br Fig Expression of circHIPK
Fig. 4. Expression of circHIPK3 in ovarian cancer NCT-501 and knockdown eﬃciency of circHIPK3 siRNA. A: qRT-PCR detection of the expression levels of circHIPK3 among 5 ovarian cancer cell lines (HO8910, A2780, OVCAR8, COC1 and SKOV3) and 1 normal ovarian epithelium cell line (IOSE80). B: RNA isolated from the cytoplasm and nucleus of A2780 ovarian cancer cells were analyzed by qRT-PCR. CircHIPK3 were mainly distributed in the cytoplasm, similar to the 12S control. C and D: Si-circHIPK3 only knocked down the circHIPK3 (C) but not their linear HIPK3 mRNA (D) in A2780 and SKOV3 cells. Data are shown as the mean ± SEM.
HIPK3 mRNA counterparts were not significantly changed in both A2780 and SKOV3 cells (Fig. 4D).
3.4. Silencing of circHIPK3 significantly promoted the invasion and migration of ovarian cancer cells and normal ovarian epithelial cells
After circHIPK3 was successfully knocked down in ovarian cancer cells, we further examined the invasion and metastasis of ovarian cancer cells. As shown in Fig. 5A and B, silencing of circHIPK3 could significantly promote the migration of ovarian cancer cells compared with that in the negative control group. Consistent with the wound-healing results, downregulation of circHIPK3 significantly promoted migration and invasion of A2780 and SKOV3 cells, as shown in the transwell assay (*p < 0.05, Fig. 5C and E). The results indicate that knockdown of circHIPK3 markedly enhanced migration and invasion of A2780 and SKOV3 cells (*p < 0.05, Fig. 5D and F). Besides, we also examined the eﬀects of circHIPK3 knock-down on the migration and invasion ability of IOSE80 cells. The results indicated that knock-down
3.5. Downregulation of circHIPK3 increased cell proliferation and decreased cell apoptosis of ovarian cancer cells and normal ovarian epithelial cells
As an important indicator of the malignant behavior of cancer cells, cell proliferation and apoptosis was further examined. As shown in Fig. 6A and B and Figure S2B, cell proliferation was markedly increased in the circHIPK3 knockdown ovarian cancer cells and normal ovarian epithelial cells at 48 h and 72 h after seeding (*p < 0.05, **p < 0.01, ***p < 0.001). Flow cytometry analysis indicated that cell apoptosis was
significantly decreased in the circHIPK3 knockdown ovarian cancer
3.6. Possible mechanism of circHIPK3 in ovarian cancer progression
According to previously reported results (Zheng et al., 2016; Shan et al., 2017; Li et al., 2017; Zeng et al., 2018), circHIPK3 (hsa_-circ_0000284) primarily functions through the circHIPK3-miRNA-mRNA axis. Therefore, ceRNA analysis of circHIPK3 was also per-formed. Our results also supported the observation that circHIPK3 might play an important role in EOC through the circHIPK3-miRNA-mRNA axis (Fig. 6E). We collected 12 miRNAs that have 1–3 binding sites in circHIPK3, as reported previously (Zheng et al., 2016), or that were expressed with high abundancy in our sequencing results, and we also predicted their possible target genes via miRanda (Table S5-6). Cytoscape was performed to visualize the network (Fig. 6E).
With the development of NGS, thousands of circRNAs have been found. Many circRNAs have been verified to play vital roles in carci-nogenesis, including cell proliferation, apoptosis, migration and inva-sion (Wang et al., 2017; Zhong et al., 2017; Chen et al., 2017). One study analyzed circRNAs expression in primary ovarian cancer tissues and their lymph nodes and in peritoneum metastatic cancer tissues (Ahmed et al., 2016). In the current study, we further explored the diﬀerently expressed circRNAs in EOC and NOT using high throughput sequencing. Compared with NOT, 2431 and 3120 circRNAs were dis-tinctly upregulated and downregulated in EOC, respectively.
To further explore circRNAs that might exert an important function
Fig. 5. Silencing of circHIPK3 promoted migration and invasion of ovarian cancer cell lines. A and B: The migration potential of cells transfected with the
circHIPK3 siRNA and control was determined by the wound healing assay at 24 h after injury in A2780 and SKOV3 cells. C: Cell migration and invasion capabilities of
A2780 cells were evaluated by transwell assay without (migration) and with (invasion) Matrigel assays. D: The protein concentration of the migrated and invaded
A2780 cells was analyzed using a microplate reader at a wavelength of 562 nm. E and F: The cell migration and invasion potential of cells transfected with circHIPK3 siRNA and control were repeated in SKOV3 cells in the transwell assay. The results are shown as the mean ± SEM. *p < 0.05, **p < 0.01.
in EOC, we validated six circRNAs that were significantly diﬀerentially expressed, and the qRT-PCR results are consistent with the sequencing data, which indicated that the sequencing data are credible.