A 3-circular RNA signature as a noninvasive biomarker for diagnosis of colorectal cancer

Abstract

Background: Circular RNAs (circRNAs), a novel type of noncoding RNAs, play critical roles in the initiation and progression of cancer. Emerging studies also shows that circRNAs may function as potential markers for cancer diagnosis and treatment. However, the diagnostic value of circRNAs in colorectal cancer (CRC) remains need to be unearthed.

Methods: CircRNA microarray was performed to detect the differentially expressed circRNAs in eight plasma samples, including four colorectal cancer (CRC) and four normal samples. Besides, the results of microarray were validated by quantitative real-time polymerase chain reaction (qRT-PCR). Moreover, ROC curve evaluation was performed to calculate the diagnostic value of significantly dysregulated circRNAs. In order to predict the potential mechanism of the significant circRNAs, circRNA-miRNA-mRNA network was constructed based on the TargetScan, miRTarBase and MIRDB database, as well as CircInteractome online software. Furthermore, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to further predict the function of meaningful circRNAs.

Results: Totally three differentially expressed circRNAs were identified in CRC plasma compared to normal plasma by circRNA microarray analysis, and the results was validated by qRT-PCR. Hsa_circ_0082182, hsa_circ_0000370 and hsa_circ_0035445 were identified and ROC curves analysis was used to calculate the single and joint diagnostic value. Furthermore, GO and KEGG analyses revealed that functions were mainly cancer-related, which indicated that the circRNAs were meaningfully associated with CRC cell proliferation and metastasis.

Conclusion: In conclusion, we have identified three circRNAs that are dysregulated in CRC plasma, including hsa_circ_0082182, hsa_circ_0000370 and hsa_circ_0035445. ROC curves showed that these circRNAs might have diagnostic value for colorectal cancer. Furthermore, bioinformatics analysis indicated that the above-mentioned circRNAs might be involved in the development of CRC.

Keywords: Biomarker; Diagnostic signature; ROC curve; circRNAs.

Fig. 1

The flow chart of the whole work

Fig. 2

Profiling of circRNAs in the plasmas from GC patients CRC and normal plasma and the biological structure of circRNAs. a Volcano plot displays the circRNAs differentially expressed (fold change > 2) between CRC and normal plasma (P < 0.05). b Heat map shows the top 20 dysregulated circRNAs between CRC and normal plasma. c Schematics shows the biological structure of hsa_circ¬_0082182, hsa_circ¬_0000370, and hsa_circ¬_0035445

Fig. 3

Relative expression of hsa_circ¬_0082182, hsa_circ¬_0000370, and hsa_circ-_0035445. a–c The expression of hsa_circ_0082182, hsa_circ_0000370 and hsa_circ_0035445 in CRC plasma and healthy control. d–f The expression of hsa_circ_0082182, hsa_circ_0000370 and hsa_circ_0035445 in cell lines. g–i The expression of hsa_circ_0082182, hsa_circ_0000370, and hsa_circ_0035445 in the 66 stages I-CRC plasma and normal group. Student’s t-test or ANOVA was used to compare the differences in circRNA expression between different groups. *** means P < 0.001, **** means P < 0.0001

Fig. 4

ROC curves analysis of hsa_circ_0082182, hsa_circ_0000370, hsa_circ-_0035445, and mixture of these three circRNAs (a–d)

Fig. 5

The expression of circRNAs (hsa_circ_0082182, hsa_circ_0000370, and hsa_circ_0035445) in the plasma of CRC patients before and after surgery (a–c)

Fig. 6

Functional analysis for three circRNAs. a GO and KEGG analysis of hsa_circ_0082182 based on the ceRNA network. b GO and KEGG analysis of hsa_circ_0000370 based on the ceRNA network. c GO and KEGG analysis of hsa_circ_0035445 based on the ceRNA network