Integrated Network Analysis Decipher ZNF384-Related miR-20b-5p and miR-424-5p in Colon Adenocarcinoma

Background

ZNF384 is a C2H2-type zinc finger protein (ZNF) which is implicated in DNA double-strand break (DSB) repair through the classical non-homologous end-joining (cNHEJ) pathway.

Aims

To clarify the regulatory mechanisms involving ZNF384 in colon adenocarcinoma (COAD).

Methods and Results

First, we conducted a differential expression gene (DEG) analysis of mRNA and lncRNA using TCGA-COAD RNA-Seq data. We also identified ZNF384-related mRNAs through Pearson's correlation coefficient calculation and conducted weighted gene co-expression network analysis (WGCNA) for these genes, leading to the identification of a cluster of 331 genes with strongly positive correlation to tumor, 84 of which overlapped with DEGs. Gene functional analysis showed enrichment of genes in DNA repair, replication fork, and cell cycle checkpoint signaling pathways. Protein–protein interaction (PPI) network analysis of these 84 genes led to the identification of the top 20 key mRNAs. Then we employed three machine learning methods to refine our selection of candidate genes from these intersecting mRNAs. We constructed a competitive endogenous RNA (ceRNA) network and identified two significant intersecting miRNAs, miR-20b-5p and miR-424-5p, which have been shown to act as a tumor suppressor gene and an oncogene, respectively. Additionally, we found that KIF14 and KIF18B are regulated by these two miRNAs in this ceRNA network, particularly in DNA damage repair and cell cycle. Finally, validation using an external dataset from the GEO database confirmed their expression patterns.

Conclusion

The current study clarifies the mechanisms of how miR-20b-5p and miR-424-5p work in colon cancer and underscores their predictive capabilities in colon cancer.