Frontiers in Oncology (Dec 2020)

Ubiquitin-Like Modifier Activating Enzyme 1 as a Novel Diagnostic and Prognostic Indicator That Correlates With Ferroptosis and the Malignant Phenotypes of Liver Cancer Cells

  • Yiru Shan,
  • Guang Yang,
  • Haixia Huang,
  • Yehan Zhou,
  • Xiangyu Hu,
  • Qiuhong Lu,
  • Peng Guo,
  • Jun Hou,
  • Li Cao,
  • Fuhua Tian,
  • Qi Pan,
  • Qi Pan

DOI
https://doi.org/10.3389/fonc.2020.592413
Journal volume & issue
Vol. 10

Abstract

Read online

PurposeFerroptosis is a type of cell death that is iron dependent, a characteristic that distinguishes it from necrosis, apoptosis, and autophagy. However, the ferroptotic mechanisms for hepatitis B virus-associated hepatocellular carcinoma (HCC) remain incompletely described.MethodsTwo hepatitis B virus-associated HCC public datasets, GSE22058 (n=192) and GSE54238 (n=23), were obtained from the NCBI Gene Expression Omnibus (GEO) database. Bioinformatics methods, including weighted gene coexpression network analysis (WGCNA), Cox regression, and LASSO analysis, were used to identify signature markers for diagnosis and prognosis. CCK8, wound healing, Transwell migration/invasion, and ferroptosis assays were employed to explore the biological function of novel candidate markers weight gene coexpression network analysis.ResultsIn total, 926 differentially expressed genes (DEGs) were common between the GSE22058 and GSE54238 datasets. Following WGCNA, 515 DEGs derived from the MEturquoise gene module were employed to establish diagnosis and prognosis models in The Cancer Genome Atlas (TCGA) HCC RNA-Seq cohort (n=423). The score of the diagnostic model was strikingly upregulated in the TCGA HCC group (p<2.2e-16). The prognostic model exhibited high specificity and sensitivity in both training and validation (AUC=0.835 and 0.626, respectively), and the high-risk group showed dismal prognostic outcomes compared with the low-risk group (training: p=1.416e-10; validation: p=4.495e-02). Ubiquitin-like modifier activating enzyme 1 (UBA1) was identified among both diagnosis and prognosis signature genes, and its overexpression was associated with poor survival. We validated the expression level of UBA1 in eight pairs of HCC patient tissues and liver cancer cell lines. UBA1 silencing decreased proliferation, migration, and invasion in Huh7 cells while elevating the Fe2+ and malondialdehyde (MDA) levels. Additionally, these biological effects were recovered by oltipraz (an Nrf2 activator). Furthermore, blocking UBA1 strikingly repressed the protein expression levels of Nrf2, HO-1, NQO1, and FTH1 in the Nrf2 signal transduction pathway.ConclusionOur findings demonstrated that UBA1 participates in the development of HCC by modulating Huh7 phenotypes and ferroptosis via the Nrf2 signal transduction pathway and might be a promising diagnostic and prognostic indicator for HCC.

Keywords