A network-based analysis to identify a piRNA-target signature related to colorectal cancer prognosis: in silico and in vitro study

Abstract

Purpose: Patients with colorectal cancer (CRC) are diagnosed in advanced stages and have worse overall survival. Also, this cancer incidence is rising in many countries. The aim of this study is to find piwi-interacting RNAs (piRNA) predicting the prognosis of patients with colorectal cancer, using bioinformatics and evaluating these results through RT-qPCR method.

Methods: The target genes of piRNAs were predicted using miRDB and TargetRank databases. Protein-protein interaction (PPI) networks were constructed by STRING and were analyzed with Cytoscape software and the MCODE tool used for module construction. Expression levels of final selected piRNAs in 18 pairs of CRC tissue and adjacent normal tissue were measured by quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR).

Results: Twenty CRC-related piRNAs and 980 target genes were included in this study. After PPI analysis 19 hub genes were identified. Then, the prognostic value of these hub genes was assessed via Kaplan-Meier survival analyses. This survival analysis indicated that the expression of six genes was significantly associated with overall survival of patients with CRC. These genes are targets of hsa-piR-487, hsa-piR-28944 and piR-hsa-8401. Also, the pathway analysis revealed the potential signal pathways of these piRNAs targets involved in CRC. RT-qPCR showed that hsa-piR-487 and hsa-piR-28944 expression significantly were down-regulated in CRC tumor tissues compared with the adjacent normal tissues (P < 0.05, P < 0.01).

Conclusion: It seems that hsa-piR-487 and hsa-piR-28944 can be considered as a potential biomarker for the diagnosis of CRC. However, it is still necessary to conduct studies with a higher statistical population and measure them in the serum of patients to confirm these results.

Keywords: Biomarkers; Colorectal neoplasms; Piwi-interacting RNA; Prognosis.

Fig. 1

Flowchart of this study

Fig. 2

Venn diagram of hub piRNAs and regulated genes in Cytoscape based on topological features (degree, betweenness centrality, and closeness centrality). a piRNA-target genes network, b Target genes network, and c Merged network

Fig. 3

Top ten GO annotation results of piRNAs targets. a Biological process (BP) enrichment analysis; b cell component (CC) enrichment analysis; c molecular function (MF) enrichment analysis. GO gene ontology

Fig. 4

Pathway enrichment results. Top 10 pathways enriched by all the targets of piRNAs

Fig. 5

PPI network of piRNAs-target genes, merged network. a The PPI network of piRNAs-target genes was constructed using Cytoscape. b The PPI network of the merged network. The green nodes represent the target genes, while the blue nodes represent piRNAs.The most significant module was selected from the PPI network. c The module with a score of 4.889, 19 nodes, and 88 edges. d The module with a score of 7.688, with 33 nodes and 246 edges. The red nodes represent the genes with high topological features (19 hub genes)

Fig. 6

Target genes are associated with overall survival. The red lines show individuals with high expression of hub genes. The black lines show individuals with low expression of hub genes

Fig. 7

Pathway enrichment results. Top 20 pathways enriched by all the target genes of a hsa-piR-487; b hsa-piR-28944 c piR-hsa-8401

Fig. 8

Comparison of a hsa-piR-28944 and b hsa-piR-487 expression levels between CRC tissues and adjacent normal tissues. Analysis using the student’s t-test showed that the relative expression levels of hsa-piR-28944 and hsa-piR-487 in the CRC tissues were significantly lower than those in adjacent normal tissues (P < 0.05, P < 0.01)