Identifying MMP14 and COL12A1 as a potential combination of prognostic biomarkers in pancreatic ductal adenocarcinoma using integrated bioinformatics analysis

Jingyi Ding; Yanxi Liu; Yu Lai

doi:10.7717/peerj.10419

Identifying MMP14 and COL12A1 as a potential combination of prognostic biomarkers in pancreatic ductal adenocarcinoma using integrated bioinformatics analysis

PeerJ. 2020 Nov 23:8:e10419. doi: 10.7717/peerj.10419. eCollection 2020.

Authors

Jingyi Ding¹, Yanxi Liu², Yu Lai³

Affiliations

¹ Hospital of Chengdu University of Traditional Chinese Medicine, Chengdu, China.
² University of California, Los Angeles, Los Angeles, CA, United States of America.
³ School of Basic Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China.

Abstract

Background: Pancreatic ductal adenocarcinoma (PDAC) is a fatal malignant neoplasm. It is necessary to improve the understanding of the underlying molecular mechanisms and identify the key genes and signaling pathways involved in PDAC.

Methods: The microarray datasets GSE28735, GSE62165, and GSE91035 were downloaded from the Gene Expression Omnibus. Differentially expressed genes (DEGs) were identified by integrated bioinformatics analysis, including protein-protein interaction (PPI) network, Gene Ontology (GO) enrichment, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses. The PPI network was established using the Search Tool for the Retrieval of Interacting Genes (STRING) and Cytoscape software. GO functional annotation and KEGG pathway analyses were performed using the Database for Annotation, Visualization, and Integrated Discovery. Hub genes were validated via the Gene Expression Profiling Interactive Analysis tool (GEPIA) and the Human Protein Atlas (HPA) website.

Results: A total of 263 DEGs (167 upregulated and 96 downregulated) were common to the three datasets. We used STRING and Cytoscape software to establish the PPI network and then identified key modules. From the PPI network, 225 nodes and 803 edges were selected. The most significant module, which comprised 11 DEGs, was identified using the Molecular Complex Detection plugin. The top 20 hub genes, which were filtered by the CytoHubba plugin, comprised FN1, COL1A1, COL3A1, BGN, POSTN, FBN1, COL5A2, COL12A1, THBS2, COL6A3, VCAN, CDH11, MMP14, LTBP1, IGFBP5, ALB, CXCL12, FAP, MATN3, and COL8A1. These genes were validated using The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases, and the encoded proteins were subsequently validated using the HPA website. The GO analysis results showed that the most significantly enriched biological process, cellular component, and molecular function terms among the 20 hub genes were cell adhesion, proteinaceous extracellular matrix, and calcium ion binding, respectively. The KEGG pathway analysis showed that the 20 hub genes were mainly enriched in ECM-receptor interaction, focal adhesion, PI3K-Akt signaling pathway, and protein digestion and absorption. These findings indicated that FBN1 and COL8A1 appear to be involved in the progression of PDAC. Moreover, patient survival analysis performed via the GEPIA using TCGA and GTEx databases demonstrated that the expression levels of COL12A1 and MMP14 were correlated with a poor prognosis in PDAC patients (p < 0.05).

Conclusions: The results demonstrated that upregulation of MMP14 and COL12A1 is associated with poor overall survival, and these might be a combination of prognostic biomarkers in PDAC.

Keywords: Bioinformatics; Biomarker; COL12A1; MMP14; Pancreatic ductal adenocarcinoma; Prognostic.

Grants and funding

The research was supported by the Special Program of Science and Technology Research of Chinese Medicine Administration Bureau of Sichuan Province (2018JC012). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.