ESPL1 is Elevated in Hepatocellular Carcinoma and Predicts Prognosis

Abstract

Purpose: The extra spindle pole bodies-like 1 (ESPL1) gene is associated with malignant biological behaviors in several tumors. Nevertheless, the correlation between hepatocellular carcinoma (HCC) and ESPL1 has not been determined. The present study analyzed the molecular function and prognostic value of ESPL1 in HCC.

Patients and methods: Samples from 121 HCCs and 119 adjacent normal tissue specimens were subjected to next-generation sequencing. Clinicopathological and genetic data of HCC patients in The Cancer Genome Atlas (TCGA) were also collected. ESPL1 expression was assessed in 20 pairs of HCC and normal liver specimens by qRT-PCR and immunohistochemistry (IHC). The prognostic value of ESPL1 expression was determined by Cox univariate and multivariate regression analyses. ESPL1-related co-expressed genes were evaluated by weighted gene co-expression network analysis (WGCNA). Processes and pathways involving ESPL1 in HCC were determined by Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. The prognostic values of hub genes were determined by joint effect survival analysis.

Results: RNA-Seq, RT-qPCR and IHC showed that ESPL1 expression was significantly higher in HCC than in normal liver tissues. Increased ESPL1 expression, greater tumor size and advanced BCLC stage were independently prognostic of poorer overall survival; and increased ESPL1 and advanced BCLC stage were independently prognostic of poorer recurrence-free survival. WGCNA showed that the top 10 co-expressed genes associated with ESPL1 were GTSE1, KIF18B, BUB1B, GINS1, PRC1, KIF23, KIF18A, TOP2A, NEK2 and FANCD2. Enrichment analysis indicated that ESPL1 and its co-expressed genes might be involved in the cell cycle and cell division of HCC. Joint effect survival analysis showed that the mortality rate was approximately 3.37 times higher in HCC patients with high than low expression of ESPL1, GTSE1, BUB1B, PRC1, KIF23, and TOP2A.

Conclusion: ESPL1 might be associated with cell cycle and might be an effective prognostic indicator in patients with HCC.

Keywords: extra spindle pole bodies-like 1; hepatocellular carcinoma; prognosis.

Figure 1

ESPL1 mRNA expression in HCC in the Guangxi RNA-seq database. (A and B) ESPL1 expression levels in tumors and adjacent normal tissues. (C and D) Association between ESPL1 expression and clinical characteristics. (E) qRT-PCR validation of ESPL1 expression in tumor tissues and adjacent normal tissues. ****P < 0.0001; **P < 0.01; *P < 0.05.

Figure 2

ESPL1 protein is greatly expressed in HCC samples. (A and E) Immunohistochemical staining of HCC tissue samples and adjacent normal tissues. (Magnification, 100× and 400×; *P < 0.05.) (B) Immunofluorescent staining of human epidermoid carcinoma (A-431), osteosarcoma (U2-OS) and glioblastoma (U-251 MG) cell lines in the HPA database. The red color represents microtubule and the green color represents ESPL1. (C and D) ESPL1 expression level and cell cycle phase in single U-2 OS cells.

Figure 3

Effects of ESPL1 mRNA expression level on survival in patients with HCC. Relationship of ESPL1 mRNA with OS (A and C) and RFS (B and D) in the (A and B) Guangxi cohort and (C and D) the GEPIA database.

Figure 4

WGCNA analysis of: (A) network topology for various soft-thresholding powers (weighted coefficient, β). The x-axis represents different soft-thresholding powers. The y-axis represents the correlation coefficient between log (k) and log [P(k)]. The red line indicates a correlation coefficient of 0.9. (B) Clustering gene dendrogram showing five co-expressed modules (different colors) in patients with HCC. (C) Scatter plot of gene and module correlation within the blue module. (D) Heatmap showing the relationships between modules and clinical parameters; the correlation coefficient and corresponding P-value are shown in each cell. (E) Top 10 genes in the blue module.

Figure 5

Enrichment analysis of GO and KEGG ontologies and pathways among DEGs in the blue module (A) GO-BP; (B) GO-CC; (C) GO-MF; (D) KEGG.

Figure 6

Kaplan–Meier analysis of OS in patients relative to the levels of expression of the genes (A) ESPL1 (B) BUB1B (C) GTSE1 (D) KIF23 (E) PRC1 (F) TOP2A. (G) Joint model based on six genes.

Figure 7

Prognostic risk score model and time-dependent ROC curve of the ESPL1 gene in HCC (A–C) Guangxi cohort; (D–F) TCGA cohort.

Figure 8

Single cell profiling of six HCCs from the TISCH database. (A–C) UMAP and violin plots showing the expression levels of ESPL1. (D) Proportions of immune cells in tumor tissue of six HCC patients. (E–G) Pathway enrichment of proliferating T cells based on the HALLMARK dataset.