Identification of glutamyl-prolyl-tRNA synthetase as a new therapeutic target in hepatocellular carcinoma via a novel bioinformatic approach

Abstract

Background: Hepatocellular carcinoma (HCC) has a high incidence, and current treatments are ineffective. We aimed to explore potential diagnostic and prognostic biomarkers for HCC by conducting bioinformatics analysis on genomic and proteomic data.

Methods: Genome and proteome data were downloaded from The Cancer Genome Atlas (TCGA) and ProteomeXchange databases, respectively. Differentially expressed genes was determined using limma package. Functional enrichment analysis was conducted by Database for Annotation, Visualization, and Integrated Discovery (DAVID). Protein-protein analysis was established by STRING dataset. Using Cytoscope for network visualization and CytoHubba for hub gene identification. The gene mRNA and protein levels were validated using GEPIA and HPA, as well as RT-qPCR and Western blot.

Results: A total of 127 up-regulated and 80 down-regulated common DEGPs were identified between the genomic and proteomic data, Mining 10 key genes/proteins(ACLY, ACACB, EPRS, CAD, HSPA4, ACACA, MTHFD1, DMGDH, ALDH2, and GLDC) through protein interaction networks. in addition, Glutamyl-prolyl-tRNA synthetase (EPRS) was highlighted as an HCC biomarker that is negatively correlated with survival. Differential EPRS expression analysis in HCC and paracancerous tissues showed that EPRS expression was elevated in HCC. RT-qPCR and Western blot analysis results showed that EPRS expression was upregulated in HCC cells.

Conclusions: Our results suggest that EPRS is a potential therapeutic target for inhibiting HCC tumorigenesis and progression.

Keywords: ATP citrate lyase (ACLY); ProteomeXchange; The Cancer Genome Atlas (TCGA); glutamyl-prolyl-tRNA synthetase (EPRS); heat shock protein family A member 4 (HSPA4).

Figure 1

Flow chart. TCGA-HCC, The Cancer Genome Atlas-hepatocellular carcinoma; PPI, protein-protein interaction; GO, gene ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; GEPIA, Gene Expression Profiling Interactive Analysis; RT-qPCR, real-time reverse transcriptase-polymerase chain reaction.

Figure 2

Data processing and differential expression analysis. Volcano plots of proteins (A) and genes (B) in HCC tissues and adjacent normal tissues. The black dots in the figures represent genes or protein that do not show significant differences. Common upregulated (C) and downregulated (D) proteins or genes. The red circles represent genes, and the blue circles represent proteins. HCC, hepatocellular carcinoma.

Figure 3

GO analysis, KEGG pathway-enrichment analyses, and PPI network construction. (A) Cellular components of upregulated and downregulated genes/proteins; (B) molecular functions of upregulated and downregulated genes/proteins; (C) biological pathway functions of upregulated and downregulated genes/proteins; (D) KEGG pathway analysis of the upregulated and downregulated genes/proteins. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein-protein network.

Figure 4

PPI network of the DEPGs. Each node represents a protein, an edge between two nodes represents some kind of relationship between them. The more connections of a node show, the more significant it is in the PPI network. PPI, protein-protein network; DEPGs, differentially expressed proteins and genes.

Figure 5

Expression differences of candidate genes and their relationship with prognosis. The ten genes with the highest connectivity in the PPI network were considered as candidate genes. (A) mRNA-expression levels of the indicated 10 genes in HCC and adjacent normal tissues, based on GEPIA; (B) Kaplan–Meier Plotter analysis of the same 10 genes. *P<0.05 was considered to reflect a statistically significant difference. PPI, protein-protein network; mRNA, messenger RNA; GEPIA, Gene Expression Profiling and Interactive Analysis; ACLY, ATP citrate lyase; HSPA4, heat shock protein family A member 4.

Figure 6

EPRS, ACLY, and HSPA4 proteins were higher in HCC tissues (patient ID: 2766) than in normal liver tissues using HE staining, based on information deposited in the HPA database. (A) Liver tissue with normal EPRS expression was obtained from a 55-year-old male (patient ID: 2429, https://images.proteinatlas.org/26490/54816_A_8_4.jpg and https://images.proteinatlas.org/30052/62918_B_7_7.jpg); (B) liver tissue with normal ACLY levels was obtained from a 30-year-old female (patient ID: 3222, https://images.proteinatlas.org/22959/49962_A_9_4.jpg and https://images.proteinatlas.org/28758/61505_B_7_8.jpg); (C) liver tissue with normal HSPA4 expression was collected from a 54-year-old female (patient ID: 2429, https://images.proteinatlas.org/10023/24928_A_8_4.jpg and https://images.proteinatlas.org/10023/24927_B_9_1.jpg). Magnification, ×200. HCC, hepatocellular carcinoma; EPRS, glutamyl-prolyl-tRNA synthetase; ACLY, ATP citrate lyase; HSPA4, heat shock protein family A member 4; HE, hematoxylin-eosin.

Figure 7

EPRS expression between fresh HCC tissues and adjacent non-tumor tissues. (A) EPRS mRNA was significantly upregulated (P=0.0401) in HCC tissues (cancer) compared with adjacent non-tumor tissues (normal). (B) EPRS protein-expression levels between HCC and adjacent normal tissues. The results indicated that EPRS was overexpressed in HCC tissues compared with that in adjacent normal tissues (P=0.0484). C = tumor tissue, N = adjacent normal tissue. EPRS, glutamyl-prolyl-tRNA synthetase; mRNA, messenger RNA; HCC, hepatocellular carcinoma.