ACOX2 is a prognostic marker and impedes the progression of hepatocellular carcinoma via PPARα pathway

Abstract

Hepatocellular carcinoma (HCC) has been extensively studied as one of the most aggressive tumors worldwide. However, its mortality rate remains high due to ideal diagnosis and treatment strategies. Uncovering novel genes with prognostic significance would shed light on improving the HCC patient's outcome. In our study, we applied data-independent acquisition (DIA) quantitative proteomics to investigate the expression landscape of 24 paired HCC patients. A total of 1029 differentially expressed proteins (DEPs) were screened. Then, we compared DEPs in our cohort with the differentially expressed genes (DEGs) in The Cancer Genome Atlas, and investigated their prognostic significance, and found 183 prognosis-related genes (PRGs). By conducting protein-protein interaction topological analysis, we identified four subnetworks with prognostic significance. Acyl-CoA oxidase 2 (ACOX2) is a novel gene in subnetwork1, encodes a peroxisomal enzyme, and its function in HCC was investigated in vivo and in vitro. The lower expression of ACOX2 was validated by real-time quantitative PCR, immunohistochemistry, and Western blot. Cell Counting Kit-8 assay, wound healing, and transwell migration assay were applied to evaluate the impact of ACOX2 overexpression on the proliferation and migration abilities in two liver cancer cell lines. ACOX2 overexpression, using a subcutaneous xenograft tumor model, indicated a tumor suppressor role in HCC. To uncover the underlying mechanism, gene set enrichment analysis was conducted, and peroxisome proliferator-activated receptor-α (PPARα) was proposed to be a potential target. In conclusion, we demonstrated a PRG ACOX2, and its overexpression reduced the proliferation and metastasis of liver cancer in vitro and in vivo through PPARα pathway.

Fig. 1. The DEP identification and correlation analysis with HCC clinicopathological traits.

A OPLS-DA analysis of protein expression in hepatocellular carcinoma (CA) and paired adjacent normal (NO) samples. B Volcano plot showing significantly differentially expressed proteins. UP upregulated proteins were labeled in purple, DW downregulated proteins were labeled in green, NoDiff proteins with no significant differential expression. C Hierarchical clustering of differentially expressed proteins in hepatocellular carcinoma (CA) and paired adjacent normal (NO) samples. Clinical traits, including PT, DBIL, TBIL, AST, ALT, tumor number, embolus, differentiation, capsule, AFP, cirrhosis, gender, and age, were shown.

Fig. 2. Functional enrichment analysis of common DEGs.

A The common genes between our DEGs and TCGA cohort DEGs, PRGs of HCC. B The heatmap of the 161 common DEGs in our cohort and TCGA cohort of HCC. The red color indicates the normal samples and blue color indicates the HCC samples. C The top 15 significantly enriched GO-biological processes of the 161 DEGs D The top 15 significantly enriched KEGG pathways of the 161 common DEGs.

Fig. 3. Protein–protein interaction network of 161 common DEGs.

A Protein–protein interaction network of 161 common DEGs. Upregulated DEGs were labeled in red and downregulated DEGs were labeled in blue; P value and fold change were log transformed, and applied as the bolder with and node fill color; topological degree was used as the criteria for node size. B Four key subnetworks from the main network. The subnetworks were based on the results of the score calculation using the MCODE plugin of Cytoscape.

Fig. 4. Forest plot of the four subnetworks.

A Forest plot of the genes in subnet 1. B Forest plot of the genes in subnet 2. C Forest plot of the genes in subnet 3. D Forest plot of the genes in subnet 4. The forest plot showed the P value, hazard ratio (HR), and 95% CI of the DEGs.

Fig. 5. ACOX2 is downregulated in liver cancer.

A The expression of ACOX2 gene in liver cancer tissues and paired paracancer tissues by using RT-qPCR. B The expression of ACOX2 protein in liver cancer tissues and normal control tissues was detected by using immunohistochemistry. The expression distribution of ACOX2 in normal liver tissue and HCC patient samples was evaluated in the THPA database. C The expression of ACOX2 mRNA in normal liver cells and liver cancer cells by using RT-qPCR (*P < 0.05 and **P < 0.01). D The protein expression of ACOX2 in normal liver cells and liver cancer cells by Western blot (***P < 0.001). E The mRNA expression of ACOX2 in the TCGA dataset (***P < 0.001). F Survival analysis of the association between ACOX2 and overall survival time in HCC patients (based on the TCGA data, P = 0.0326).

Fig. 6. ACOX2 overexpression inhibits liver cancer cell proliferation and migration.

A ACOX2 overexpression inhibits HepG2 and SMCC-7721 cells growth by CCK-8. B ACOX2 overexpression inhibits migration of HepG2 and SMCC-7721 cells. The wound-healing ability was determined 24 h after scratching. C ACOX2 overexpression inhibits invasion of HepG2 and SMCC-7721 cells. The effects of ACOX2 on liver cancer cell invasion were determined by transwell assays. D ACOX2 inhibits liver cancer cells in vivo. ACOX2 inhibits tumor growth of HepG2 cells in nude mice. The control, ACOX2-overexpressing, or vector-infected HepG2 cells were injected into nude mice (n = 6). Tumor weight was displayed above. *P < 0.05 and **P < 0.01.

Fig. 7. ACOX2 overexpression inhibits tumor lung metastasis.

A Representative in vivo imaging of control, ACOX2-overexpressing, or vector-infected HepG2 cells in nude mice after 36 days. B Representative HE staining images of lung metastatic nodules (indicated by red arrow).

Fig. 8. ACOX2 inhibits tumor through the PPARα signaling pathway.

A The GSEA showed that PPAR (P = 0.004) and FATTY_ACID_METABOLISM (P = 0.006) were significantly changed. B The PPI network of ACOX2, PPARA, PPARD, and PPARG was constructed by the String database. C The expression of PPARA, PPARD, and PPARG genes in liver cancer tissues and adjacent noncancerou tissues by using RT-qPCR. D The mRNA expression of PPARA, PPARD, and PPARG was detected by RT-qPCR in HepG2 and SMCC-7721 (NC, the normal HCC cell lines; ACOX2-Ov, overexpression of ACOX2; Sh- ACOX2, knockdown ACOX2). E The protein levels of PPARA, PPARD, and PPARG were detected by Western blot in HepG2 and SMCC-7721 (NC, the normal HCC cell lines; ACOX2-Ov, overexpression of ACOX2; Sh- ACOX2, knockdown ACOX2, *P < 0.05; **P < 0.01; n.s. not significant). F The Pearson correlation analysis between ACOX2 and PPARA, PPARD, and PPARG by the GEPIA database.