RNA sequencing reveals the expression profiles of circRNA and indicates that circDDX17 acts as a tumor suppressor in colorectal cancer

Abstract

Background: Circular RNA (circRNA) is a novel class of noncoding RNAs with functions in various pathophysiological activities. However, the expression profiles and functions of circRNAs in colorectal cancer (CRC) remain largely unknown.

Methods: High-throughput RNA sequencing (RNA-seq) was performed to assess circRNA expression profiles in 4 paired CRC tissues, and significantly dysregulated circRNAs were validated by quantitative real-time polymerase chain reaction (qRT-PCR). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to predict the potential functions of dysregulated circRNAs. Target miRNAs of circRNAs were predicted using miRanda software, and were further analyzed combining DIANA-miRPath v.3 platform (Reverse Search module) with KEGG pathways of COLORECTAL CANCER and MicroRNAs in cancer (Entry: map05210 and map05206). CircRNA-miRNA interaction networks were constructed using Cytoscape software. Expression levels of a significantly down-regulated circRNA, circDDX17 (hsa_circ_0002211), was detected by qRT-PCR in 60 paired CRC tissues. CircDDX17 was knockdown by siRNA, and the biological functions of circDDX17 were examined in CRC cell lines.

Results: Totally 448 differentially expressed circRNAs were identified, including 394 up-regulated and 54 down-regulated circRNAs. qRT-PCR validation confirmed the reliability of the RNA-Seq data. GO and KEGG analyses revealed that these dysregulated circRNAs were potentially implicated in CRC pathogenesis. Analyses by combining miRanda and miRPath softwares with KEGG pathways suggested that the miRNAs targeted by the top 10 dysregulated circRNAs were associated with the KEGG pathways of COLORECTAL CANCER and MicroRNAs in cancer, indicating that circRNA-miRNA interactions might play important functional roles in the initiation and progression of CRC. The results of qRT-PCR for circDDX17 in 60 paired CRC tissues showed that circDDX17 was significantly down-regulated in CRC tissues and associated with unfavorable clinicopathological parameters. In vitro experiments showed that silencing of circDDX17 promoted CRC cell proliferation, migration, invasion, and inhibited apoptosis.

Conclusions: In conclusion, we have identified numerous circRNAs that are dysregulated in CRC tissues compared with adjacent normal mucosa tissues. Bioinformatic analyses suggested that these dysregulated circRNAs might play important functional roles in CRC tumorigenesis. CircDDX17 functions as a tumor suppressor and could serve as a potential biomarker and a therapeutic target for CRC.

Keywords: Bioinformatic analysis; CircDDX17; Circular RNAs; Colorectal cancer; High-throughput sequencing; Tumor suppressor.

Fig. 1

a Circos plot showed the locations of cricRNA on human chromosomes. The outermost layer was a chromosome map of the human genome. The inner 8 circles represented all circRNAs of each sample detected by RNA-seq. The inner circles from outside to inside corresponded to sample N3, N4, N5, N6, T3, T4, T5, and T6, respectively. The bar chart presented the expression levels of circRNA. b The scatter plot presented the circRNA expression variations between CRC and control groups. The values of the X and Y axes represented the normalized circRNA signal values (log2 scaled). The circRNAs above the top black line and below the bottom black line displayed greater than two-fold change of up- and down-regulation. c Heatmap of inter-sample correlation showed there was an obvious difference of significantly dysregulated circRNA expression levels between CRC and control groups, but the difference was slight within each group. The Pearson’s correlation coefficient was represented by a color scale. The intensity increased from blue (relatively lower correlation) to red (relatively higher correlation). Correlation was evaluated by Pearson’s correlation coefficient of significantly dysregulated circRNA expression levels. d Volcano plot displayed the dysregulated circRNAs between CRC and control groups. The vertical gray lines corresponded to two-fold up- and down-regulation (log2 scaled), and the horizontal gray line represented a p value of 0.05. The red points represented significantly up-regulated circRNAs in CRC, and the green points represented significantly down-regulated circRNAs in CRC. e Classification of the significantly dysregulated circRNAs based on genomic origin. f Hierarchical clustering of the dysregulated circRNAs in CRC. The expression values were represented by a color scale. The intensity increased from green (relatively lower expression) to red (relatively higher expression). Each column represented one tissue sample, and each row represented a single circRNA. T, CRC tumor tissues; N, adjacent normal mucosa tissues

Fig. 2

qRT-PCR validation of ten differentially expressed circRNAs in 20 pairs of CRC samples. a The relative expression levels of 5 up-regulated and 5 down-regulated circRNAs (selected from the top 10 dysregulated circRNAs) in 20 pairs of CRC and adjacent normal mucosa tissues. b Sanger sequencing confirmed the back-splice junction sites of circRNAs. c. Comparison of log₂FC in ten differentially expressed circRNAs between RNA-Seq and qRT-PCR. Data are shown as means ± s.d. of at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001. NS, not significant. T, CRC tumor tissues; N, adjacent normal mucosa tissues

Fig. 3

GO and KEGG analyses of the host genes of differentially expressed circRNAs. a GO annotations of the host genes of differentially expressed circRNAs. The bar plot presented the enrichment scores (−log_e[p value]) of the top 10 significantly enriched GO terms in biological processes, cellular components and molecular functions. b Bulb map of KEGG analysis for the host genes of differentially expressed circRNAs. Rich factor represented the enrichment degree of differentially expressed genes. Y axis showed the name of enriched pathways. The area of each node represented the number of the enriched host genes of differentially expressed circRNAs. The p-value was represented by a color scale. The statistical significance increased from purple (relatively lower significance) to orange (relatively higher significance)

Fig. 4

CircRNA-miRNA interactions are potentially involved in the initiation and progression of CRC. a Network of circRNA-miRNA interactions potentially associated with the KEGG pathway of COLORECTAL CANCER. The network was based on the top 10 dysregulated circRNAs in CRC and their predicted target miRNAs with experimentally supported or in silico predicted target genes which were involved in the KEGG pathway of COLORECTAL CANCER. miRanda software was used to predict the target miRNAs of circRNAs, and miRPath software was used to identify miRNAs associated with the COLORECTAL CANCER pathway. The purple square node represented up-regulated circRNAs. The cyan square node represented down-regulated circRNAs. The red star node represented miRNAs associated with COLORECTAL CANCER pathway. The log₂FC was represented by a color scale, increased from cyan (relatively lower log₂FC) to purple (relatively higher log₂FC). b The top 10 dysregulated circRNAs were potentially associated with the KEGG pathway of MicroRNAs in cancer (section of “Colorectal cancer”). Target miRNAs of top 10 dysregulated circRNAs were screened according to the KEGG pathway of MicroRNAs in cancer, and target miRNAs associated with the KEGG pathway of MicroRNAs in cancer were identified. Among the top 10 dysregulated circRNAs, circPRKAG2, circITGAL, circDDX17, circASPHD1, circSPIDR, and circHOMER1 could potentially bind to miRNAs involved in the KEGG pathway of MicroRNAs in cancer. Modified from the KEGG pathway map of MicroRNAs in cancer

Fig. 5

Relative expression of circDDX17 in 60 pairs of CRC tissues. a Expression levels of circDDX17 were decreased in CRC tissues compared with adjacent normal mucosa tissues. b 71.67% (43/60) CRC patients presented decreased expression of circDDX17. The bar chart presented the log₂FC. c Expression levels of circDDX17 decreased with advanced TNM stages. d According to the median expression level of circDDX17, the 60 CRC patients were divided into two groups. The blue bars represented the low expression group, and the red bars represented the high expression group. Data are shown as means ± s.d. of at least three independent experiments. *p < 0.05, ***p < 0.001. T, CRC tumor tissues; N, adjacent normal mucosa tissues

Fig. 6

Silencing of circDDX17 promotes CRC cell proliferation, migration, invasion, and inhibits apoptosis. a Relative expression of circDDX17 in 6 CRC cell lines. b Schematic representation of the sequence around the back-splice junction site of circDDX17 and the siRNAs targeting the junction site (si-circDDX17#1 and si-circDDX17#2) (top). Results of qRT-PCR for circDDX17 and its linear isoform in SW480 and SW620 cells treated with siRNAs. c CCK-8 assay of SW480 and SW620 cells transfected with negative control siRNA (si-NC) or si-circDDX17#1 at the indicated days. d Colony formation assays of SW480 and SW620 cells transfected with si-NC or si-circDDX17#1. e Cell apoptosis assay by flow cytometry of SW480 and SW620 cells transfected with si-NC or si-circDDX17#1. f, g Transwell migration and invasion assays of SW480 and SW620 cells transfected with si-NC or si-circDDX17#1. Data are shown as means ± s.d. of at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001. Scale bar, 20 μm