in vitro and in vivo experimental validation before concluding their exact molecular mechanisms.

Conclusion

HCC is a global health problem that affects millions of people each year. The primary goal of this work was to highlight the relevant hub genes identification, which are involved in HCC. It will be helpful in the risk assessment of HCC and also in the prevention approach. Results revealed that the mRNA and protein thresholds of the five hub genes are associated with several important signaling pathways, which are significantly diagnosed in HCC patients. Based on the data related to the identification, level of expression, and survival patterns of the five hub genes in HCC, our study provides significant information with relevance for the management and treatment of HCC patients.

Methods

Data mining

The relevant HCC targets were obtained from the human gene directory GeneCards (https://www.genecards.org/)34 and DisGeNET (Version 7.0) (https://www.disgenet.org/)35. The parameter for selecting disease target genes in this analysis of the data was "score" ≥ mean score.

Data processing and target screening

The Venn chart was created with the FunRich software36 to screen for intersecting genes between GeneCards and DisGeNET database genes of HCC. The intersection of anticipated targets of both databases was used to obtain the targets of the HCC.

Generation of protein protein interaction (PPI) networks and recognition of hub genes

The PPI network was first plotted using the online Search Tool for the Acquisition of Interacting Genes (STRING)37 (https://string-db.org/). It was further displayed and interpreted through Cytoscape software38. The CytoNCA was also used to find the hub genes with the highest degree of association, which have been filtered with regard to degree, closeness, or betweenness.

Investigations of candidate genes enrichment and pathway

This analysis was performed with the help of GO and KEGG pathways. GO is a functional procedure that helps in the classification of hub genes on the basis of biological processes, molecule functioning, and cellular comportment, while the KEGG39 (https://www.genome.jp/kegg) and DAVID databases (https://david.ncifcrf.gov/) help in the enrichment analysis for mechanism of action and to analyze the possible function of the preferred hub genes40,41.

Verification and survivability assessment of hub genes

The GEPIA2 database was selected to evaluate the impact of hub gene biomarkers on HCC patient survival rates. It depicts the analysis of normal and tumour sample data, which were collected from TCGA and GTEx42. The genes found significantly (p < 0.05) in the formats were chosen for further investigation. The biomarkers were translated into expression profiles for selected hub genes in normal as well as HCC samples and further visualized and evaluated using GEPIA2.

In addition, for protein expression in the human protein atlas, a web interface (http://www.proteinatlas.org) has been utilized to assess the role of genes in normal as well as HCC samples. Subsequently, another interface, UALCAN(http://ualcan.path.uab.edu/), was used for the examination of relationships between different stages of cancer for mRNA expression of the genes of interest using the TCGA set of data43.

Immune infiltration

TIMER is a cancer-targeted web-accessible immune infiltration database, which is a robust and efficient database that analyzes the immune cell infiltration (TIMER 2/) by utilizing data from TCGA from 32 different cancers40. The tumor-infiltrating immune cells (TIICs) comprise CD4+ T cells, dendritic cells (DCs), macrophages, CD8+ T cells, B cells, and neutrophil cells. TIMER was used for the correlation of gene expression between immune cells and hub genes.

Ethics declarations

All methods were carried out in accordance with relevant guidelines and regulations.