Integrated bioinformatical analysis of lncRNA-mRNA co-expression profiles of cervical cancer

Abstract

Background: Aberrant gene expression, including protein-coding and non-coding genes (like long non-coding RNA, lncRNA), is associated with cervical cancer development. To reveal the possible molecular mechanisms of cervical carcinogenesis, this study conducted high throughput sequencing along with a bioinformatics analysis. Methods: The differentially expressed lncRNAs and mRNAs were assessed using a microarray technique in three pairs of cervical cancer and paracancerous tissues and analyzed using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotations. Their co-expression profiles, containing 653 nodes and 400 edges, were constructed and analyzed using Pearson’s correlation and lncRNA-mRNA co-expression network analyses. Results: There were 242 lncRNAs and 169 mRNAs upregulated and 1204 lncRNAs and 1131 mRNAs downregulated in cervical cancer (fold change ≥2 or ≤–2; p ≤ 0.05). The KEGG pathway analytic data showed that these differentially expressed mRNAs were primarily enriched in the ubiquitin-mediated proteolysis and cancer pathways. The differentially expressed cis-targeted lncRNAs were associated with the TGF-β signaling and cell adhesion molecules, while the trans-targeted lncRNAs were related to p53 signaling. Functional analysis of the integrated mRNA-lncRNA co-expression networks were linked to autophagy, regulation of vascular genesis, transcriptional repressor complex, and regulation of gene expression. Conclusions: These differentially expressed lncRNAs and mRNAs could be associated with cervical cancer development and/or progression.

Keywords

• cervical cancer
• long non-coding rna
• microarray
• pathway analysis
• co-expression network